Re: [freenet-support] Freenet 0.7 Issue

[Matthew Toseland]

On Monday 07 Nov 2011 22:13:30 Dave Riley wrote:
 I use an Intel iMac running 10.6.8. I used freenet.jnlp, then 
 new_installer_offline_1404.jar to Install the Freenet directory onto a 3GB 
 TrueCrypt volume, and the tray into /Applications -- this configuration 
 worked fine in an iMac PPC running Leopard (in fact, I later deleted the 
 tray; I started / stopped Freenet via the Terminal.).When I tried to use the 
 Wizard (both times)I received this page after the Data Storage page:
  
  Internal error: please report
  
  java.nio.channels.OverlappingFileLockException
  at 
 sun.nio.ch.FileChannelImpl$SharedFileLockTable.checkList(FileChannelImpl.java:1166)
  at 
 sun.nio.ch.FileChannelImpl$SharedFileLockTable.add(FileChannelImpl.java:1068)
  at sun.nio.ch.FileChannelImpl.lock(FileChannelImpl.java:824)
  at java.nio.channels.FileChannel.lock(FileChannel.java:860)
  at 
 freenet.store.saltedhash.SaltedHashFreenetStore.openStoreFiles(SaltedHashFreenetStore.java:831)
  at 
 freenet.store.saltedhash.SaltedHashFreenetStore.(SaltedHashFreenetStore.java:189)
  at 
 freenet.store.saltedhash.SaltedHashFreenetStore.construct(SaltedHashFreenetStore.java:147)
  at freenet.node.Node.makeStore(Node.java:3591)
  at freenet.node.Node.makeStore(Node.java:3584)
  at freenet.node.Node.initSaltHashFS(Node.java:3395)
  at freenet.node.Node.makeStore(Node.java:900)
  at freenet.node.Node$StoreTypeCallback.set(Node.java:365)
  at freenet.node.Node$StoreTypeCallback.set(Node.java:338)
  at freenet.config.Option.set(Option.java:65)
  at freenet.config.Option.setValue(Option.java:57)
  at freenet.config.SubConfig.set(SubConfig.java:301)
  at 
 freenet.clients.http.wizardsteps.DATASTORE_SIZE._setDatastoreSize(DATASTORE_SIZE.java:122)
  at 
 freenet.clients.http.wizardsteps.DATASTORE_SIZE.postStep(DATASTORE_SIZE.java:93)
  at 
 freenet.clients.http.FirstTimeWizardToadlet.handleMethodPOST(FirstTimeWizardToadlet.java:231)
  at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
  at 
 sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
  at 
 sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
  at java.lang.reflect.Method.invoke(Method.java:597)
  at 
 freenet.clients.http.ToadletContextImpl.handle(ToadletContextImpl.java:550)
  at 
 freenet.clients.http.SimpleToadletServer$SocketHandler.run(SimpleToadletServer.java:989)
  at 
 freenet.support.PooledExecutor$MyThread.realRun(PooledExecutor.java:233)
  at freenet.support.io.NativeThread.run(NativeThread.java:130)
  
  What am I missing? Thanks for your time.
  
Did you ever fix this? It might be trying to run more than one copy - did it 
work if you open it in a web browser?

Sorry for taking a long time to get back to you. You might like to try 
installing a more recent version if you had given up on it!


signature.asc
Description: This is a digitally signed message part.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1232

[Matthew Toseland]

Sorry folks, don't use 1231 (fortunately the upload to the auto-update didn't 
complete), it had a severe bug preventing startup. 1232 fixes this and also has 
some minor work on plugins.


signature.asc
Description: This is a digitally signed message part.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1229, 1230 and 1231

[Matthew Toseland]

Please upgrade to 1231. Changelogs:
1229:
- XML vulnerability warnings fixes.
- Fix an NPE in plugins.
- Minor internal stuff.
1230:
- Detect the XML vulnerability on OS/X. Try to detect it on OpenJDK, maybe not 
very well.
- Clarify an english string (Completed downloads to temporary space not 
directory).
- Sync before closing the new peers file. Some wierd filesystems might need 
this and we're still getting reports of losing peers on power loss.
- Minor internal stuff.
1231:
- Increase the maximum peers limit and make it scale with your upstream 
bandwidth. 11 peers at 10KB/sec, 15 at 20KB/sec, 19 at 30KB/sec, 26 at 
60KB/sec, 34 at 100KB/sec and 40 and 140KB/sec. Show the limit on the stats 
page.
- Support BMPs in fproxy (from kurmi's Summer of Code project).
- Slightly better (X)HTML support in fproxy.
- When changing the store type (most notably when setting the store size in the 
first time wizard), don't stall for minutes or more preallocating the datastore 
- do it in the background. IMHO this was a serious problem for new users with a 
lot of disk space.
- Internal changes to the web interface which will make it easier to implement 
WebDAV in plugins. Requires a new version of XMLSpider.
- Persist overall total output/input bytes in the database.
- Translation updates and minor english string fixes.
- Some workarounds for cruft left in the database by old bugs or by GCJ.
- Snoop callbacks for data and metadata, will allow easier listing of the files 
on a freesite, or make it easier to write things like KeyExplorer.
- Saces' multi-container site insert code: many changes merged, still not 
enabled by default because it still leaks stuff in the database.
- Can specify in FCP what compression codecs to use.
- Internal and build fixes and more comments and javadocs.

Thanks to:

saces
toad
Tommy[D]
Markus
3BUIb3S50i
platy
sdiz
Artefact2
infinity0
juiceman


signature.asc
Description: This is a digitally signed message part.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1228

[Matthew Toseland]

Freenet 0.7 build 1228 is now available. Apologies for not posting a changelog 
for 1227, I will now. Please upgrade ASAP as 1228 is mandatory on Friday and 
warns the user about a critical security problem in Sun JVMs.
1228:
- Warn user about, and refuse to load plugins handling XML, a severe 
vulnerability in Java.
- Some robustness code in the client layer (half-deleted stuff left over by 
bugs causing problems).
- Avoid node-to-node text messages leaking memory when peers are removed.
- If we have peers, set the completed-the-first-time-wizard flag to true. This 
affects encryption of the client layer database, and also whether we show the 
wizard on later startups before managing to connect.
- Fix a thread leak in FCP, usually triggered by FMS.
- Remove TargetNodeName in FCP message ReceivedN2NFeedMessage.
- Minor improvement to javadocs generation, and indenting.

infinity0
ljb
toad

1227:
- Fix failure to start when trying to defrag a big node.db4o.
- Some robustness code in the client layer (half-deleted stuff left over by 
bugs causing problems).
- Fix ?forcedownload headers. On some systems, forcedownload was not working.
- New FCP messages for plugin management: LoadPlugin, RemovePlugin, 
ReloadPlugin, response message PluginRemoved.
- Some refactoring, logging and minor internal changes.
- Fix a wierd NPE breaking site inserts when heavy logging is enabled.

saces
volodya
toad

With regards to the JVM thing, Windows users will generally have auto-update 
for their JVM; Linux and Mac users are more problematic, initially the 
vulnerability was much less serious so may not be deployed quickly, also it 
might be possible to fix it without increasing the JVM version in which case we 
would not be able to detect the fix...


signature.asc
Description: This is a digitally signed message part.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1226

[Matthew Toseland]

Freenet 0.7 build 1225 is now available (1224 originally, 1225 and 1226 fix 
regressions, sorry). Many changes, the most important ones being:
- Finally fix the attack identified by The Register *four years* ago, by not 
caching data requested either locally or nearby. This is also necessary in 
preparation for Bloom filter sharing. This had a number of knock-on effects 
including changes to HTL and new caches to minimise performance losses.
- Two new stores, the client cache and the slashdot cache. The client cache 
stores mostly freesites visited through fproxy, and persists but can be 
encrypted. The slashdot cache stores anything that doesn't go into the other 
caches, but only for 30 minutes.
- Encrypt the client cache and downloads/uploads database (node.db4o), 
according to the physical security level.
- Physical security levels now: LOW (don't encrypt anything), NORMAL (encrypt 
and keep the keys in master.keys), HIGH (need a password to unlock master 
keys), or MAXIMUM (no persistence).
- Fix the panic button, this now works.
- Set the datastore initially to a 10MB RAM-only store, set it to salted-hash 
in the post-install wizard when the size is chosen, automatically migrate the 
data. Avoids a lot of unnecessary disk accesses on new installs.
- Support defragmenting node.db4o. There are two config settings, the first is 
whether to defrag on startup, the second is whether to reset the first setting. 
Both are true by default.
- Aggregate download notifications on the home page. This is the tip of the 
iceberg of ljb's changes to user-alerts, introducing events...
- Changes to ULPRs behaviour. ULPRs are a mechanism whereby a node can remember 
recent requests and if it finds the data nodes have been looking for later on, 
it can forward it to those nodes originally looking for it. This helps to 
improve performance of FMS, Frost, etc. Some security issues have been fixed 
with ULPRs, which will now generally respond faster, also the slashdot cache 
was needed to make ULPRs work with the caching changes.
- Some internal fixes and optimisations.

Freenet will attempt to defragment your node.db4o on startup, and then if your 
physical security level is NORMAL, and there is enough disk space or you are 
running on java 1.5, it will attempt to encrypt it (disk space usage being 
equal to the size of node.db4o). You can turn this off using the following 
settings (in freenet.ini, shut down the node first, then edit the freenet.ini, 
then update the node, then start it back up):
node.defragDatabaseOnStartup=false
node.autoChangeDatabaseEncryption=false

It is likely that there will be a small performance reduction in the short term 
as a result of these changes, and there may be an adjustment shock due to the 
HTL changes. However Bloom filter sharing will greatly improve performance, and 
the sooner we make the caching changes the more data will be flagged as usable 
by Bloom filter sharing (as we cannot share data that might have been requested 
locally).

If you find any bugs, please let us know, post to the bug tracker at 
https://bugs.freenetproject.org/ (create an account first so we can contact 
you), or report them to the support list. Thanks!

Please upgrade! Thanks. This build is not mandatory but it should have been; 
the next one probably will make this one mandatory.


signature.asc
Description: This is a digitally signed message part.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1217

[Matthew Toseland]

Freenet 0.7 build 1217 is now available, please upgrade. This build will 
probably be 0.7.5-beta1. Changes are minimal:
- Translation updates to German and Italian, change Chat and Chat Forums to 
Discussion and Forums.
- Work around overflowing progress bars.
- Link to the FAFS pages on Frost and FMS on the Forums page.
- Fix the fetch a key box fetching a key and forcing its mime type to , so 
that web pages would show a type warning.
- Put a comment in when filtering a page and truncating it, so that the cause 
is more obvious. E.g. when there is an extra  in human-edited HTML, this can 
cause the whole page to get chopped off.


signature.asc
Description: This is a digitally signed message part.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1218

[Matthew Toseland]

Freenet 0.7 build 1218: last minute fixes:
- Don't use javascript on the progress page on Chrome and Safari, as it doesn't 
work (the progress page stalls).
- Bookmark edition updates.
- Fix the confirmation page for setting high friends security level, was 
showing the wrong string.
- Spell compromise with an s.


signature.asc
Description: This is a digitally signed message part.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

Re: [freenet-support] Freenet 0.7 build 1211, 1212

[Matthew Toseland]

On Monday 01 June 2009 22:33:29 SmallSister development wrote:
  Another issue: I have several downloads that report being at 100%,
  but they'll never report completed; even over freenet restarts.
  This could be caused by a freenet crash while in the process of
  writing out the files. (using db4o)
  
  When were they added? If you add a new download for the same key (for
  one of them), do they both complete?
 
 I am still waiting (after more than 24 hours):
   SizeProgress
 New   Unknown 16 (100%??)
 Old   87.1 KiB100%
 
 I started the original downloads somewhere around the 1207, 1208 build
 (not the first distributed build with db4o, but some later build.) There
 are some 470 keys on the download page.

I think this might be a bug recently fixed in git. Please run update.sh testing 
and then add the same key again. What happens to the requests? If you open that 
key in fproxy (without queueing it), does the fetch stall indefinitely? (It 
does in the bug I fixed, which particularly affects XMLSpider). This is caused 
by us mishandling corrupted compressed data...


signature.asc
Description: This is a digitally signed message part.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

Re: [freenet-support] Freenet 0.7 build 1211, 1212

[Matthew Toseland]

On Wednesday 27 May 2009 15:41:47 SmallSister development wrote:
 Matthew Toseland wrote:
  Freenet 0.7 build 1212 is now available. Please upgrade.
 [...]
  1212: - Recently we have had several reports of bogus claims by peers
   that the auto-update key has been blown. The nodes would not
  transfer the revocation key, because they didn't have it, they would
  just cause panic. This appears to be a bug or a malicious attack. We
  now give peers a reasonable time to start transferring the
  auto-update revocation key, and if they don't, we cancel the alert
  and let the update through. If 3 of our peers have failed to transfer
  the key, we still treat that as broken.
 
 I had three peers with the message before upgrading to 1211...
 
 Another issue: I have several downloads that report being at 100%, but
 they'll never report completed; even over freenet restarts. This could
 be caused by a freenet crash while in the process of writing out the
 files. (using db4o)

When were they added? If you add a new download for the same key (for one of 
them), do they both complete?


signature.asc
Description: This is a digitally signed message part.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1210

[Matthew Toseland]

Freenet 0.7 build 1210 is out. Sorry for the delay in announcing it. It will 
be mandatory at midnight. There was also a minor git mess relating to 
tagging - it was released from a local branch because I forgot to push before 
tagging, although it has been merged back into master now. If you are 
building from source you need to either clone the tag or do git fetch -t 
origin to fetch orphan tags.

Changes:
- Startup fix: you can now move the persistent temp directory, as long as you 
update freenet.ini accordingly. This is also an issue when moving the node.
- Startup fix: Deleting the persistent temp dir and moving the node failed to 
startup
- Let the node start up even if the database is unrecoverably corrupted. Tell 
the user what the situation is, and refuse to do any persistent requests or 
inserts.
- Don't commit every database job immediately, unless we need to for memory 
reasons (some job types), for UI reasons (e.g. starting a request), every 30 
seconds, or after every job if the new alwaysCommit config option is enabled. 
This should reduce disk I/O and improve performance significantly on some 
systems with big queues.
- Fix various client layer bugs, in freesite inserts and other things.
- SSK insert collision handling fix - store the data as well as forwarding it.
- Minor crypto optimisation in handling SSKs.
- First time wizard: remove the memory limit selection page, add a page to ask 
the user whether to auto-update Freenet and whether to load JSTUN and UPnP 
(this may reduce performance on a new node *slightly*, but we don't want to 
have to ask in the installer), warn the user about monthly bandwidth usage, 
always show the bandwidth and store size pages, minor fixes.
- Enable javascript support by default in fproxy, minor fixes, get rid of the 
MIME type column on the queue page in simple mode, more tolerant parsing of 
noderefs and bulk download lists.
- Fix various connection level bugs, which hopefully once everyone has 1210 
will fix the forced disconnected because not acknowledging packets bug. In 
simulation this fixes Unmatchable packet from blah, which really 
shouldn't happen!
- Minor plugin code improvements (new APIs, require XMLLibrarian version 22).
- Internal bugfix which hopefully only was breaking IPv6 address detection 
from peers.
- Internal bugfix relating to moving files, might have caused bugs in several 
places.
- Don't let default changes override the local option for whether to 
preallocate the store.
- Lots of internal cleanup and some work on the many-nodes-one-VM simulators / 
automated network tests, including a new long-term data retrievability test.

Also, the XMLLibrarian plugin now shows a search's progress without using 
javascript, and immediately from typing a search on the homepage (thanks 
mikeb). KeyExplorer 0.4 beta is out, and there has been more work on the WoT 
web of trust plugin (thanks saces). A bugfix in the UPnP plugin, and as 
previously mentioned we deployed the new windows installer (thanks Zero3).

Thanks to:
sdiz
3BUIb3S50i
platy
nextgens
saces
juiceman


signature.asc
Description: This is a digitally signed message part.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1209

[Dennis Nezic]

On Fri, 1 May 2009 11:23:39 -0400, Dennis Nezic wrote:
> On Thu, 30 Apr 2009 19:08:30 +0100, Matthew Toseland wrote:
> > In other news, Freenet has moved from using Subversion to keep its
> > source code to Git.
> 
> What's the link to the git repository? (Maybe the website should be
> updated to include it?)

git clone git://github.com/freenet/fred-staging.git

or

git clone git://github.com/freenet/fred-official.git

Excellent.

Re: [freenet-support] Freenet 0.7 build 1209

[Dennis Nezic]

On Thu, 30 Apr 2009 19:08:30 +0100, Matthew Toseland wrote:
 In other news, Freenet has moved from using Subversion to keep its
 source code to Git.

What's the link to the git repository? (Maybe the website should be
updated to include it?)
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

Re: [freenet-support] Freenet 0.7 build 1209

[Dennis Nezic]

On Fri, 1 May 2009 11:23:39 -0400, Dennis Nezic wrote:
 On Thu, 30 Apr 2009 19:08:30 +0100, Matthew Toseland wrote:
  In other news, Freenet has moved from using Subversion to keep its
  source code to Git.
 
 What's the link to the git repository? (Maybe the website should be
 updated to include it?)

git clone git://github.com/freenet/fred-staging.git

or

git clone git://github.com/freenet/fred-official.git

Excellent.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1209

[Matthew Toseland]

Freenet 0.7 build 1209 is out (sorry it took so long, I understand there were 
some fairly bad bugs in 1208). Please upgrade. Lots of changes, including:
- Lots of bugfixes related to the db4o merge.
- A progress page shown when a page or file fetched through fproxy will take 
more than 5 seconds to load. This works both with and without javascript.
- Various improvements to the USK polling code. This is used to poll recently 
visited freesites for updates, to check which slot to insert to when 
inserting a USK, to find ARKs for disconnected nodes and so on.
- A few changes to FCP (SendingToNetwork, UnsubscribeUSK, changes to 
SubscribeUSK behaviour and SubscribedUSKUpdate contents).
- Substantial work on multi-container freesites, but this is not yet the 
default. Currently, all freesites have one container which is an up to 2MB 
tar.lzma file, and then all other files are inserted separately. The new code 
should make large freesites (and hg inserts) much more efficient.
- Various plugin infrastructure improvements.
- A new content filter for BMP images, this is not yet wired in but hopefully 
will be soon.
- A bugfix in ULPRs (our mechanism for automatic propagation of widely polled 
data).
- A bugfix in adding nodes from opennet (path folding or announcement), and a 
minor change to opennet behaviour on restarting.
- Make the javascript enabled config option not be forgotten on restarting the 
node.
- Lots of minor optimisations, code cleanups etc.

Other recent work:

XMLLibrarian now supports asynchronous searches. Searches from the homepage 
are not asynchronous yet, but if you let a search complete, or open 
XMLLibrarian from the plugins page, it will (if you have javascript turned 
on) show you the progress of the search in real time rather than just 
blocking your browser for minutes as it does now.

KeyExplorer now supports downloading manifest zip files, and has many other 
improvements. (KeyExplorer is a tool for investigating exactly how freesites 
etc are put together).

A bugfix in the auto-discovery-on-a-LAN plugin (MDNSDiscovery).

Much work on native 64-bit FEC libraries. Splitfiles are encoded using Forward 
Error Correction codes to add redundancy so that if a few blocks drop out of 
the network, they can be reconstructed from the rest, and reinserted. 
Unfortunately this operation was dramatically slower on 64-bit systems. The 
next freenet-ext.jar (not released yet) will fix this. Much of this was 
actually before 1208 was released, but I have recently got around to 
reviewing the code.

Increase the initial memory limit in the installer to 192MB.

Lots of people have contributed to this build, I have indicated major 
contributions next to some of them:
bombe
dconrado
infinity0
juiceman
kurmiashish (BMP filter)
luke771 (italian translation)
mikeb (XMLLibrarian changes)
nextgens
saces (multi-container freesites and KeyExplorer)
sashee (javascript on the progress page)
sdiz (loads of stuff, help with git migration)
sich (french translation)
thesnark
toad
tommy
xor (freetalk, not mentioned above but really important)
zothar

In other news, Freenet has moved from using Subversion to keep its source code 
to Git. We are hoping to release at least a beta of 0.8.0 in the near future 
(i.e. before the money runs out). If you find any bugs, please let us know, 
preferably via the bug tracker: https://bugs.freenetproject.org/

Thanks for using Freenet!
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 835 bytes
Desc: This is a digitally signed message part.
URL: 

[freenet-support] Freenet 0.7 build 1209

[Matthew Toseland]

Freenet 0.7 build 1209 is out (sorry it took so long, I understand there were 
some fairly bad bugs in 1208). Please upgrade. Lots of changes, including:
- Lots of bugfixes related to the db4o merge.
- A progress page shown when a page or file fetched through fproxy will take 
more than 5 seconds to load. This works both with and without javascript.
- Various improvements to the USK polling code. This is used to poll recently 
visited freesites for updates, to check which slot to insert to when 
inserting a USK, to find ARKs for disconnected nodes and so on.
- A few changes to FCP (SendingToNetwork, UnsubscribeUSK, changes to 
SubscribeUSK behaviour and SubscribedUSKUpdate contents).
- Substantial work on multi-container freesites, but this is not yet the 
default. Currently, all freesites have one container which is an up to 2MB 
tar.lzma file, and then all other files are inserted separately. The new code 
should make large freesites (and hg inserts) much more efficient.
- Various plugin infrastructure improvements.
- A new content filter for BMP images, this is not yet wired in but hopefully 
will be soon.
- A bugfix in ULPRs (our mechanism for automatic propagation of widely polled 
data).
- A bugfix in adding nodes from opennet (path folding or announcement), and a 
minor change to opennet behaviour on restarting.
- Make the javascript enabled config option not be forgotten on restarting the 
node.
- Lots of minor optimisations, code cleanups etc.

Other recent work:

XMLLibrarian now supports asynchronous searches. Searches from the homepage 
are not asynchronous yet, but if you let a search complete, or open 
XMLLibrarian from the plugins page, it will (if you have javascript turned 
on) show you the progress of the search in real time rather than just 
blocking your browser for minutes as it does now.

KeyExplorer now supports downloading manifest zip files, and has many other 
improvements. (KeyExplorer is a tool for investigating exactly how freesites 
etc are put together).

A bugfix in the auto-discovery-on-a-LAN plugin (MDNSDiscovery).

Much work on native 64-bit FEC libraries. Splitfiles are encoded using Forward 
Error Correction codes to add redundancy so that if a few blocks drop out of 
the network, they can be reconstructed from the rest, and reinserted. 
Unfortunately this operation was dramatically slower on 64-bit systems. The 
next freenet-ext.jar (not released yet) will fix this. Much of this was 
actually before 1208 was released, but I have recently got around to 
reviewing the code.

Increase the initial memory limit in the installer to 192MB.

Lots of people have contributed to this build, I have indicated major 
contributions next to some of them:
bombe
dconrado
infinity0
juiceman
kurmiashish (BMP filter)
luke771 (italian translation)
mikeb (XMLLibrarian changes)
nextgens
saces (multi-container freesites and KeyExplorer)
sashee (javascript on the progress page)
sdiz (loads of stuff, help with git migration)
sich (french translation)
thesnark
toad
tommy
xor (freetalk, not mentioned above but really important)
zothar

In other news, Freenet has moved from using Subversion to keep its source code 
to Git. We are hoping to release at least a beta of 0.8.0 in the near future 
(i.e. before the money runs out). If you find any bugs, please let us know, 
preferably via the bug tracker: https://bugs.freenetproject.org/

Thanks for using Freenet!


signature.asc
Description: This is a digitally signed message part.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1207 and 1208: DB4O MERGED!

[Matthew Toseland]

I released 1207 yesterday, but didn't insert it to the auto-updater. At the 
time new nodes were failing to load any of the default plugins. I discovered 
that there was a bug breaking FCP inserts in 1207, so fixed it and released 
1208, and have now started both inserting to the auto-updater.

Please upgrade! In Freenet 0.7 build 1207, the db4o branch was finally merged 
to stable (please get 1208 because of the above bug). The db4o branch uses a 
database to store the status of persistent requests, radically reducing 
memory usage and making requests resume instantly on startup. Like most big 
database driven projects, this massively overran, taking nearly a year (on 
and off), but it is finally merged now (and fairly stable IMHO). It should 
Just Work, automatically moving your existing queue over, and should be 
reasonably stable, but obviously there are likely to be new bugs, and we'd 
like to hear about them. Also, it is possible that bugs will result in data 
being left over in your node.db4o concerning previous requests; generally 
when this happens there will be ERRORs in the logs (NullPointerException's 
usually); if you are very paranoid this may be a problem for you as db4o does 
not implement any form of garbage collection, so we have to remove the data 
manually and if a bug breaks that, it is possible to leak old data, which 
might conceivably have privacy implications.

Apart from the db4o branch, 1207 also includes various other changes:
- A proper numerical version for plugins, which can be easily compared by the 
node. Most plugins have this new interface, FredPluginRealVersioned.
- This enables us to have minimum versions for each official plugin. Hence 
when you upgrade from a pre-db4o node, some of your old plugins 
(ThawIndexBrowser, KeyExplorer, XMLSpider) will fail to load, but Freenet 
will offer to re-download them for you from the project webserver. (In future 
we will support plugin update over Freenet, this is a step towards it, but 
not yet).
- SSK inserts were sometimes finishing but not completing properly, this may 
have resulted in a very gradual slowdown of the node over time.
- Various bugfixes in fproxy including: preserve ?max-size= through links, 
make the form to restart after a config change work, allow adding more than 
one node at a time on the darknet page.
- Allow docName's within KSKs (that is, allow KSKs to be "directories" - 
KSK at sitename/file-in-site).
- Salted hash datastore (the default): Make sure the datastore config file 
gets written to disk, if it is corrupted on startup create a new one.
- Fix a rare division by zero on the stats page, change the calculation of 
darknet network sizes slightly.
- Fix a bug in the security levels code. (I'm not sure why this didn't happen 
in practice for more people, it shouldn't have a major impact but you may as 
well check that your security levels are set to what you want).
- Fix calculation of how many FEC decode threads to use on Windows.
- Record the ARK number for the opennet noderef to disk when we insert.
- Some minor USK polling changes.
- Show whether a node is a seednode in the FCP commands for listing nodes.
- Various optimisations in the random number generator and other places.
- More unit tests.
- Remove various unused code, other minor internal changes.

Thanks to (I hope I haven't missed anyone out):
j16sdiz
bombe
sashee
infinity0
nextgens
p0s
toad

Meanwhile, Artefact2 has contributed a patch to make XMLSpider (the crawler 
that generates indexes for the search box on the homepage) able to run at 
configurable times of day, and sdiz has made some more minor changes to the 
spider. Zero3 continues his work on the new windows installer, which 
hopefully we will be able to deploy before 0.8, and p0s continues to work on 
Freetalk/WoT, which are rapidly becoming usable.

PLEASE UPGRADE! And if (when?) you find bugs, please report them, ideally to 
the bug tracker, but alternatively to the mailing lists.
https://bugs.freenetproject.org/
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 835 bytes
Desc: This is a digitally signed message part.
URL: 

[freenet-support] Freenet 0.7 build 1207 and 1208: DB4O MERGED!

[Matthew Toseland]

I released 1207 yesterday, but didn't insert it to the auto-updater. At the 
time new nodes were failing to load any of the default plugins. I discovered 
that there was a bug breaking FCP inserts in 1207, so fixed it and released 
1208, and have now started both inserting to the auto-updater.

Please upgrade! In Freenet 0.7 build 1207, the db4o branch was finally merged 
to stable (please get 1208 because of the above bug). The db4o branch uses a 
database to store the status of persistent requests, radically reducing 
memory usage and making requests resume instantly on startup. Like most big 
database driven projects, this massively overran, taking nearly a year (on 
and off), but it is finally merged now (and fairly stable IMHO). It should 
Just Work, automatically moving your existing queue over, and should be 
reasonably stable, but obviously there are likely to be new bugs, and we'd 
like to hear about them. Also, it is possible that bugs will result in data 
being left over in your node.db4o concerning previous requests; generally 
when this happens there will be ERRORs in the logs (NullPointerException's 
usually); if you are very paranoid this may be a problem for you as db4o does 
not implement any form of garbage collection, so we have to remove the data 
manually and if a bug breaks that, it is possible to leak old data, which 
might conceivably have privacy implications.

Apart from the db4o branch, 1207 also includes various other changes:
- A proper numerical version for plugins, which can be easily compared by the 
node. Most plugins have this new interface, FredPluginRealVersioned.
- This enables us to have minimum versions for each official plugin. Hence 
when you upgrade from a pre-db4o node, some of your old plugins 
(ThawIndexBrowser, KeyExplorer, XMLSpider) will fail to load, but Freenet 
will offer to re-download them for you from the project webserver. (In future 
we will support plugin update over Freenet, this is a step towards it, but 
not yet).
- SSK inserts were sometimes finishing but not completing properly, this may 
have resulted in a very gradual slowdown of the node over time.
- Various bugfixes in fproxy including: preserve ?max-size= through links, 
make the form to restart after a config change work, allow adding more than 
one node at a time on the darknet page.
- Allow docName's within KSKs (that is, allow KSKs to be directories - 
k...@sitename/file-in-site).
- Salted hash datastore (the default): Make sure the datastore config file 
gets written to disk, if it is corrupted on startup create a new one.
- Fix a rare division by zero on the stats page, change the calculation of 
darknet network sizes slightly.
- Fix a bug in the security levels code. (I'm not sure why this didn't happen 
in practice for more people, it shouldn't have a major impact but you may as 
well check that your security levels are set to what you want).
- Fix calculation of how many FEC decode threads to use on Windows.
- Record the ARK number for the opennet noderef to disk when we insert.
- Some minor USK polling changes.
- Show whether a node is a seednode in the FCP commands for listing nodes.
- Various optimisations in the random number generator and other places.
- More unit tests.
- Remove various unused code, other minor internal changes.

Thanks to (I hope I haven't missed anyone out):
j16sdiz
bombe
sashee
infinity0
nextgens
p0s
toad

Meanwhile, Artefact2 has contributed a patch to make XMLSpider (the crawler 
that generates indexes for the search box on the homepage) able to run at 
configurable times of day, and sdiz has made some more minor changes to the 
spider. Zero3 continues his work on the new windows installer, which 
hopefully we will be able to deploy before 0.8, and p0s continues to work on 
Freetalk/WoT, which are rapidly becoming usable.

PLEASE UPGRADE! And if (when?) you find bugs, please report them, ideally to 
the bug tracker, but alternatively to the mailing lists.
https://bugs.freenetproject.org/


signature.asc
Description: This is a digitally signed message part.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1205

[Matthew Toseland]

Freenet 0.7 build 1205 is now available. Please upgrade. Mostly this is 
optimisations and minor bugfixes. There are also some changes related to 
inserting big freesites (or librarian indexes), backported from the db4o 
branch, and some minor changes.

If you find a bug, please let us know! Report it on the support mailing list, 
or the bug tracker at https://bugs.freenetproject.org/ . Thanks.
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 835 bytes
Desc: This is a digitally signed message part.
URL: 

[freenet-support] Freenet 0.7 build 1205

[Matthew Toseland]

Freenet 0.7 build 1205 is now available. Please upgrade. Mostly this is 
optimisations and minor bugfixes. There are also some changes related to 
inserting big freesites (or librarian indexes), backported from the db4o 
branch, and some minor changes.

If you find a bug, please let us know! Report it on the support mailing list, 
or the bug tracker at https://bugs.freenetproject.org/ . Thanks.


signature.asc
Description: This is a digitally signed message part.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1203

[Matthew Toseland]

On Saturday 24 January 2009 18:18, Dennis Nezic wrote:
> On Sat, 24 Jan 2009 18:07:24 +, Matthew Toseland wrote:
> > fproxy can still be probed for, just not individual sites... We
> > should also warn about it in the README...
> 
> You mean the executable/jar can still be probed for on the filesystem?
> Maybe. But it can't be done via the network interface, if that iptables
> rule is in place--the port would appear as dead as any other unused
> port, as far as anyone except freenetuid would know.

I mean javascript port scanning. But yes you can do it with iptables rules.
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1203

[Matthew Toseland]

On Saturday 24 January 2009 17:41, Dennis Nezic wrote:
> On Sat, 24 Jan 2009 13:05:41 +, Matthew Toseland wrote:
> > There have been some question marks over whether it is possible to
> > load an image from an external domain and get a callback when it is
> > loaded - if so, it may be possible to time fetches of specific sites
> > from javascript on an unrelated site. Meaning running a web browser
> > on a system with access to fproxy is dangerous. I haven't tested
> > this, maybe you'd like to?
> 
> It's a well known attack--"cache timing attacks". Pretty similar to
> css-history attacks. And it's also not hard to prevent. (For history
> attacks, simply disable history in your freenet profile.) For cache
> attacks, simply restrict access to fproxy to a separate freenet user on
> your system. (And, of course, do not use that user to surf the
> dangerous web--unless, of course, you use a safe browser, like one with
> javascript disabled. Javascript is, after all, the root of all
> (website) evil.)

The cross-domain rules don't prevent it then?
> 
> Fproxy access can be restricted on a per-user basis very simply with
> iptables:
> 
> iptables -A OUTPUT -p tcp --dport  -m owner ! --uid-owner
> $FREENETUID -j DROP

We have different definitions of easy! This is a good reason IMHO to not 
include fproxy by default once we have a dedicated browser. And even if we do 
that, fproxy can still be probed for, just not individual sites... We should 
also warn about it in the README...
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1203

[Dennis Nezic]

On Sat, 24 Jan 2009 18:07:24 +, Matthew Toseland wrote:
> fproxy can still be probed for, just not individual sites... We
> should also warn about it in the README...

You mean the executable/jar can still be probed for on the filesystem?
Maybe. But it can't be done via the network interface, if that iptables
rule is in place--the port would appear as dead as any other unused
port, as far as anyone except freenetuid would know.

[freenet-support] Freenet 0.7 build 1203

[Matthew Toseland]

On Friday 23 January 2009 01:53, [Anon] Anon User > wrote:
> -BEGIN TYPE III ANONYMOUS MESSAGE-
> Message-type: plaintext
> 
> In  Matthew Toseland  wrote:
> 
> >>> > > The real solution to browser history stealing is simply to use a 
separate
> >>> > > browser for Freenet than the one you use for the wider web. We now 
warn 
> > users
> >>> > > about this at the beginning of the first time wizard.
> >> > 
> >> > ^^ That is one more reason for the suggestion of a separate UI that I
> >> > made on freenet.uservoice.com. 
> > 
> > This is now largely accepted in the development team as a long-term goal. 
> > saces is working on something like this based on wxWindows. However, a 
> > dedicated browser/client (somewhere in between probably, e.g. there is no 
> > point in converting all the config pages to tabbed dialogs) would likely 
be a 
> > significant amount of work and require a significant amount of 
maintenance. 
> > So it's not a priority for 0.8.0.
> 
> I hope that as Freenet moves to this new UI, somewhere in the days 0.8+++, 
that
> plain old browser support is not discontinued.  There's plenty of us who 
know how
> to be secure with a browser and don't mind doing things like using a 
separate
> profile for the purpose with tighter settings.

And with various "features" turned off, and with lots of connections enabled, 
and so on.

There have been some question marks over whether it is possible to load an 
image from an external domain and get a callback when it is loaded - if so, 
it may be possible to time fetches of specific sites from javascript on an 
unrelated site. Meaning running a web browser on a system with access to 
fproxy is dangerous. I haven't tested this, maybe you'd like to?
> 
> I think there's a lot of use who prefer to stay with that functionality 
rather
> than a shiny new whiz-bang UI.

Fproxy will still be available as a plugin in any case.
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1203

[Dennis Nezic]

On Sat, 24 Jan 2009 13:05:41 +, Matthew Toseland wrote:
> Meaning running a web browser on a system with access to fproxy is
> dangerous. I haven't tested this, maybe you'd like to?

I would imagine that the same problem exists on a system with access to
fcp? or telnet if it's enabled? Ie. aren't flash/javascript/
java/insert-evil-scripting-that-doesn't-belong-in-a-document capable
of issuing commands to fcp/telnet?

In which case, the same solution would trivially fix it.

Unless some kind of login-mechanism is implemented in all vectors to
freenet.

[freenet-support] Freenet 0.7 build 1203

[Dennis Nezic]

On Sat, 24 Jan 2009 13:05:41 +, Matthew Toseland wrote:
> There have been some question marks over whether it is possible to
> load an image from an external domain and get a callback when it is
> loaded - if so, it may be possible to time fetches of specific sites
> from javascript on an unrelated site. Meaning running a web browser
> on a system with access to fproxy is dangerous. I haven't tested
> this, maybe you'd like to?

It's a well known attack--"cache timing attacks". Pretty similar to
css-history attacks. And it's also not hard to prevent. (For history
attacks, simply disable history in your freenet profile.) For cache
attacks, simply restrict access to fproxy to a separate freenet user on
your system. (And, of course, do not use that user to surf the
dangerous web--unless, of course, you use a safe browser, like one with
javascript disabled. Javascript is, after all, the root of all
(website) evil.)

Fproxy access can be restricted on a per-user basis very simply with
iptables:

iptables -A OUTPUT -p tcp --dport  -m owner ! --uid-owner
$FREENETUID -j DROP

Re: [freenet-support] Freenet 0.7 build 1203

[[Anon] Anon User]

-BEGIN TYPE III ANONYMOUS MESSAGE-
Message-type: plaintext

In  Matthew Toseland t...@amphibian.dyndns.org wrote:

   The real solution to browser history stealing is simply to use a 
   separate
   browser for Freenet than the one you use for the wider web. We now warn 
 users
   about this at the beginning of the first time wizard.
  
  ^^ That is one more reason for the suggestion of a separate UI that I
  made on freenet.uservoice.com. 
 
 This is now largely accepted in the development team as a long-term goal. 
 saces is working on something like this based on wxWindows. However, a 
 dedicated browser/client (somewhere in between probably, e.g. there is no 
 point in converting all the config pages to tabbed dialogs) would likely be a 
 significant amount of work and require a significant amount of maintenance. 
 So it's not a priority for 0.8.0.

I hope that as Freenet moves to this new UI, somewhere in the days 0.8+++, that
plain old browser support is not discontinued.  There's plenty of us who know 
how
to be secure with a browser and don't mind doing things like using a separate
profile for the purpose with tighter settings.

I think there's a lot of use who prefer to stay with that functionality rather
than a shiny new whiz-bang UI.


-END TYPE III ANONYMOUS MESSAGE-
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

Re: [freenet-support] Freenet 0.7 build 1203

[Matthew Toseland]

On Friday 23 January 2009 01:53, [Anon] Anon User  wrote:
 -BEGIN TYPE III ANONYMOUS MESSAGE-
 Message-type: plaintext
 
 In  Matthew Toseland t...@amphibian.dyndns.org wrote:
 
The real solution to browser history stealing is simply to use a 
separate
browser for Freenet than the one you use for the wider web. We now 
warn 
  users
about this at the beginning of the first time wizard.
   
   ^^ That is one more reason for the suggestion of a separate UI that I
   made on freenet.uservoice.com. 
  
  This is now largely accepted in the development team as a long-term goal. 
  saces is working on something like this based on wxWindows. However, a 
  dedicated browser/client (somewhere in between probably, e.g. there is no 
  point in converting all the config pages to tabbed dialogs) would likely 
be a 
  significant amount of work and require a significant amount of 
maintenance. 
  So it's not a priority for 0.8.0.
 
 I hope that as Freenet moves to this new UI, somewhere in the days 0.8+++, 
that
 plain old browser support is not discontinued.  There's plenty of us who 
know how
 to be secure with a browser and don't mind doing things like using a 
separate
 profile for the purpose with tighter settings.

And with various features turned off, and with lots of connections enabled, 
and so on.

There have been some question marks over whether it is possible to load an 
image from an external domain and get a callback when it is loaded - if so, 
it may be possible to time fetches of specific sites from javascript on an 
unrelated site. Meaning running a web browser on a system with access to 
fproxy is dangerous. I haven't tested this, maybe you'd like to?
 
 I think there's a lot of use who prefer to stay with that functionality 
rather
 than a shiny new whiz-bang UI.

Fproxy will still be available as a plugin in any case.


pgp57uE7WFSp8.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

Re: [freenet-support] Freenet 0.7 build 1203

[Dennis Nezic]

On Sat, 24 Jan 2009 13:05:41 +, Matthew Toseland wrote:
 There have been some question marks over whether it is possible to
 load an image from an external domain and get a callback when it is
 loaded - if so, it may be possible to time fetches of specific sites
 from javascript on an unrelated site. Meaning running a web browser
 on a system with access to fproxy is dangerous. I haven't tested
 this, maybe you'd like to?

It's a well known attack--cache timing attacks. Pretty similar to
css-history attacks. And it's also not hard to prevent. (For history
attacks, simply disable history in your freenet profile.) For cache
attacks, simply restrict access to fproxy to a separate freenet user on
your system. (And, of course, do not use that user to surf the
dangerous web--unless, of course, you use a safe browser, like one with
javascript disabled. Javascript is, after all, the root of all
(website) evil.)

Fproxy access can be restricted on a per-user basis very simply with
iptables:

iptables -A OUTPUT -p tcp --dport  -m owner ! --uid-owner
$FREENETUID -j DROP
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

Re: [freenet-support] Freenet 0.7 build 1203

[Dennis Nezic]

On Sat, 24 Jan 2009 13:05:41 +, Matthew Toseland wrote:
 Meaning running a web browser on a system with access to fproxy is
 dangerous. I haven't tested this, maybe you'd like to?

I would imagine that the same problem exists on a system with access to
fcp? or telnet if it's enabled? Ie. aren't flash/javascript/
java/insert-evil-scripting-that-doesn't-belong-in-a-document capable
of issuing commands to fcp/telnet?

In which case, the same solution would trivially fix it.

Unless some kind of login-mechanism is implemented in all vectors to
freenet.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

Re: [freenet-support] Freenet 0.7 build 1203

[Matthew Toseland]

On Saturday 24 January 2009 17:41, Dennis Nezic wrote:
 On Sat, 24 Jan 2009 13:05:41 +, Matthew Toseland wrote:
  There have been some question marks over whether it is possible to
  load an image from an external domain and get a callback when it is
  loaded - if so, it may be possible to time fetches of specific sites
  from javascript on an unrelated site. Meaning running a web browser
  on a system with access to fproxy is dangerous. I haven't tested
  this, maybe you'd like to?
 
 It's a well known attack--cache timing attacks. Pretty similar to
 css-history attacks. And it's also not hard to prevent. (For history
 attacks, simply disable history in your freenet profile.) For cache
 attacks, simply restrict access to fproxy to a separate freenet user on
 your system. (And, of course, do not use that user to surf the
 dangerous web--unless, of course, you use a safe browser, like one with
 javascript disabled. Javascript is, after all, the root of all
 (website) evil.)

The cross-domain rules don't prevent it then?
 
 Fproxy access can be restricted on a per-user basis very simply with
 iptables:
 
 iptables -A OUTPUT -p tcp --dport  -m owner ! --uid-owner
 $FREENETUID -j DROP

We have different definitions of easy! This is a good reason IMHO to not 
include fproxy by default once we have a dedicated browser. And even if we do 
that, fproxy can still be probed for, just not individual sites... We should 
also warn about it in the README...


pgpjFCgOV4uVo.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

Re: [freenet-support] Freenet 0.7 build 1203

[Dennis Nezic]

On Sat, 24 Jan 2009 18:07:24 +, Matthew Toseland wrote:
 fproxy can still be probed for, just not individual sites... We
 should also warn about it in the README...

You mean the executable/jar can still be probed for on the filesystem?
Maybe. But it can't be done via the network interface, if that iptables
rule is in place--the port would appear as dead as any other unused
port, as far as anyone except freenetuid would know.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

Re: [freenet-support] Freenet 0.7 build 1203

[Matthew Toseland]

On Saturday 24 January 2009 18:18, Dennis Nezic wrote:
 On Sat, 24 Jan 2009 18:07:24 +, Matthew Toseland wrote:
  fproxy can still be probed for, just not individual sites... We
  should also warn about it in the README...
 
 You mean the executable/jar can still be probed for on the filesystem?
 Maybe. But it can't be done via the network interface, if that iptables
 rule is in place--the port would appear as dead as any other unused
 port, as far as anyone except freenetuid would know.

I mean javascript port scanning. But yes you can do it with iptables rules.


pgpjukOL2qI3o.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1203

[Ancoron Luciferis]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1


Matthew Toseland wrote:
> On Thursday 22 January 2009 22:30, Ancoron Luciferis wrote:
>> Matthew Toseland wrote:
>>> Freenet 0.7 build 1203 is now available. Please upgrade.
>>>
>>> The main change in 1203 is that history cloaking is removed. It is very 
> messy
>>> code-wise and does not really solve the problem - for example, if a user
>>> posted the key for something they had inserted, and forgot to remove
>>> the ?secureid= added by history cloaking, a malicious website could then
>>> probe for that key with the secureid.
>>>
>>> The real solution to browser history stealing is simply to use a separate
>>> browser for Freenet than the one you use for the wider web. We now warn 
> users
>>> about this at the beginning of the first time wizard.
>> ^^ That is one more reason for the suggestion of a separate UI that I
>> made on freenet.uservoice.com. 
> 
> This is now largely accepted in the development team as a long-term goal. 
> saces is working on something like this based on wxWindows. However, a 
> dedicated browser/client (somewhere in between probably, e.g. there is no 
> point in converting all the config pages to tabbed dialogs) would likely be a 
> significant amount of work and require a significant amount of maintenance. 

^^ Well, and there I could help out.

> So it's not a priority for 0.8.0.
> 
>> And I feel that I have to restate what I 
>> said there to prefer XUL for such a client as for the following reasons:
>> 1.) widely used already (stable language, although not community driven)
>> 2.) easy to learn (it's just some XML paired with ECMA-Script - even a
>> lazy JEE/web developer like me was able to master that)
> 
> We do actually need some help with javascript, I don't know js...

^^ OK then, I'm ready to help on that for sure.

> 
>> 3.) common look and feel intregrated (most of the users already use some
>> other XUL based apps like Firefox, Thunderbird, aso.)
>> 4.) easy to extend (regarding to plugins, extensions, themes, aso.)
> 
> It will probably require some C++ (or java with mozswing) code, we would want 
> it to speak FCP for various reasons, and we'd want callbacks from incoming 
> FCP packets to some UI elements.

^^ That is why Mozilla created XPCOM. Its interfaces can be called
through privileged JavaScript which is the way all Mozilla apps work.
Today XPCOM components can be written in the following languages:
- - JavaScript
(https://developer.mozilla.org/en/How_to_Build_an_XPCOM_Component_in_Javascript)
- - C++ (https://developer.mozilla.org/en/XPCOM)
- - Java (https://developer.mozilla.org/en/JavaXPCOM)
- - Perl (https://developer.mozilla.org/en/PlXPCOM)
- - Python (https://developer.mozilla.org/en/PyXPCOM)
- - Ruby (https://developer.mozilla.org/en/RbXPCOM)

And that's why I really like it. Having the choice to use the most
simple for what you need.

> 
>> 5.) runs on nearly every platform
>>
>>> There are also some
>>> German translation updates by an anonymous contributor, and some work on 
> the
>>> README and the website.
>>>
>>> Apart from the above, Zero3 has started to commit his new windows 
> installer.
>>> saces has continued to work on his wxFCP project, which hopefully will 
> result
>>> in a custom browser for Freenet, which we may or may not use when it is
>>> finished, and robert has committed a spec file for generating RPMs for
>>> Freenet.
>> ^^ RPM? I would rather need DEB or S5R4 datastreams. JRPM over at
>> sourceforge is a straight forward library for creating/parsing RPMs
>> regardless of the platform it runs on. The software I'm developing on at
>> work makes extensive use of the JRPM library and it is used in many
>> production systems so it can be considered as stable. Also it supports
>> noarch packages. I was already about to start working on something
>> similar for Solaris PKG, as the very basics (CPIO streams) is the same
>> in both the RPM and S5R4 package system. Only the stuff around that
>> differs a lot but even on windows one can unpack a datastream PKG with
>> the standalone windows builds of GNU dd and cpio. So there's not much
>> magic. I don't like projects that deliver just one package format (and
>> RPM is really not my favorite one). When you are planning to release
>> freenet in package formats please do it for all or for none. But I would
>> suspect that creating a good MSI would be the hardest task anyway.
> 
> What are S5R4? We cannot support every package format, and we will have to 
> keep the java installer around because of this fact. But we would like to 
> support the major formats. Let us know if you are interested in helping in 
> any of the areas I have mentioned.

^^ S5R4 is for System 5 Release 4 packages which is the package system
used by Solaris and other Unixes. Here I can help out too as I am
familiar with the Solaris packaging system due to the fact that I work
in a company that makes money with software management.  :)

Regards,

AncoL
-BEGIN PGP 

[freenet-support] Freenet 0.7 build 1203

[[Anon] Anon User]

-BEGIN TYPE III ANONYMOUS MESSAGE-
Message-type: plaintext

In  Matthew Toseland  wrote:

>>> > > The real solution to browser history stealing is simply to use a 
>>> > > separate
>>> > > browser for Freenet than the one you use for the wider web. We now warn 
> users
>>> > > about this at the beginning of the first time wizard.
>> > 
>> > ^^ That is one more reason for the suggestion of a separate UI that I
>> > made on freenet.uservoice.com. 
> 
> This is now largely accepted in the development team as a long-term goal. 
> saces is working on something like this based on wxWindows. However, a 
> dedicated browser/client (somewhere in between probably, e.g. there is no 
> point in converting all the config pages to tabbed dialogs) would likely be a 
> significant amount of work and require a significant amount of maintenance. 
> So it's not a priority for 0.8.0.

I hope that as Freenet moves to this new UI, somewhere in the days 0.8+++, that
plain old browser support is not discontinued.  There's plenty of us who know 
how
to be secure with a browser and don't mind doing things like using a separate
profile for the purpose with tighter settings.

I think there's a lot of use who prefer to stay with that functionality rather
than a shiny new whiz-bang UI.


-END TYPE III ANONYMOUS MESSAGE-

[freenet-support] Freenet 0.7 build 1203

[Matthew Toseland]

On Thursday 22 January 2009 22:30, Ancoron Luciferis wrote:
> Matthew Toseland wrote:
> > Freenet 0.7 build 1203 is now available. Please upgrade.
> >
> > The main change in 1203 is that history cloaking is removed. It is very 
messy
> > code-wise and does not really solve the problem - for example, if a user
> > posted the key for something they had inserted, and forgot to remove
> > the ?secureid= added by history cloaking, a malicious website could then
> > probe for that key with the secureid.
> >
> > The real solution to browser history stealing is simply to use a separate
> > browser for Freenet than the one you use for the wider web. We now warn 
users
> > about this at the beginning of the first time wizard.
> 
> ^^ That is one more reason for the suggestion of a separate UI that I
> made on freenet.uservoice.com. 

This is now largely accepted in the development team as a long-term goal. 
saces is working on something like this based on wxWindows. However, a 
dedicated browser/client (somewhere in between probably, e.g. there is no 
point in converting all the config pages to tabbed dialogs) would likely be a 
significant amount of work and require a significant amount of maintenance. 
So it's not a priority for 0.8.0.

> And I feel that I have to restate what I 
> said there to prefer XUL for such a client as for the following reasons:
> 1.) widely used already (stable language, although not community driven)
> 2.) easy to learn (it's just some XML paired with ECMA-Script - even a
> lazy JEE/web developer like me was able to master that)

We do actually need some help with javascript, I don't know js...

> 3.) common look and feel intregrated (most of the users already use some
> other XUL based apps like Firefox, Thunderbird, aso.)
> 4.) easy to extend (regarding to plugins, extensions, themes, aso.)

It will probably require some C++ (or java with mozswing) code, we would want 
it to speak FCP for various reasons, and we'd want callbacks from incoming 
FCP packets to some UI elements.

> 5.) runs on nearly every platform
> 
> > There are also some
> > German translation updates by an anonymous contributor, and some work on 
the
> > README and the website.
> >
> > Apart from the above, Zero3 has started to commit his new windows 
installer.
> > saces has continued to work on his wxFCP project, which hopefully will 
result
> > in a custom browser for Freenet, which we may or may not use when it is
> > finished, and robert has committed a spec file for generating RPMs for
> > Freenet.
> 
> ^^ RPM? I would rather need DEB or S5R4 datastreams. JRPM over at
> sourceforge is a straight forward library for creating/parsing RPMs
> regardless of the platform it runs on. The software I'm developing on at
> work makes extensive use of the JRPM library and it is used in many
> production systems so it can be considered as stable. Also it supports
> noarch packages. I was already about to start working on something
> similar for Solaris PKG, as the very basics (CPIO streams) is the same
> in both the RPM and S5R4 package system. Only the stuff around that
> differs a lot but even on windows one can unpack a datastream PKG with
> the standalone windows builds of GNU dd and cpio. So there's not much
> magic. I don't like projects that deliver just one package format (and
> RPM is really not my favorite one). When you are planning to release
> freenet in package formats please do it for all or for none. But I would
> suspect that creating a good MSI would be the hardest task anyway.

What are S5R4? We cannot support every package format, and we will have to 
keep the java installer around because of this fact. But we would like to 
support the major formats. Let us know if you are interested in helping in 
any of the areas I have mentioned.
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

Re: [freenet-support] Freenet 0.7 build 1203

[Ancoron Luciferis]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1


Matthew Toseland wrote:
 On Thursday 22 January 2009 22:30, Ancoron Luciferis wrote:
 Matthew Toseland wrote:
 Freenet 0.7 build 1203 is now available. Please upgrade.

 The main change in 1203 is that history cloaking is removed. It is very 
 messy
 code-wise and does not really solve the problem - for example, if a user
 posted the key for something they had inserted, and forgot to remove
 the ?secureid= added by history cloaking, a malicious website could then
 probe for that key with the secureid.

 The real solution to browser history stealing is simply to use a separate
 browser for Freenet than the one you use for the wider web. We now warn 
 users
 about this at the beginning of the first time wizard.
 ^^ That is one more reason for the suggestion of a separate UI that I
 made on freenet.uservoice.com. 
 
 This is now largely accepted in the development team as a long-term goal. 
 saces is working on something like this based on wxWindows. However, a 
 dedicated browser/client (somewhere in between probably, e.g. there is no 
 point in converting all the config pages to tabbed dialogs) would likely be a 
 significant amount of work and require a significant amount of maintenance. 

^^ Well, and there I could help out.

 So it's not a priority for 0.8.0.
 
 And I feel that I have to restate what I 
 said there to prefer XUL for such a client as for the following reasons:
 1.) widely used already (stable language, although not community driven)
 2.) easy to learn (it's just some XML paired with ECMA-Script - even a
 lazy JEE/web developer like me was able to master that)
 
 We do actually need some help with javascript, I don't know js...

^^ OK then, I'm ready to help on that for sure.

 
 3.) common look and feel intregrated (most of the users already use some
 other XUL based apps like Firefox, Thunderbird, aso.)
 4.) easy to extend (regarding to plugins, extensions, themes, aso.)
 
 It will probably require some C++ (or java with mozswing) code, we would want 
 it to speak FCP for various reasons, and we'd want callbacks from incoming 
 FCP packets to some UI elements.

^^ That is why Mozilla created XPCOM. Its interfaces can be called
through privileged JavaScript which is the way all Mozilla apps work.
Today XPCOM components can be written in the following languages:
- - JavaScript
(https://developer.mozilla.org/en/How_to_Build_an_XPCOM_Component_in_Javascript)
- - C++ (https://developer.mozilla.org/en/XPCOM)
- - Java (https://developer.mozilla.org/en/JavaXPCOM)
- - Perl (https://developer.mozilla.org/en/PlXPCOM)
- - Python (https://developer.mozilla.org/en/PyXPCOM)
- - Ruby (https://developer.mozilla.org/en/RbXPCOM)

And that's why I really like it. Having the choice to use the most
simple for what you need.

 
 5.) runs on nearly every platform

 There are also some
 German translation updates by an anonymous contributor, and some work on 
 the
 README and the website.

 Apart from the above, Zero3 has started to commit his new windows 
 installer.
 saces has continued to work on his wxFCP project, which hopefully will 
 result
 in a custom browser for Freenet, which we may or may not use when it is
 finished, and robert has committed a spec file for generating RPMs for
 Freenet.
 ^^ RPM? I would rather need DEB or S5R4 datastreams. JRPM over at
 sourceforge is a straight forward library for creating/parsing RPMs
 regardless of the platform it runs on. The software I'm developing on at
 work makes extensive use of the JRPM library and it is used in many
 production systems so it can be considered as stable. Also it supports
 noarch packages. I was already about to start working on something
 similar for Solaris PKG, as the very basics (CPIO streams) is the same
 in both the RPM and S5R4 package system. Only the stuff around that
 differs a lot but even on windows one can unpack a datastream PKG with
 the standalone windows builds of GNU dd and cpio. So there's not much
 magic. I don't like projects that deliver just one package format (and
 RPM is really not my favorite one). When you are planning to release
 freenet in package formats please do it for all or for none. But I would
 suspect that creating a good MSI would be the hardest task anyway.
 
 What are S5R4? We cannot support every package format, and we will have to 
 keep the java installer around because of this fact. But we would like to 
 support the major formats. Let us know if you are interested in helping in 
 any of the areas I have mentioned.

^^ S5R4 is for System 5 Release 4 packages which is the package system
used by Solaris and other Unixes. Here I can help out too as I am
familiar with the Solaris packaging system due to the fact that I work
in a company that makes money with software management.  :)

Regards,

AncoL
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

[freenet-support] Freenet 0.7 build 1203

[Ancoron Luciferis]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Matthew Toseland wrote:
> Freenet 0.7 build 1203 is now available. Please upgrade.
> 
> The main change in 1203 is that history cloaking is removed. It is very messy 
> code-wise and does not really solve the problem - for example, if a user 
> posted the key for something they had inserted, and forgot to remove 
> the ?secureid= added by history cloaking, a malicious website could then 
> probe for that key with the secureid.
> 
> The real solution to browser history stealing is simply to use a separate 
> browser for Freenet than the one you use for the wider web. We now warn users 
> about this at the beginning of the first time wizard. 

^^ That is one more reason for the suggestion of a separate UI that I
made on freenet.uservoice.com. And I feel that I have to restate what I
said there to prefer XUL for such a client as for the following reasons:
1.) widely used already (stable language, although not community driven)
2.) easy to learn (it's just some XML paired with ECMA-Script - even a
lazy JEE/web developer like me was able to master that)
3.) common look and feel intregrated (most of the users already use some
other XUL based apps like Firefox, Thunderbird, aso.)
4.) easy to extend (regarding to plugins, extensions, themes, aso.)
5.) runs on nearly every platform

> There are also some 
> German translation updates by an anonymous contributor, and some work on the 
> README and the website. 
> 
> Apart from the above, Zero3 has started to commit his new windows installer. 
> saces has continued to work on his wxFCP project, which hopefully will result 
> in a custom browser for Freenet, which we may or may not use when it is 
> finished, and robert has committed a spec file for generating RPMs for 
> Freenet.

^^ RPM? I would rather need DEB or S5R4 datastreams. JRPM over at
sourceforge is a straight forward library for creating/parsing RPMs
regardless of the platform it runs on. The software I'm developing on at
work makes extensive use of the JRPM library and it is used in many
production systems so it can be considered as stable. Also it supports
noarch packages. I was already about to start working on something
similar for Solaris PKG, as the very basics (CPIO streams) is the same
in both the RPM and S5R4 package system. Only the stuff around that
differs a lot but even on windows one can unpack a datastream PKG with
the standalone windows builds of GNU dd and cpio. So there's not much
magic. I don't like projects that deliver just one package format (and
RPM is really not my favorite one). When you are planning to release
freenet in package formats please do it for all or for none. But I would
suspect that creating a good MSI would be the hardest task anyway.

Greetz,

AncoL

> 
> 1202 was related to history cloaking (making it configurable), and 1201 fixed 
> a bug causing the activelinks enabled setting not to be read on startup.
> 
> If you find any bugs, please report them on the bug tracker:
> https://bugs.freenetproject.org/
> 
> Thanks.
> 
> 
> 
> 
> ___
> Support mailing list
> Support at freenetproject.org
> http://news.gmane.org/gmane.network.freenet.support
> Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
> Or mailto:support-request at freenetproject.org?subject=unsubscribe
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkl485QACgkQHwxOsqv2bG3R6wCeMSd/xhz33Lp4VltG1tx31mwQ
XWQAoMWo3uDplXNLHTsTuiV2mMfA5rDV
=kzG2
-END PGP SIGNATURE-

Re: [freenet-support] Freenet 0.7 build 1203

[Ancoron Luciferis]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Matthew Toseland wrote:
 Freenet 0.7 build 1203 is now available. Please upgrade.
 
 The main change in 1203 is that history cloaking is removed. It is very messy 
 code-wise and does not really solve the problem - for example, if a user 
 posted the key for something they had inserted, and forgot to remove 
 the ?secureid= added by history cloaking, a malicious website could then 
 probe for that key with the secureid.
 
 The real solution to browser history stealing is simply to use a separate 
 browser for Freenet than the one you use for the wider web. We now warn users 
 about this at the beginning of the first time wizard. 

^^ That is one more reason for the suggestion of a separate UI that I
made on freenet.uservoice.com. And I feel that I have to restate what I
said there to prefer XUL for such a client as for the following reasons:
1.) widely used already (stable language, although not community driven)
2.) easy to learn (it's just some XML paired with ECMA-Script - even a
lazy JEE/web developer like me was able to master that)
3.) common look and feel intregrated (most of the users already use some
other XUL based apps like Firefox, Thunderbird, aso.)
4.) easy to extend (regarding to plugins, extensions, themes, aso.)
5.) runs on nearly every platform

 There are also some 
 German translation updates by an anonymous contributor, and some work on the 
 README and the website. 
 
 Apart from the above, Zero3 has started to commit his new windows installer. 
 saces has continued to work on his wxFCP project, which hopefully will result 
 in a custom browser for Freenet, which we may or may not use when it is 
 finished, and robert has committed a spec file for generating RPMs for 
 Freenet.

^^ RPM? I would rather need DEB or S5R4 datastreams. JRPM over at
sourceforge is a straight forward library for creating/parsing RPMs
regardless of the platform it runs on. The software I'm developing on at
work makes extensive use of the JRPM library and it is used in many
production systems so it can be considered as stable. Also it supports
noarch packages. I was already about to start working on something
similar for Solaris PKG, as the very basics (CPIO streams) is the same
in both the RPM and S5R4 package system. Only the stuff around that
differs a lot but even on windows one can unpack a datastream PKG with
the standalone windows builds of GNU dd and cpio. So there's not much
magic. I don't like projects that deliver just one package format (and
RPM is really not my favorite one). When you are planning to release
freenet in package formats please do it for all or for none. But I would
suspect that creating a good MSI would be the hardest task anyway.

Greetz,

AncoL

 
 1202 was related to history cloaking (making it configurable), and 1201 fixed 
 a bug causing the activelinks enabled setting not to be read on startup.
 
 If you find any bugs, please report them on the bug tracker:
 https://bugs.freenetproject.org/
 
 Thanks.
 
 
 
 
 ___
 Support mailing list
 Support@freenetproject.org
 http://news.gmane.org/gmane.network.freenet.support
 Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
 Or mailto:support-requ...@freenetproject.org?subject=unsubscribe
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkl485QACgkQHwxOsqv2bG3R6wCeMSd/xhz33Lp4VltG1tx31mwQ
XWQAoMWo3uDplXNLHTsTuiV2mMfA5rDV
=kzG2
-END PGP SIGNATURE-
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

Re: [freenet-support] Freenet 0.7 build 1203

[Matthew Toseland]

On Thursday 22 January 2009 22:30, Ancoron Luciferis wrote:
 Matthew Toseland wrote:
  Freenet 0.7 build 1203 is now available. Please upgrade.
 
  The main change in 1203 is that history cloaking is removed. It is very 
messy
  code-wise and does not really solve the problem - for example, if a user
  posted the key for something they had inserted, and forgot to remove
  the ?secureid= added by history cloaking, a malicious website could then
  probe for that key with the secureid.
 
  The real solution to browser history stealing is simply to use a separate
  browser for Freenet than the one you use for the wider web. We now warn 
users
  about this at the beginning of the first time wizard.
 
 ^^ That is one more reason for the suggestion of a separate UI that I
 made on freenet.uservoice.com. 

This is now largely accepted in the development team as a long-term goal. 
saces is working on something like this based on wxWindows. However, a 
dedicated browser/client (somewhere in between probably, e.g. there is no 
point in converting all the config pages to tabbed dialogs) would likely be a 
significant amount of work and require a significant amount of maintenance. 
So it's not a priority for 0.8.0.

 And I feel that I have to restate what I 
 said there to prefer XUL for such a client as for the following reasons:
 1.) widely used already (stable language, although not community driven)
 2.) easy to learn (it's just some XML paired with ECMA-Script - even a
 lazy JEE/web developer like me was able to master that)

We do actually need some help with javascript, I don't know js...

 3.) common look and feel intregrated (most of the users already use some
 other XUL based apps like Firefox, Thunderbird, aso.)
 4.) easy to extend (regarding to plugins, extensions, themes, aso.)

It will probably require some C++ (or java with mozswing) code, we would want 
it to speak FCP for various reasons, and we'd want callbacks from incoming 
FCP packets to some UI elements.

 5.) runs on nearly every platform
 
  There are also some
  German translation updates by an anonymous contributor, and some work on 
the
  README and the website.
 
  Apart from the above, Zero3 has started to commit his new windows 
installer.
  saces has continued to work on his wxFCP project, which hopefully will 
result
  in a custom browser for Freenet, which we may or may not use when it is
  finished, and robert has committed a spec file for generating RPMs for
  Freenet.
 
 ^^ RPM? I would rather need DEB or S5R4 datastreams. JRPM over at
 sourceforge is a straight forward library for creating/parsing RPMs
 regardless of the platform it runs on. The software I'm developing on at
 work makes extensive use of the JRPM library and it is used in many
 production systems so it can be considered as stable. Also it supports
 noarch packages. I was already about to start working on something
 similar for Solaris PKG, as the very basics (CPIO streams) is the same
 in both the RPM and S5R4 package system. Only the stuff around that
 differs a lot but even on windows one can unpack a datastream PKG with
 the standalone windows builds of GNU dd and cpio. So there's not much
 magic. I don't like projects that deliver just one package format (and
 RPM is really not my favorite one). When you are planning to release
 freenet in package formats please do it for all or for none. But I would
 suspect that creating a good MSI would be the hardest task anyway.

What are S5R4? We cannot support every package format, and we will have to 
keep the java installer around because of this fact. But we would like to 
support the major formats. Let us know if you are interested in helping in 
any of the areas I have mentioned.


pgp9XMOt8wlO6.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1203

[Matthew Toseland]

Freenet 0.7 build 1203 is now available. Please upgrade.

The main change in 1203 is that history cloaking is removed. It is very messy 
code-wise and does not really solve the problem - for example, if a user 
posted the key for something they had inserted, and forgot to remove 
the ?secureid= added by history cloaking, a malicious website could then 
probe for that key with the secureid.

The real solution to browser history stealing is simply to use a separate 
browser for Freenet than the one you use for the wider web. We now warn users 
about this at the beginning of the first time wizard. There are also some 
German translation updates by an anonymous contributor, and some work on the 
README and the website. 

Apart from the above, Zero3 has started to commit his new windows installer. 
saces has continued to work on his wxFCP project, which hopefully will result 
in a custom browser for Freenet, which we may or may not use when it is 
finished, and robert has committed a spec file for generating RPMs for 
Freenet.

1202 was related to history cloaking (making it configurable), and 1201 fixed 
a bug causing the activelinks enabled setting not to be read on startup.

If you find any bugs, please report them on the bug tracker:
https://bugs.freenetproject.org/

Thanks.
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1203

[Matthew Toseland]

Freenet 0.7 build 1203 is now available. Please upgrade.

The main change in 1203 is that history cloaking is removed. It is very messy 
code-wise and does not really solve the problem - for example, if a user 
posted the key for something they had inserted, and forgot to remove 
the ?secureid= added by history cloaking, a malicious website could then 
probe for that key with the secureid.

The real solution to browser history stealing is simply to use a separate 
browser for Freenet than the one you use for the wider web. We now warn users 
about this at the beginning of the first time wizard. There are also some 
German translation updates by an anonymous contributor, and some work on the 
README and the website. 

Apart from the above, Zero3 has started to commit his new windows installer. 
saces has continued to work on his wxFCP project, which hopefully will result 
in a custom browser for Freenet, which we may or may not use when it is 
finished, and robert has committed a spec file for generating RPMs for 
Freenet.

1202 was related to history cloaking (making it configurable), and 1201 fixed 
a bug causing the activelinks enabled setting not to be read on startup.

If you find any bugs, please report them on the bug tracker:
https://bugs.freenetproject.org/

Thanks.


pgp9zEPzw3Ymv.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1199, 1200 and installer changes

[Matthew Toseland]

Freenet 0.7 build 1199, 1200 and 1201 are now available (sorry, I needed some 
changes to Freenet for the new installer...). Please upgrade to 1201. Also, 
there have been some important changes to the installer.

1199:
- Minor changes to turtling: Backoff from nodes which send a turtle transfer 
which then times out, add them to the per-node failure table if it fails 
without timing out.
- Fix some bugs in history cloaking.
- Don't skip the last data block in a splitfile if it is the full expected 
size. This means that splitfiles inserted by 1195 or later, which are padded 
on insert, will download slightly more quickly and shouldn't get stuck at 
100% or slightly over 100%.
- Statistics, logging fixes

1200:
- More history cloaking bugfixes.
- Make activelinks configurable and turn them off by default. Ian thinks they 
are ugly. I'm not sure, any opinions would be welcome. It is however clear 
that they slow down loading the homepage.

1201:
- Fix activelinks - setting wasn't being read on startup. Doh!

Installer:
- Enable history cloaking, update browse.sh / browse.cmd to open the correct 
starter page, minor related fixes.
- Turn off persistent connection support, because by default Firefox (and 
probably other browsers) allows up to 8 non-persistent connections per server 
but only 2 persistent ones.
- Remove the MDNSDiscovery plugin from the default plugins. This plugin 
announces the Freenet node to the LAN in order to make it easy to connect to 
nodes on the same LAN. We do not automatically open Freenet to the LAN, and 
this plugin is the cause of Freenet hanging during shutdown in many cases.

Plugins:
- Fix compile of XMLSpider, add more extensions to the blacklist.
- Fix links (in history cloaking mode, prevent a warning showing up by adding 
the ?secureid= to links) in XMLLibrarian and KeyExplorer.
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1198

[Matthew Toseland]

Freenet 0.7 build 1198 is now available, and will be mandatory at midnight 
GMT. 

The main change is request turtling. This is an attempt to reduce the average 
request latency by cutting off request transfers at 60 seconds - we cancel 
the downstream transfers, and backoff the node, but let it continue to 
transfer at whatever rate it makes, and if the transfer succeeds, we offer it 
to the nodes which have requested the data. There are various limits 
on turtle transfers, and they should be only a minority of requests - but 
hopefully they should cut the latency of the average request significantly, 
at the cost of a small number of requests failing early. If the data is 
queued, it will be propagated after the turtle transfer completes; if it was 
an fproxy request, it will be propagated but of course it won't be displayed.

Apart from turtling, there is also a minor fix to request handling (which 
might just have been part of the bandwidth collapsing bug), and a minor 
tempfile leak fix, and some new stats for turtling, and a fix to block 
transfer statistics.

Please upgrade!


pgppI2FUKXTxd.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1197

[Matthew Toseland]

Freenet 0.7 build 1197 is now available, and is mandatory at midnight. Please 
upgrade. Nodes with auto-deploy-update disabled may have problems, since in 
1195 a bug was introduced which broke the Update Now button, which is fixed 
in 1197. If you are in this situation please either update manually with 
update.sh/update.cmd, or turn on "Automatically install new versions" in the 
updater category (you can turn it back off once the node is running 1197).

Apart from that bugfix and related ones, and a small French update from 
regivanx, I have reverted the recent changes related to long transfer times. 
Freenet should recover and use as much bandwidth as it was using before now. 
If you have adjusted your bandwidth limit to compensate for the problems, 
please change it back. However, this is not the end of the story - it is 
likely that 1198 will have some changes aimed at eliminating really slow 
request transfers, but hopefully these changes will be less disruptive. 
Thanks.
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1196 and current status

[Matthew Toseland]

Freenet 0.7 build 1196 is now available, and is mandatory. It should be 
downloaded by your node automatically. Please upgrade.

We are still working on the request latency optimisations that we started on 
Saturday. We have discovered that the median time taken by a typical CHK 
request is vastly lower than the mean time taken by one, because every so 
often a data transfer takes a long time (many minutes in some cases). In this 
build we have extended the transfer backoff system to try to isolate nodes 
causing this trouble. Apologies for the continued disruption particularly to 
throughput, but if we can crack this one it should greatly improve latency 
and fproxy performance, hopefully permanently.

Thanks for testing Freenet! Freenet remains an unfinished and in-progress 
system, and sometimes the best way to move things forward is to try things 
out on the live network ... None of the recent changes have in any way 
jeopardised anonymity as far as I know. This build also includes a fix for a 
NullPointerException introduced in the stalled request detection code in 
1195, which was largely harmless, and a tweak to some code used by some 
plugins. Upgrade! Let us know if you fail to upgrade or notice any bugs...
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1197

[Matthew Toseland]

Freenet 0.7 build 1197 is now available, and is mandatory at midnight. Please 
upgrade. Nodes with auto-deploy-update disabled may have problems, since in 
1195 a bug was introduced which broke the Update Now button, which is fixed 
in 1197. If you are in this situation please either update manually with 
update.sh/update.cmd, or turn on Automatically install new versions in the 
updater category (you can turn it back off once the node is running 1197).

Apart from that bugfix and related ones, and a small French update from 
regivanx, I have reverted the recent changes related to long transfer times. 
Freenet should recover and use as much bandwidth as it was using before now. 
If you have adjusted your bandwidth limit to compensate for the problems, 
please change it back. However, this is not the end of the story - it is 
likely that 1198 will have some changes aimed at eliminating really slow 
request transfers, but hopefully these changes will be less disruptive. 
Thanks.


pgpmEbgUqoxoC.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1196 and current status

[Matthew Toseland]

Freenet 0.7 build 1196 is now available, and is mandatory. It should be 
downloaded by your node automatically. Please upgrade.

We are still working on the request latency optimisations that we started on 
Saturday. We have discovered that the median time taken by a typical CHK 
request is vastly lower than the mean time taken by one, because every so 
often a data transfer takes a long time (many minutes in some cases). In this 
build we have extended the transfer backoff system to try to isolate nodes 
causing this trouble. Apologies for the continued disruption particularly to 
throughput, but if we can crack this one it should greatly improve latency 
and fproxy performance, hopefully permanently.

Thanks for testing Freenet! Freenet remains an unfinished and in-progress 
system, and sometimes the best way to move things forward is to try things 
out on the live network ... None of the recent changes have in any way 
jeopardised anonymity as far as I know. This build also includes a fix for a 
NullPointerException introduced in the stalled request detection code in 
1195, which was largely harmless, and a tweak to some code used by some 
plugins. Upgrade! Let us know if you fail to upgrade or notice any bugs...


pgpv2T6GLaqkH.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1195 (and warning)

[Matthew Toseland]

Freenet 0.7 build 1195 is now available. Please upgrade ASAP. There are 
several important changes in this build and recently, the first two of which 
have caveats:
- Freenet nodes now accept fewer parallel requests. This is an attempt to 
reduce the average time taken by a successful CHK fetch, which was around 18 
seconds on my node (see your stats page in advanced mode).

CAVEAT: In the short term, this may result in a significant loss of 
performance: once the network has settled, hopefully requests will be much 
faster in the sense of having significantly lower latency (we are aiming for 
an average of 5 seconds), with comparable overall throughput to now. If so, 
this will improve fproxy performance significantly. If this results in 
sustained loss of performance after a short period, we will revert it.

CAVEAT: I accidentally released 1195 as self-mandatory, however in retrospect 
it is probably for the best, so 1195 is self-mandatory. Upgrade or be left 
behind!

- We no longer ship the customised Firefox profile in the installer, due to 
its causing severe problems for a minority of users.

CAVEAT: First, this means that new users will generally use the same browser 
to browse Freenet as to browse the web, possibly resulting in security 
breaches via e.g. browser history stealing. The best solution is to use a 
separate browser for browsing Freenet vs browsing the web. Another solution 
is to turn on history cloaking, see the next item. The second problem is that 
we are now limited by browser connection limits: Web browsers generally have 
a very low limit on the number of connections per server. A partial solution 
is to implement a progress screen when loading a page; this will be done in 
the near future. If you have expertise in Javascript, please contact us, we 
could really do with some help for the next stage.

(End of caveats)

- Some work on the bandwidth collapsing bug. Hopefully it won't be as severe, 
and ERRORs should be logged with enough information to track down the 
original cause.

- Pad the last block of a splitfile before inserting it. This changes the CHKs 
yet again, but should make us immune to any future problems with padding 
algorithms; we have had 4 different padding algorithms, each generating 
different CHKs, and until we implemented a really ugly workaround a few 
builds ago, we had silent data corruption caused by this.

- Minor changes to the web interface and themes, content flter, URI handling, 
connection setup, random number generator initialisation, plugins system, 
unit tests, installer (fix not hiding the freenet user from the login 
screen), website, various minor optimisations, and various code cleanups.

- FCP support for getting the status of all peers, not just darknet peers.

In addition, sdiz has done a massive amount of work on XMLSpider, the plugin 
that generates indexes for the search box on the home page (also known as 
XMLLibrarian), and a significant amount of work on XMLLibrarian. The spider 
in particular is massively improved, using a database and persisting its 
state across restarts, writing indexes only when needed, with a great deal of 
work on optimisation, and many bugfixes. Also, since the last build, I have 
done some work on the database branch, apart from having a week off over 
Christmas. Happy new year, and I apologise for the (necessary) length of this 
announcement!
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1195 (and warning)

[Ian Clarke]

On Sat, Jan 10, 2009 at 4:58 PM, Matthew Toseland
 wrote:
> A partial solution
> is to implement a progress screen when loading a page; this will be done in
> the near future. If you have expertise in Javascript, please contact us, we
> could really do with some help for the next stage.

You should look into GWT for this.

Ian.

-- 
Ian Clarke
CEO, Uprizer Labs
Email: ian at uprizer.com
Ph: +1 512 422 3588
Fax: +1 512 276 6674

[freenet-support] Freenet 0.7 build 1195 (and warning)

[Matthew Toseland]

Freenet 0.7 build 1195 is now available. Please upgrade ASAP. There are 
several important changes in this build and recently, the first two of which 
have caveats:
- Freenet nodes now accept fewer parallel requests. This is an attempt to 
reduce the average time taken by a successful CHK fetch, which was around 18 
seconds on my node (see your stats page in advanced mode).

CAVEAT: In the short term, this may result in a significant loss of 
performance: once the network has settled, hopefully requests will be much 
faster in the sense of having significantly lower latency (we are aiming for 
an average of 5 seconds), with comparable overall throughput to now. If so, 
this will improve fproxy performance significantly. If this results in 
sustained loss of performance after a short period, we will revert it.

CAVEAT: I accidentally released 1195 as self-mandatory, however in retrospect 
it is probably for the best, so 1195 is self-mandatory. Upgrade or be left 
behind!

- We no longer ship the customised Firefox profile in the installer, due to 
its causing severe problems for a minority of users.

CAVEAT: First, this means that new users will generally use the same browser 
to browse Freenet as to browse the web, possibly resulting in security 
breaches via e.g. browser history stealing. The best solution is to use a 
separate browser for browsing Freenet vs browsing the web. Another solution 
is to turn on history cloaking, see the next item. The second problem is that 
we are now limited by browser connection limits: Web browsers generally have 
a very low limit on the number of connections per server. A partial solution 
is to implement a progress screen when loading a page; this will be done in 
the near future. If you have expertise in Javascript, please contact us, we 
could really do with some help for the next stage.

(End of caveats)

- Some work on the bandwidth collapsing bug. Hopefully it won't be as severe, 
and ERRORs should be logged with enough information to track down the 
original cause.

- Pad the last block of a splitfile before inserting it. This changes the CHKs 
yet again, but should make us immune to any future problems with padding 
algorithms; we have had 4 different padding algorithms, each generating 
different CHKs, and until we implemented a really ugly workaround a few 
builds ago, we had silent data corruption caused by this.

- Minor changes to the web interface and themes, content flter, URI handling, 
connection setup, random number generator initialisation, plugins system, 
unit tests, installer (fix not hiding the freenet user from the login 
screen), website, various minor optimisations, and various code cleanups.

- FCP support for getting the status of all peers, not just darknet peers.

In addition, sdiz has done a massive amount of work on XMLSpider, the plugin 
that generates indexes for the search box on the home page (also known as 
XMLLibrarian), and a significant amount of work on XMLLibrarian. The spider 
in particular is massively improved, using a database and persisting its 
state across restarts, writing indexes only when needed, with a great deal of 
work on optimisation, and many bugfixes. Also, since the last build, I have 
done some work on the database branch, apart from having a week off over 
Christmas. Happy new year, and I apologise for the (necessary) length of this 
announcement!


pgpXCS2oXBhHZ.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

Re: [freenet-support] Freenet 0.7 build 1195 (and warning)

[Ian Clarke]

On Sat, Jan 10, 2009 at 4:58 PM, Matthew Toseland
t...@amphibian.dyndns.org wrote:
 A partial solution
 is to implement a progress screen when loading a page; this will be done in
 the near future. If you have expertise in Javascript, please contact us, we
 could really do with some help for the next stage.

You should look into GWT for this.

Ian.

-- 
Ian Clarke
CEO, Uprizer Labs
Email: i...@uprizer.com
Ph: +1 512 422 3588
Fax: +1 512 276 6674
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1194

[Matthew Toseland]

On Wednesday 24 December 2008 18:34, bbackde at googlemail.com wrote:
> I still get this:
> 
> * Probably a bug: please report: 2 peers forcibly disconnected due
> to not acknowledging packets.
> * Probably a bug: please report: 1 peers forcibly disconnected due
> to not acknowledging packets.
> 
> with
> 
> * Freenet 0.7 Build #1194 r24387
> * Freenet-ext Build #26 r23771
> 
> after this
> 
>nodeUptime: 1d21h

We still seem to get it occasionally but IMHO it is no longer a big 
problem ... but it will need to be looked into at some point before 0.8.
> 
> 
> Merry Christmas
> 
> 
> On Wed, Dec 24, 2008 at 17:13, SmallSister development
>  wrote:
> > Matthew Toseland wrote:
> >> Freenet 0.7 build 1194 is now available. Please upgrade, this will be
> >> mandatory on Friday. Some fairly substantial changes:
> >
> > It looks like the infamous "# peers forcibly disconnected" bug is gone
> > in this release, thanks and congratulations! (I see far less package
> > handling related messages in the logfiles.)
> >
> > While I'm at it: Merry Christmas and a Happy New Year!
> >
> > Peter.
> > ___
> > Support mailing list
> > Support at freenetproject.org
> > http://news.gmane.org/gmane.network.freenet.support
> > Unsubscribe at 
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
> > Or mailto:support-request at freenetproject.org?subject=unsubscribe
> >
> 
> 
> 
> -- 
> __
> GnuPG key:   (0x48DBFA8A)
> Keyserver:   pgpkeys.pca.dfn.de
> Fingerprint:
> 477D F057 1BD4 1AE7 8A54 8679 6690 E2EC 48DB FA8A
> __
> ___
> Support mailing list
> Support at freenetproject.org
> http://news.gmane.org/gmane.network.freenet.support
> Unsubscribe at 
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
> Or mailto:support-request at freenetproject.org?subject=unsubscribe
> 
> 
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

Re: [freenet-support] Freenet 0.7 build 1194

[Matthew Toseland]

On Wednesday 24 December 2008 18:34, bbac...@googlemail.com wrote:
 I still get this:
 
 * Probably a bug: please report: 2 peers forcibly disconnected due
 to not acknowledging packets.
 * Probably a bug: please report: 1 peers forcibly disconnected due
 to not acknowledging packets.
 
 with
 
 * Freenet 0.7 Build #1194 r24387
 * Freenet-ext Build #26 r23771
 
 after this
 
nodeUptime: 1d21h

We still seem to get it occasionally but IMHO it is no longer a big 
problem ... but it will need to be looked into at some point before 0.8.
 
 
 Merry Christmas
 
 
 On Wed, Dec 24, 2008 at 17:13, SmallSister development
 smallsis...@xs4all.nl wrote:
  Matthew Toseland wrote:
  Freenet 0.7 build 1194 is now available. Please upgrade, this will be
  mandatory on Friday. Some fairly substantial changes:
 
  It looks like the infamous # peers forcibly disconnected bug is gone
  in this release, thanks and congratulations! (I see far less package
  handling related messages in the logfiles.)
 
  While I'm at it: Merry Christmas and a Happy New Year!
 
  Peter.
  ___
  Support mailing list
  Support@freenetproject.org
  http://news.gmane.org/gmane.network.freenet.support
  Unsubscribe at 
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
  Or mailto:support-requ...@freenetproject.org?subject=unsubscribe
 
 
 
 
 -- 
 __
 GnuPG key:   (0x48DBFA8A)
 Keyserver:   pgpkeys.pca.dfn.de
 Fingerprint:
 477D F057 1BD4 1AE7 8A54 8679 6690 E2EC 48DB FA8A
 __
 ___
 Support mailing list
 Support@freenetproject.org
 http://news.gmane.org/gmane.network.freenet.support
 Unsubscribe at 
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
 Or mailto:support-requ...@freenetproject.org?subject=unsubscribe
 
 


pgpaLvGjDC1X9.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1194

[3BUIb3S50i 3BUIb3S50i]

This problem is resolved for me. Thanks!


On Wed, Dec 24, 2008 at 7:34 PM,  wrote:

> I still get this:
>
>* Probably a bug: please report: 2 peers forcibly disconnected due
> to not acknowledging packets.
>* Probably a bug: please report: 1 peers forcibly disconnected due
> to not acknowledging packets.
>
> with
>
>* Freenet 0.7 Build #1194 r24387
>* Freenet-ext Build #26 r23771
>
> after this
>
>   nodeUptime: 1d21h
>
>
> Merry Christmas
>
>
> On Wed, Dec 24, 2008 at 17:13, SmallSister development
>  wrote:
> > Matthew Toseland wrote:
> >> Freenet 0.7 build 1194 is now available. Please upgrade, this will be
> >> mandatory on Friday. Some fairly substantial changes:
> >
> > It looks like the infamous "# peers forcibly disconnected" bug is gone
> > in this release, thanks and congratulations! (I see far less package
> > handling related messages in the logfiles.)
> >
> > While I'm at it: Merry Christmas and a Happy New Year!
> >
> > Peter.
> > ___
> > Support mailing list
> > Support at freenetproject.org
> > http://news.gmane.org/gmane.network.freenet.support
> > Unsubscribe at
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
> > Or mailto:support-request at freenetproject.org?subject=unsubscribe
> >
>
>
>
> --
> __
> GnuPG key:   (0x48DBFA8A)
> Keyserver:   pgpkeys.pca.dfn.de
> Fingerprint:
> 477D F057 1BD4 1AE7 8A54 8679 6690 E2EC 48DB FA8A
> __
> ___
> Support mailing list
> Support at freenetproject.org
> http://news.gmane.org/gmane.network.freenet.support
> Unsubscribe at
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
> Or mailto:support-request at freenetproject.org?subject=unsubscribe
>
-- next part --
An HTML attachment was scrubbed...
URL: 

[freenet-support] Freenet 0.7 build 1194

[SmallSister development]

Matthew Toseland wrote:
> Freenet 0.7 build 1194 is now available. Please upgrade, this will be 
> mandatory on Friday. Some fairly substantial changes:

It looks like the infamous "# peers forcibly disconnected" bug is gone
in this release, thanks and congratulations! (I see far less package
handling related messages in the logfiles.)

While I'm at it: Merry Christmas and a Happy New Year!

Peter.

Re: [freenet-support] Freenet 0.7 build 1194

[SmallSister development]

Matthew Toseland wrote:
 Freenet 0.7 build 1194 is now available. Please upgrade, this will be 
 mandatory on Friday. Some fairly substantial changes:

It looks like the infamous # peers forcibly disconnected bug is gone
in this release, thanks and congratulations! (I see far less package
handling related messages in the logfiles.)

While I'm at it: Merry Christmas and a Happy New Year!

Peter.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

Re: [freenet-support] Freenet 0.7 build 1194

[bbackde]

I still get this:

* Probably a bug: please report: 2 peers forcibly disconnected due
to not acknowledging packets.
* Probably a bug: please report: 1 peers forcibly disconnected due
to not acknowledging packets.

with

* Freenet 0.7 Build #1194 r24387
* Freenet-ext Build #26 r23771

after this

   nodeUptime: 1d21h


Merry Christmas


On Wed, Dec 24, 2008 at 17:13, SmallSister development
smallsis...@xs4all.nl wrote:
 Matthew Toseland wrote:
 Freenet 0.7 build 1194 is now available. Please upgrade, this will be
 mandatory on Friday. Some fairly substantial changes:

 It looks like the infamous # peers forcibly disconnected bug is gone
 in this release, thanks and congratulations! (I see far less package
 handling related messages in the logfiles.)

 While I'm at it: Merry Christmas and a Happy New Year!

 Peter.
 ___
 Support mailing list
 Support@freenetproject.org
 http://news.gmane.org/gmane.network.freenet.support
 Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
 Or mailto:support-requ...@freenetproject.org?subject=unsubscribe




-- 
__
GnuPG key:   (0x48DBFA8A)
Keyserver:   pgpkeys.pca.dfn.de
Fingerprint:
477D F057 1BD4 1AE7 8A54 8679 6690 E2EC 48DB FA8A
__
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1194

[Matthew Toseland]

Freenet 0.7 build 1194 is now available. Please upgrade, this will be 
mandatory on Friday. Some fairly substantial changes:
- Fix a serious bug causing the packet sending loop to be very unfair to nodes 
late on in the peers table.
- When reconnecting or rekeying, if the other side decides not to reuse the 
PacketTracker (sequence of packets, tracks which packets we have acked etc), 
assume it is because they don't have the old one any more. This should fix 
the "Forced disconnect of N peers because failed to acknowledge packets after 
10 minutes" bug once and for all.
- Fix various other packet level bugs.
- When deploying an update, parse the main jar to find out which 
freenet-ext.jar versions it is compatible with. Do not deploy the update 
until the main jar and freenet-ext.jar we are going to deploy are compatible. 
Fetch new versions of freenet-ext.jar if needed.
- Allow too-old opennet nodes to connect for an hour so they can download the 
latest update, and don't count them towards the connection limit.
- Some fixes to bulk transfer. This is the system used to transfer large files 
from one node to another - particularly for Update Over Mandatory, the 
mechanism that lets a node auto-update despite being too old to talk to its 
peers. Many UOM transfers were failing, this and the last change should fix 
that.
- Finnish translation update.
- Minor web interface changes.
- Various changes useful for plugins.

There have also been some big changes by sdiz to the XMLSpider plugin, the 
tool used to create search indexes used by the XMLLibrarian plugin / the 
search box on the Browse Freenet page. The XMLSpider plugin now uses db4o to 
store its current progress, should use much less memory, and should resume 
from where it left when it is restarted. This work is still in progress so it 
may not be fully debugged yet. And esr has contributed some more 
documentation fixes.

Credits:
esr
FloppusMaximus
Smar
toad
xor (aka p0s)
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1194

[Matthew Toseland]

Freenet 0.7 build 1194 is now available. Please upgrade, this will be 
mandatory on Friday. Some fairly substantial changes:
- Fix a serious bug causing the packet sending loop to be very unfair to nodes 
late on in the peers table.
- When reconnecting or rekeying, if the other side decides not to reuse the 
PacketTracker (sequence of packets, tracks which packets we have acked etc), 
assume it is because they don't have the old one any more. This should fix 
the Forced disconnect of N peers because failed to acknowledge packets after 
10 minutes bug once and for all.
- Fix various other packet level bugs.
- When deploying an update, parse the main jar to find out which 
freenet-ext.jar versions it is compatible with. Do not deploy the update 
until the main jar and freenet-ext.jar we are going to deploy are compatible. 
Fetch new versions of freenet-ext.jar if needed.
- Allow too-old opennet nodes to connect for an hour so they can download the 
latest update, and don't count them towards the connection limit.
- Some fixes to bulk transfer. This is the system used to transfer large files 
from one node to another - particularly for Update Over Mandatory, the 
mechanism that lets a node auto-update despite being too old to talk to its 
peers. Many UOM transfers were failing, this and the last change should fix 
that.
- Finnish translation update.
- Minor web interface changes.
- Various changes useful for plugins.

There have also been some big changes by sdiz to the XMLSpider plugin, the 
tool used to create search indexes used by the XMLLibrarian plugin / the 
search box on the Browse Freenet page. The XMLSpider plugin now uses db4o to 
store its current progress, should use much less memory, and should resume 
from where it left when it is restarted. This work is still in progress so it 
may not be fully debugged yet. And esr has contributed some more 
documentation fixes.

Credits:
esr
FloppusMaximus
Smar
toad
xor (aka p0s)


pgpKaiu0nC3w6.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1193

[Matthew Toseland]

Freenet 0.7 build 1193 is now available. Please upgrade. Changes:
- Significant changes to the old-opennet-peers mechanism, should make 
reconnecting after downtime much quicker, nodes should get back to where they 
were on the network more quickly even after long downtimes. This will only 
really be testable after Tuesday when 1193 is mandatory. We now keep 25 
old-opennet-peers, which had been useful to us (i.e. weren't dropped from the 
bottom of the LRU by either us or the other peer, just disconnected when the 
node went down etc). We constantly try to reconnect to them for up to a 
month.
- Various other opennet tweaks and bugfixes, some may be significant: we were 
making far too few connection offers on nodes receiving a lot of 
un-decodeable or old-opennet-peer packets, and also fixes to going over the 
20 peers limit.
- Some connectivity fixes, for nodes with multiple IP addresses (e.g. if 
connecting to local/LAN IPs is enabled), and sending packets when we had 
decided not to send packets.
- Public gateway mode: If enabled, users who are allowed to access fproxy but 
don't have full access to the node cannot access the global queue and cannot 
fetch files over 2MB.
- Minor fixes to fproxy.

Credits:
ian
esr
sdiz
toad

Meanwhile, sdiz has been doing a lot of work on XMLSpider, which now uses db4o 
to reduce memory usage, although it is not yet able to resume across 
restarts. And p0s has the Freetalk plugin actually loadable and testable, 
with one message successfully posted; see the wot list for more info. Ian and 
esr have done some work on the website.
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1193

[Matthew Toseland]

Freenet 0.7 build 1193 is now available. Please upgrade. Changes:
- Significant changes to the old-opennet-peers mechanism, should make 
reconnecting after downtime much quicker, nodes should get back to where they 
were on the network more quickly even after long downtimes. This will only 
really be testable after Tuesday when 1193 is mandatory. We now keep 25 
old-opennet-peers, which had been useful to us (i.e. weren't dropped from the 
bottom of the LRU by either us or the other peer, just disconnected when the 
node went down etc). We constantly try to reconnect to them for up to a 
month.
- Various other opennet tweaks and bugfixes, some may be significant: we were 
making far too few connection offers on nodes receiving a lot of 
un-decodeable or old-opennet-peer packets, and also fixes to going over the 
20 peers limit.
- Some connectivity fixes, for nodes with multiple IP addresses (e.g. if 
connecting to local/LAN IPs is enabled), and sending packets when we had 
decided not to send packets.
- Public gateway mode: If enabled, users who are allowed to access fproxy but 
don't have full access to the node cannot access the global queue and cannot 
fetch files over 2MB.
- Minor fixes to fproxy.

Credits:
ian
esr
sdiz
toad

Meanwhile, sdiz has been doing a lot of work on XMLSpider, which now uses db4o 
to reduce memory usage, although it is not yet able to resume across 
restarts. And p0s has the Freetalk plugin actually loadable and testable, 
with one message successfully posted; see the wot list for more info. Ian and 
esr have done some work on the website.


pgponRzMG4ZP6.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

[freenet-support] Freenet 0.7 build 1190, 1191 and 1192

[Matthew Toseland]

Sorry for the belated changelogs ... Please upgrade!

1190:
- 1189 broke seednodes. Fixed in 1190.
1191:
- Fix startup failures due to problems with updater temp files. Delete the old 
temp files.
1192:
- Fix a major unfairness bug in the packet sending scheduler code. This may 
have caused some nodes to be sent far more traffic than others, likely 
causing various nasty connectivity bugs possibly including the forced 
disconnect due to not acking packets bug.

1192 is mandatory on Friday. After that it will be interesting to know whether 
the forced disconnect due to not acknowledging packets bug is still around - 
it probably will be, but at least we're down one possible cause... :|

If you find any bugs, please let us know, preferably via the bug tracker: 
https://bugs.freenetproject.org/

Thanks for using Freenet, and apologies for the problems we've had lately. 
Note that we have additional changes in trunk which have not been released 
yet; all the above are single-bugfix builds. We always need more trunk 
testers (update.sh/update.cmd testing), but you need to keep up to date; 
joining the IRC channel is helpful but not essential.


pgpmIsnoHacGO.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[EMAIL PROTECTED]

[freenet-support] Freenet 0.7 build 1189

[Bill Shutt]

Many, many thanks for all the hard work you guys put into this project. i only 
wish i had skills to help you, sadly i do not. i will certainly contribute as 
and when i can affpord it, but that will have to wait until January with the 
holiday season looming

kindest regards, and grateful thanks

bill

> From: toad at amphibian.dyndns.org
> To: devl at freenetproject.org; support at freenetproject.org
> Date: Sat, 6 Dec 2008 14:49:14 +
> Subject: [freenet-support] Freenet 0.7 build 1189
> 
> Freenet 0.7 build 1189 is now available. Please upgrade, it will be mandatory 
> on Tuesday. Changelog:
> - Another new version of the connection setup protocol which hopefully will 
> be 
> the last for some time, which should fix the current disconnections problems 
> by negotiating whether to reuse the retransmission data and thus avoiding one 
> side choosing to reuse it and the other side not reusing it, resulting in a 
> disconnection.
> - Fix a bug causing us to fail to connect to nodes with DNS addresses (e.g. 
> darknet peers on dyndns) in some cases.
> - Various other improvements to the connection setup layer, especially error 
> handling.
> - Fix a serious bug in opennet, which was causing us to treat handling a 
> packet from an old opennet peer (or which was not identifiable at all) as 
> sending a connection offer, resulting in way too few connection offers being 
> sent.
> - Fix some bugs relating to ARKs.
> - Refactoring, clarify a string.
> 
> Also, XMLSpider and XMLLibrarian (the searching code) have been updated to 
> deal better with non-english languages (as long as they have spaces between 
> words i.e. not chinese yet), and a minor fix to KeyExplorer.
> 
> Thanks to:
> p0s
> sdiz
> toad
> 
> Please upgrade, and report any bugs you find on the bug tracker:
> https://bugs.freenetproject.org/
> 
> With regards to disconnections, please wait until Tuesday, then restart your 
> node. If you get any disconnection notices on the web interface after that, 
> please report them, it means we still haven't solved the problem.
> 
> Thanks for using Freenet! Sorry for all the problems lately, we are working 
> to 
> stabilise it...

_
Get Windows Live Messenger on your Mobile
http://clk.atdmt.com/UKM/go/msnnkmgl001001ukm/direct/01/
-- next part --
An HTML attachment was scrubbed...
URL: 
<https://emu.freenetproject.org/pipermail/support/attachments/20081206/c8603ffa/attachment.html>

[freenet-support] Freenet 0.7 build 1189

[Matthew Toseland]

Freenet 0.7 build 1189 is now available. Please upgrade, it will be mandatory 
on Tuesday. Changelog:
- Another new version of the connection setup protocol which hopefully will be 
the last for some time, which should fix the current disconnections problems 
by negotiating whether to reuse the retransmission data and thus avoiding one 
side choosing to reuse it and the other side not reusing it, resulting in a 
disconnection.
- Fix a bug causing us to fail to connect to nodes with DNS addresses (e.g. 
darknet peers on dyndns) in some cases.
- Various other improvements to the connection setup layer, especially error 
handling.
- Fix a serious bug in opennet, which was causing us to treat handling a 
packet from an old opennet peer (or which was not identifiable at all) as 
sending a connection offer, resulting in way too few connection offers being 
sent.
- Fix some bugs relating to ARKs.
- Refactoring, clarify a string.

Also, XMLSpider and XMLLibrarian (the searching code) have been updated to 
deal better with non-english languages (as long as they have spaces between 
words i.e. not chinese yet), and a minor fix to KeyExplorer.

Thanks to:
p0s
sdiz
toad

Please upgrade, and report any bugs you find on the bug tracker:
https://bugs.freenetproject.org/

With regards to disconnections, please wait until Tuesday, then restart your 
node. If you get any disconnection notices on the web interface after that, 
please report them, it means we still haven't solved the problem.

Thanks for using Freenet! Sorry for all the problems lately, we are working to 
stabilise it...
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1189

[Matthew Toseland]

Freenet 0.7 build 1189 is now available. Please upgrade, it will be mandatory 
on Tuesday. Changelog:
- Another new version of the connection setup protocol which hopefully will be 
the last for some time, which should fix the current disconnections problems 
by negotiating whether to reuse the retransmission data and thus avoiding one 
side choosing to reuse it and the other side not reusing it, resulting in a 
disconnection.
- Fix a bug causing us to fail to connect to nodes with DNS addresses (e.g. 
darknet peers on dyndns) in some cases.
- Various other improvements to the connection setup layer, especially error 
handling.
- Fix a serious bug in opennet, which was causing us to treat handling a 
packet from an old opennet peer (or which was not identifiable at all) as 
sending a connection offer, resulting in way too few connection offers being 
sent.
- Fix some bugs relating to ARKs.
- Refactoring, clarify a string.

Also, XMLSpider and XMLLibrarian (the searching code) have been updated to 
deal better with non-english languages (as long as they have spaces between 
words i.e. not chinese yet), and a minor fix to KeyExplorer.

Thanks to:
p0s
sdiz
toad

Please upgrade, and report any bugs you find on the bug tracker:
https://bugs.freenetproject.org/

With regards to disconnections, please wait until Tuesday, then restart your 
node. If you get any disconnection notices on the web interface after that, 
please report them, it means we still haven't solved the problem.

Thanks for using Freenet! Sorry for all the problems lately, we are working to 
stabilise it...


pgpYq8s6ozSwY.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[EMAIL PROTECTED]

Re: [freenet-support] Freenet 0.7 build 1189

[Bill Shutt]

Many, many thanks for all the hard work you guys put into this project. i only 
wish i had skills to help you, sadly i do not. i will certainly contribute as 
and when i can affpord it, but that will have to wait until January with the 
holiday season looming

kindest regards, and grateful thanks

bill

 From: [EMAIL PROTECTED]
 To: [EMAIL PROTECTED]; support@freenetproject.org
 Date: Sat, 6 Dec 2008 14:49:14 +
 Subject: [freenet-support] Freenet 0.7 build 1189
 
 Freenet 0.7 build 1189 is now available. Please upgrade, it will be mandatory 
 on Tuesday. Changelog:
 - Another new version of the connection setup protocol which hopefully will 
 be 
 the last for some time, which should fix the current disconnections problems 
 by negotiating whether to reuse the retransmission data and thus avoiding one 
 side choosing to reuse it and the other side not reusing it, resulting in a 
 disconnection.
 - Fix a bug causing us to fail to connect to nodes with DNS addresses (e.g. 
 darknet peers on dyndns) in some cases.
 - Various other improvements to the connection setup layer, especially error 
 handling.
 - Fix a serious bug in opennet, which was causing us to treat handling a 
 packet from an old opennet peer (or which was not identifiable at all) as 
 sending a connection offer, resulting in way too few connection offers being 
 sent.
 - Fix some bugs relating to ARKs.
 - Refactoring, clarify a string.
 
 Also, XMLSpider and XMLLibrarian (the searching code) have been updated to 
 deal better with non-english languages (as long as they have spaces between 
 words i.e. not chinese yet), and a minor fix to KeyExplorer.
 
 Thanks to:
 p0s
 sdiz
 toad
 
 Please upgrade, and report any bugs you find on the bug tracker:
 https://bugs.freenetproject.org/
 
 With regards to disconnections, please wait until Tuesday, then restart your 
 node. If you get any disconnection notices on the web interface after that, 
 please report them, it means we still haven't solved the problem.
 
 Thanks for using Freenet! Sorry for all the problems lately, we are working 
 to 
 stabilise it...

_
Get Windows Live Messenger on your Mobile
http://clk.atdmt.com/UKM/go/msnnkmgl001001ukm/direct/01/___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[EMAIL PROTECTED]

[freenet-support] Freenet 0.7 build 1188

[Matthew Toseland]

Freenet 0.7 build 1188 is now available, and will be mandatory on Saturday. 
Please upgrade. The main change in this build is a change to the connection 
setup code: Every hour or so we change the connection encryption key for 
various reasons. Before 1188, we had tied this key to the retransmission and 
packet acknowledgement logic, after 1188 we can change the key without making 
any difference to packet acknowledgements. This should largely solve the "N 
peers forcibly disconnected due to not acknowledging packets" bug, once it is 
mandatory. Please restart your node after midnight GMT on Saturday, and if 
you still get this bug, let us know. Apart from that there are some bugfixes 
and tweaks to the salted hash datastore, IP detection, and a Finnish 
translation update. If you find any bugs please report them to the bug 
tracker at https://bugs.freenetproject.org/ or the support list. Thanks.
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1188

[Matthew Toseland]

Freenet 0.7 build 1188 is now available, and will be mandatory on Saturday. 
Please upgrade. The main change in this build is a change to the connection 
setup code: Every hour or so we change the connection encryption key for 
various reasons. Before 1188, we had tied this key to the retransmission and 
packet acknowledgement logic, after 1188 we can change the key without making 
any difference to packet acknowledgements. This should largely solve the N 
peers forcibly disconnected due to not acknowledging packets bug, once it is 
mandatory. Please restart your node after midnight GMT on Saturday, and if 
you still get this bug, let us know. Apart from that there are some bugfixes 
and tweaks to the salted hash datastore, IP detection, and a Finnish 
translation update. If you find any bugs please report them to the bug 
tracker at https://bugs.freenetproject.org/ or the support list. Thanks.


pgpxiRpYSqcYY.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[EMAIL PROTECTED]

[freenet-support] Freenet 0.7 build 1187

[Matthew Toseland]

Freenet 0.7 build 1187 is now available. Please upgrade. Changes:
- Maybe fix a bug causing IP addresses to be lost for nodes, especially on 
darknet.
- When the Bloom filter size is set to -1 (the default, meaning autodetect), 
keep it as -1, so when we change the store size the bloom filter size will 
also be updated (although we'll need a restart and a bloom reconstruct for 
this to complete at the moment). New nodes were previously setting the bloom 
filter size to 50K (in 1186), or 512K (in 1185), and never changing it when 
setting up much larger stores than the initial 100M/1G in the first-time 
config wizard.
- Allow MIME types on redirects in ClientPutComplexDir (useful for some client 
apps).
- Italian translation update from Luke771, and a slightly better description 
for the always-allowed-dirs.
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1187

[Matthew Toseland]

Freenet 0.7 build 1187 is now available. Please upgrade. Changes:
- Maybe fix a bug causing IP addresses to be lost for nodes, especially on 
darknet.
- When the Bloom filter size is set to -1 (the default, meaning autodetect), 
keep it as -1, so when we change the store size the bloom filter size will 
also be updated (although we'll need a restart and a bloom reconstruct for 
this to complete at the moment). New nodes were previously setting the bloom 
filter size to 50K (in 1186), or 512K (in 1185), and never changing it when 
setting up much larger stores than the initial 100M/1G in the first-time 
config wizard.
- Allow MIME types on redirects in ClientPutComplexDir (useful for some client 
apps).
- Italian translation update from Luke771, and a slightly better description 
for the always-allowed-dirs.


pgpOzBZUWI0Qv.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[EMAIL PROTECTED]

[freenet-support] Freenet 0.7 build 1186

[Matthew Toseland]

On Monday 01 December 2008 07:11, Ronald Wiplinger (Lists) wrote:
> How to upgrade?
> 
> Automatic upgrade did not work.
> Now it does not even start anymore.

Please send me your wrapper.log. This will be in the directory you installed 
Freenet to. Thanks.
> 
> bye
> 
> Ronald
> 
> 
> Matthew Toseland wrote:
> > Freenet 0.7 build 1186 is now available. Please upgrade, it will be 
mandatory 
> > on Tuesday. Changes:
> > - Fix a nasty race condition in the message handling code. Impact is 
uncertain 
> > but could be large - it broke an update-over-mandatory between nodes on 
the 
> > same computer in testing.
> > - Implement (long overdue) update over mandatory for freenet-ext.jar. 
Normally 
> > this will fetch both the main jar and the ext jar at the same time, if the 
> > node is too old to update over Freenet, or if it takes more than 3 hours 
to 
> > update. 
> > - Include the earliest compatible version of freenet-ext.jar in the 
manifest 
> > for the main jar. We will use this to ensure compatibility when updating 
in a 
> > near future build.
> > - Limit the size of the auto-detected datastore size to 256GB. This will 
> > ensure the Bloom filter is of a reasonable size (128MB at most), and avoid 
> > some crazy situations. Don't limit the Bloom filter size to 256MB any 
more, 
> > defaults to 1/2048th of the size of the store up to 2GB filter.
> > - Make the datastore size default to 100MB. The first-time wizard will 
> > configure it to a larger size, but the resize will occur online, and this 
> > gains us a few seconds on the first startup by not having to wait so long 
for 
> > preallocating it.
> > - Minor change to plugin loading and a minor bugfix to connection setup.
> >
> > Please upgrade! If you find any bugs, please report them on the bug 
tracker. 
> > The "not acknowledging packets even after 10 minutes" bug is sadly still 
with 
> > us, it will be fixed soon.
> >   
> > 
> >
> > ___
> > Support mailing list
> > Support at freenetproject.org
> > http://news.gmane.org/gmane.network.freenet.support
> > Unsubscribe at 
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
> > Or mailto:support-request at freenetproject.org?subject=unsubscribe
> 
> ___
> Support mailing list
> Support at freenetproject.org
> http://news.gmane.org/gmane.network.freenet.support
> Unsubscribe at 
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
> Or mailto:support-request at freenetproject.org?subject=unsubscribe
> 
> 
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1186

[Luke771]

On Mon, 01 Dec 2008 15:11:07 +0800
"Ronald Wiplinger (Lists)"  wrote:

> How to upgrade?

On Windows:
Click My Computer, Browse to Freenet folder, double click on update.cmd
-or-
Start => (type) cmd [enter] => (type) cd c:\program files\freenet (or wherever 
you installed Freenet)[enter] => (type) update.cmd [enter]

NOTE: If you get 'too many parameters' error, type progra~1 instead of program 
files (cd c:\progra~1\freenet)



On any other OS (including Mac, Ubuntu)
- Open a terminal, cd to the Freenet directory, type ./update.sh ,  hit enter

NOTE: you may need to set execute permission for file update.sh with the 
command chmod +x update.sh


> Automatic upgrade did not work.
> Now it does not even start anymore.

It would be nice to know what happens exactly
Any ERROR lines in wrapper.conf?

[freenet-support] Freenet 0.7 build 1186

[Ronald Wiplinger (Lists)]

How to upgrade?

Automatic upgrade did not work.
Now it does not even start anymore.

bye

Ronald


Matthew Toseland wrote:
> Freenet 0.7 build 1186 is now available. Please upgrade, it will be mandatory 
> on Tuesday. Changes:
> - Fix a nasty race condition in the message handling code. Impact is 
> uncertain 
> but could be large - it broke an update-over-mandatory between nodes on the 
> same computer in testing.
> - Implement (long overdue) update over mandatory for freenet-ext.jar. 
> Normally 
> this will fetch both the main jar and the ext jar at the same time, if the 
> node is too old to update over Freenet, or if it takes more than 3 hours to 
> update. 
> - Include the earliest compatible version of freenet-ext.jar in the manifest 
> for the main jar. We will use this to ensure compatibility when updating in a 
> near future build.
> - Limit the size of the auto-detected datastore size to 256GB. This will 
> ensure the Bloom filter is of a reasonable size (128MB at most), and avoid 
> some crazy situations. Don't limit the Bloom filter size to 256MB any more, 
> defaults to 1/2048th of the size of the store up to 2GB filter.
> - Make the datastore size default to 100MB. The first-time wizard will 
> configure it to a larger size, but the resize will occur online, and this 
> gains us a few seconds on the first startup by not having to wait so long for 
> preallocating it.
> - Minor change to plugin loading and a minor bugfix to connection setup.
>
> Please upgrade! If you find any bugs, please report them on the bug tracker. 
> The "not acknowledging packets even after 10 minutes" bug is sadly still with 
> us, it will be fixed soon.
>   
> 
>
> ___
> Support mailing list
> Support at freenetproject.org
> http://news.gmane.org/gmane.network.freenet.support
> Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
> Or mailto:support-request at freenetproject.org?subject=unsubscribe

Re: [freenet-support] Freenet 0.7 build 1186

[Luke771]

On Mon, 01 Dec 2008 15:11:07 +0800
Ronald Wiplinger (Lists) [EMAIL PROTECTED] wrote:

 How to upgrade?

On Windows:
Click My Computer, Browse to Freenet folder, double click on update.cmd
-or-
Start = (type) cmd [enter] = (type) cd c:\program files\freenet (or wherever 
you installed Freenet)[enter] = (type) update.cmd [enter]

NOTE: If you get 'too many parameters' error, type progra~1 instead of program 
files (cd c:\progra~1\freenet)



On any other OS (including Mac, Ubuntu)
- Open a terminal, cd to the Freenet directory, type ./update.sh ,  hit enter

NOTE: you may need to set execute permission for file update.sh with the 
command chmod +x update.sh

 
 Automatic upgrade did not work.
 Now it does not even start anymore.

It would be nice to know what happens exactly
Any ERROR lines in wrapper.conf?
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[EMAIL PROTECTED]

Re: [freenet-support] Freenet 0.7 build 1186

[Matthew Toseland]

On Monday 01 December 2008 07:11, Ronald Wiplinger (Lists) wrote:
 How to upgrade?
 
 Automatic upgrade did not work.
 Now it does not even start anymore.

Please send me your wrapper.log. This will be in the directory you installed 
Freenet to. Thanks.
 
 bye
 
 Ronald
 
 
 Matthew Toseland wrote:
  Freenet 0.7 build 1186 is now available. Please upgrade, it will be 
mandatory 
  on Tuesday. Changes:
  - Fix a nasty race condition in the message handling code. Impact is 
uncertain 
  but could be large - it broke an update-over-mandatory between nodes on 
the 
  same computer in testing.
  - Implement (long overdue) update over mandatory for freenet-ext.jar. 
Normally 
  this will fetch both the main jar and the ext jar at the same time, if the 
  node is too old to update over Freenet, or if it takes more than 3 hours 
to 
  update. 
  - Include the earliest compatible version of freenet-ext.jar in the 
manifest 
  for the main jar. We will use this to ensure compatibility when updating 
in a 
  near future build.
  - Limit the size of the auto-detected datastore size to 256GB. This will 
  ensure the Bloom filter is of a reasonable size (128MB at most), and avoid 
  some crazy situations. Don't limit the Bloom filter size to 256MB any 
more, 
  defaults to 1/2048th of the size of the store up to 2GB filter.
  - Make the datastore size default to 100MB. The first-time wizard will 
  configure it to a larger size, but the resize will occur online, and this 
  gains us a few seconds on the first startup by not having to wait so long 
for 
  preallocating it.
  - Minor change to plugin loading and a minor bugfix to connection setup.
 
  Please upgrade! If you find any bugs, please report them on the bug 
tracker. 
  The not acknowledging packets even after 10 minutes bug is sadly still 
with 
  us, it will be fixed soon.

  
 
  ___
  Support mailing list
  Support@freenetproject.org
  http://news.gmane.org/gmane.network.freenet.support
  Unsubscribe at 
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
  Or mailto:[EMAIL PROTECTED]
 
 ___
 Support mailing list
 Support@freenetproject.org
 http://news.gmane.org/gmane.network.freenet.support
 Unsubscribe at 
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
 Or mailto:[EMAIL PROTECTED]
 
 


pgp65A1IoeBK0.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[EMAIL PROTECTED]

Re: [freenet-support] Freenet 0.7 build 1186

[Ronald Wiplinger (Lists)]

How to upgrade?

Automatic upgrade did not work.
Now it does not even start anymore.

bye

Ronald


Matthew Toseland wrote:
 Freenet 0.7 build 1186 is now available. Please upgrade, it will be mandatory 
 on Tuesday. Changes:
 - Fix a nasty race condition in the message handling code. Impact is 
 uncertain 
 but could be large - it broke an update-over-mandatory between nodes on the 
 same computer in testing.
 - Implement (long overdue) update over mandatory for freenet-ext.jar. 
 Normally 
 this will fetch both the main jar and the ext jar at the same time, if the 
 node is too old to update over Freenet, or if it takes more than 3 hours to 
 update. 
 - Include the earliest compatible version of freenet-ext.jar in the manifest 
 for the main jar. We will use this to ensure compatibility when updating in a 
 near future build.
 - Limit the size of the auto-detected datastore size to 256GB. This will 
 ensure the Bloom filter is of a reasonable size (128MB at most), and avoid 
 some crazy situations. Don't limit the Bloom filter size to 256MB any more, 
 defaults to 1/2048th of the size of the store up to 2GB filter.
 - Make the datastore size default to 100MB. The first-time wizard will 
 configure it to a larger size, but the resize will occur online, and this 
 gains us a few seconds on the first startup by not having to wait so long for 
 preallocating it.
 - Minor change to plugin loading and a minor bugfix to connection setup.

 Please upgrade! If you find any bugs, please report them on the bug tracker. 
 The not acknowledging packets even after 10 minutes bug is sadly still with 
 us, it will be fixed soon.
   
 

 ___
 Support mailing list
 Support@freenetproject.org
 http://news.gmane.org/gmane.network.freenet.support
 Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
 Or mailto:[EMAIL PROTECTED]

___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[EMAIL PROTECTED]

[freenet-support] Freenet 0.7 build 1186

[Matthew Toseland]

Freenet 0.7 build 1186 is now available. Please upgrade, it will be mandatory 
on Tuesday. Changes:
- Fix a nasty race condition in the message handling code. Impact is uncertain 
but could be large - it broke an update-over-mandatory between nodes on the 
same computer in testing.
- Implement (long overdue) update over mandatory for freenet-ext.jar. Normally 
this will fetch both the main jar and the ext jar at the same time, if the 
node is too old to update over Freenet, or if it takes more than 3 hours to 
update. 
- Include the earliest compatible version of freenet-ext.jar in the manifest 
for the main jar. We will use this to ensure compatibility when updating in a 
near future build.
- Limit the size of the auto-detected datastore size to 256GB. This will 
ensure the Bloom filter is of a reasonable size (128MB at most), and avoid 
some crazy situations. Don't limit the Bloom filter size to 256MB any more, 
defaults to 1/2048th of the size of the store up to 2GB filter.
- Make the datastore size default to 100MB. The first-time wizard will 
configure it to a larger size, but the resize will occur online, and this 
gains us a few seconds on the first startup by not having to wait so long for 
preallocating it.
- Minor change to plugin loading and a minor bugfix to connection setup.

Please upgrade! If you find any bugs, please report them on the bug tracker. 
The "not acknowledging packets even after 10 minutes" bug is sadly still with 
us, it will be fixed soon.
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1186

[Matthew Toseland]

Freenet 0.7 build 1186 is now available. Please upgrade, it will be mandatory 
on Tuesday. Changes:
- Fix a nasty race condition in the message handling code. Impact is uncertain 
but could be large - it broke an update-over-mandatory between nodes on the 
same computer in testing.
- Implement (long overdue) update over mandatory for freenet-ext.jar. Normally 
this will fetch both the main jar and the ext jar at the same time, if the 
node is too old to update over Freenet, or if it takes more than 3 hours to 
update. 
- Include the earliest compatible version of freenet-ext.jar in the manifest 
for the main jar. We will use this to ensure compatibility when updating in a 
near future build.
- Limit the size of the auto-detected datastore size to 256GB. This will 
ensure the Bloom filter is of a reasonable size (128MB at most), and avoid 
some crazy situations. Don't limit the Bloom filter size to 256MB any more, 
defaults to 1/2048th of the size of the store up to 2GB filter.
- Make the datastore size default to 100MB. The first-time wizard will 
configure it to a larger size, but the resize will occur online, and this 
gains us a few seconds on the first startup by not having to wait so long for 
preallocating it.
- Minor change to plugin loading and a minor bugfix to connection setup.

Please upgrade! If you find any bugs, please report them on the bug tracker. 
The not acknowledging packets even after 10 minutes bug is sadly still with 
us, it will be fixed soon.


pgplBydJowJd5.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[EMAIL PROTECTED]

[freenet-support] Freenet 0.7 build 1187

[Matthew Toseland]

Freenet 0.7 build 1187 is now available. Mostly minor changes but one was 
important for newly installed nodes:
- Maybe fix darknet peers losing their IP addresses (bug 2688). We were losing 
IPs on connect and on rekeying (every hour!).
- Keep a -1 setting for the Bloom filter size as -1. New nodes start off with 
a 100MB store (prior to 1186 with a 1GB store). The old behaviour would set 
the Bloom filter size once to 1/2048th of the store size: 50K, or 500K for 
pre-1186. Then the post-install wizard would change the store size, but *not* 
the Bloom filter size, resulting in far more false positives than needed, so 
far more disk access than needed.
- Allow MIME types on redirects in ClientPutComplexDir (useful for some client 
apps).
- Italian update from Luke771, and a couple of explanations improved.


pgp19VihLD2Rh.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[EMAIL PROTECTED]

[freenet-support] Freenet 0.7 build 1182-1185

[Dennis Nezic]

On Thu, 27 Nov 2008 20:55:50 +, Matthew Toseland
 wrote:

> Changes in 1185:
> - Maybe fix the bandwidth collapsing bug. When a connection stalled,
> this bug caused us to lose messages, resulting in finished requests
> not being removed, resulting in the node thinking it is more
> overloaded than it is and rejecting requests, node bandwidth usage
> collapses. (Bug 2720, thanks dennisn, petenix and others).

YES!!!111! Success. Merci much!

Re: [freenet-support] Freenet 0.7 build 1182-1185

[Dennis Nezic]

On Thu, 27 Nov 2008 20:55:50 +, Matthew Toseland
[EMAIL PROTECTED] wrote:

 Changes in 1185:
 - Maybe fix the bandwidth collapsing bug. When a connection stalled,
 this bug caused us to lose messages, resulting in finished requests
 not being removed, resulting in the node thinking it is more
 overloaded than it is and rejecting requests, node bandwidth usage
 collapses. (Bug 2720, thanks dennisn, petenix and others).

YES!!!111! Success. Merci much!
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[EMAIL PROTECTED]

[freenet-support] Freenet 0.7 build 1182-1185

[Matthew Toseland]

Freenet 0.7 build 1185 is now available. Please upgrade. This build will be 
mandatory on Monday. If you have any problems updating please let us know, 
include your wrapper.log, and update manually using the update scripts 
(update.sh or update.cmd). If you discover any bugs please report them, 
preferably to the bug tracker: https://bugs.freenetproject.org/

Changes in 1182:
- Greatly improve the data corruption bug workaround introduced in 1181.
- Faster compression.
- Minor bugfix in the PNG filter.
- Some unit tests to try to avoid the data corruption mess in future.

Changes in 1183: (Note if you are still on 1183 and you have automatic 
updating disabled you might have to update manually)
- Fix a space leak related to plugins that do requests (e.g. WoT) hanging in 
unload.
- First part of FMS/Frost message insertion fix.
- Temporary file handling improvements, make sure temp files are cleaned up, 
log an error when they leak.
- Minor optimisations, minor internal changes...

Changes in 1184:
- Fix a temporary file related bug affecting the web interface. Most forms 
were broken, including the shutdown button and dismissing useralerts.
- Fix an internal error caused by problems with temporary buckets
- Fix the bug causing FMS and Frost message inserts to fail (we hope; if you 
still have problems with this, let us know...).
- Lots of leaking temp files fixes.

Changes in 1185:
- Maybe fix the bandwidth collapsing bug. When a connection stalled, this bug 
caused us to lose messages, resulting in finished requests not being removed, 
resulting in the node thinking it is more overloaded than it is and rejecting 
requests, node bandwidth usage collapses. (Bug 2720, thanks dennisn, petenix 
and others).
- Fetch freenet-ext.jar version 26. This is the first part of a permanent 
solution to the data corruption bug.
- Minor fix to the salted hash datastore alert shown when there are still 
files from bdbje left over and the user should delete them.
- Minor NullPointerException fix on loading the homepage too soon after 
startup.
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1182-1185

[Matthew Toseland]

Freenet 0.7 build 1185 is now available. Please upgrade. This build will be 
mandatory on Monday. If you have any problems updating please let us know, 
include your wrapper.log, and update manually using the update scripts 
(update.sh or update.cmd). If you discover any bugs please report them, 
preferably to the bug tracker: https://bugs.freenetproject.org/

Changes in 1182:
- Greatly improve the data corruption bug workaround introduced in 1181.
- Faster compression.
- Minor bugfix in the PNG filter.
- Some unit tests to try to avoid the data corruption mess in future.

Changes in 1183: (Note if you are still on 1183 and you have automatic 
updating disabled you might have to update manually)
- Fix a space leak related to plugins that do requests (e.g. WoT) hanging in 
unload.
- First part of FMS/Frost message insertion fix.
- Temporary file handling improvements, make sure temp files are cleaned up, 
log an error when they leak.
- Minor optimisations, minor internal changes...

Changes in 1184:
- Fix a temporary file related bug affecting the web interface. Most forms 
were broken, including the shutdown button and dismissing useralerts.
- Fix an internal error caused by problems with temporary buckets
- Fix the bug causing FMS and Frost message inserts to fail (we hope; if you 
still have problems with this, let us know...).
- Lots of leaking temp files fixes.

Changes in 1185:
- Maybe fix the bandwidth collapsing bug. When a connection stalled, this bug 
caused us to lose messages, resulting in finished requests not being removed, 
resulting in the node thinking it is more overloaded than it is and rejecting 
requests, node bandwidth usage collapses. (Bug 2720, thanks dennisn, petenix 
and others).
- Fetch freenet-ext.jar version 26. This is the first part of a permanent 
solution to the data corruption bug.
- Minor fix to the salted hash datastore alert shown when there are still 
files from bdbje left over and the user should delete them.
- Minor NullPointerException fix on loading the homepage too soon after 
startup.


pgpDs7jDoUT6N.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[EMAIL PROTECTED]

[freenet-support] Freenet 0.7 build 1181

[Matthew Toseland]

Freenet 0.7 build 1181 is now available. It will be mandatory on Sunday. 
Please upgrade. 

The main change in 1181 is a partial fix (workaround) to the download 
corruption bug. All sites should now load without decompression errors and 
without silent data corruption. However, we implement this by ignoring the 
last data block of each segment, so there may be a small cost to performance 
and/or data reachability. 

The origin of the problem is that we have apparently had 3 different padding 
algorithms for the last block of the splitfile, all of them based on 
different incorrect byte[] to int[] conversions (yes, I know this is 
incredibly dumb). The new freenet-ext.jar (#26) finally implements the 
correct conversion algorithm. Unfortunately, old data will be padded 
incorrectly, in one of three different ways, and there's no easy way to 
identify which. Meaning if we download a splitfile and by chance happen to 
include the last data block in the decode, we can get data corruption. So the 
solution is to ignore the last data block when FEC decoding. 

1182 will provide a more permanent solution, involving a new metadata format 
and enforced post-decode post-decompress checksums. However 1181 should fix 
the problem. PLEASE UPGRADE! If your node has problems upgrading (it might 
well!), please run update.sh / update.cmd manually.

If you find any bugs in 1181, please let us know, preferably via the bug 
tracker: https://bugs.freenetproject.org/
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1181

[Matthew Toseland]

Freenet 0.7 build 1181 is now available. It will be mandatory on Sunday. 
Please upgrade. 

The main change in 1181 is a partial fix (workaround) to the download 
corruption bug. All sites should now load without decompression errors and 
without silent data corruption. However, we implement this by ignoring the 
last data block of each segment, so there may be a small cost to performance 
and/or data reachability. 

The origin of the problem is that we have apparently had 3 different padding 
algorithms for the last block of the splitfile, all of them based on 
different incorrect byte[] to int[] conversions (yes, I know this is 
incredibly dumb). The new freenet-ext.jar (#26) finally implements the 
correct conversion algorithm. Unfortunately, old data will be padded 
incorrectly, in one of three different ways, and there's no easy way to 
identify which. Meaning if we download a splitfile and by chance happen to 
include the last data block in the decode, we can get data corruption. So the 
solution is to ignore the last data block when FEC decoding. 

1182 will provide a more permanent solution, involving a new metadata format 
and enforced post-decode post-decompress checksums. However 1181 should fix 
the problem. PLEASE UPGRADE! If your node has problems upgrading (it might 
well!), please run update.sh / update.cmd manually.

If you find any bugs in 1181, please let us know, preferably via the bug 
tracker: https://bugs.freenetproject.org/


pgp06gLX7t651.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[EMAIL PROTECTED]

[freenet-support] Freenet 0.7 build 1180 and freenet-ext.jar #25

[Matthew Toseland]

Freenet 0.7 build 1180 is now available. Please upgrade. This build, and the 
associated freenet-ext.jar, fix two major bugs (in buckets and block padding) 
which we believe to be related to the recent download/browsing problems (bug 
2694). 1180 is mandatory on Thursday. Note that one or two freesites may 
continue to have problems (temporary bucket errors etc). The problem was that 
we accidentally changed the algorithm for padding the last block in a 
splitfile for FEC decoding. This caused the whole download to fail whenever 
the last block was involved in a decode. We have changed the algorithm back, 
but some recently inserted files and freesites may be broken by the change 
back (rather than many older files and freesites). Please would site authors 
reinsert any recently inserted sites with 1180. We have had problems with 
freenet-ext.jar updates in the past, so please tell us if you have any 
problems updating; include your wrapper.log. Thanks.
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1180 and freenet-ext.jar #25

[Matthew Toseland]

Freenet 0.7 build 1180 is now available. Please upgrade. This build, and the 
associated freenet-ext.jar, fix two major bugs (in buckets and block padding) 
which we believe to be related to the recent download/browsing problems (bug 
2694). 1180 is mandatory on Thursday. Note that one or two freesites may 
continue to have problems (temporary bucket errors etc). The problem was that 
we accidentally changed the algorithm for padding the last block in a 
splitfile for FEC decoding. This caused the whole download to fail whenever 
the last block was involved in a decode. We have changed the algorithm back, 
but some recently inserted files and freesites may be broken by the change 
back (rather than many older files and freesites). Please would site authors 
reinsert any recently inserted sites with 1180. We have had problems with 
freenet-ext.jar updates in the past, so please tell us if you have any 
problems updating; include your wrapper.log. Thanks.


pgpMh0ClabNFa.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[EMAIL PROTECTED]

[freenet-support] Freenet 0.7 build 1178 (and 1176, 1177)

[Dennis Nezic]

On Fri, 14 Nov 2008 18:21:50 -0500, Dennis Nezic
 wrote:

> On Fri, 14 Nov 2008 19:39:45 +, Matthew Toseland
>  wrote:
> 
> > Freenet 0.7 build 1178 is now available.

I spoke too soon. It works fine for an hour or two, but then, at least
for me, all communcation (net input/output) STOPS. (Well, ~100Bytes/sec
both ways--but, for all intents and purposes, it's offline.)

After almost exactly 3 hours, it begins Announcing itself
(successfully, apparently) although nothing really changes.

And this is repeatable. Restarting the node leads to the exact same
behaviour.

There are very few errors in the freenet logs. A couple per hour,
maybe. But, the "common" one that I spotted freenet.node.PeerManager,
UdpSocketHandler ERROR addPeerNodeStatus(): "node already in
peerNodeStatuses" ... "status DISCONNECTED". But, it's ip address is
still listed as CONNECTED, and data is still being transferred with it,
although the port number in the log file is different than the one that
is listed as connected. (I'm not sure if this is related.)

Re: [freenet-support] Freenet 0.7 build 1178 (and 1176, 1177)

[Dennis Nezic]

On Fri, 14 Nov 2008 18:21:50 -0500, Dennis Nezic
[EMAIL PROTECTED] wrote:

 On Fri, 14 Nov 2008 19:39:45 +, Matthew Toseland
 [EMAIL PROTECTED] wrote:
 
  Freenet 0.7 build 1178 is now available.

I spoke too soon. It works fine for an hour or two, but then, at least
for me, all communcation (net input/output) STOPS. (Well, ~100Bytes/sec
both ways--but, for all intents and purposes, it's offline.)

After almost exactly 3 hours, it begins Announcing itself
(successfully, apparently) although nothing really changes.

And this is repeatable. Restarting the node leads to the exact same
behaviour.

There are very few errors in the freenet logs. A couple per hour,
maybe. But, the common one that I spotted freenet.node.PeerManager,
UdpSocketHandler ERROR addPeerNodeStatus(): node already in
peerNodeStatuses ... status DISCONNECTED. But, it's ip address is
still listed as CONNECTED, and data is still being transferred with it,
although the port number in the log file is different than the one that
is listed as connected. (I'm not sure if this is related.)
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[EMAIL PROTECTED]

[freenet-support] Freenet 0.7 build 1178 (and 1176, 1177)

[Matthew Toseland]

On Friday 14 November 2008 23:21, Dennis Nezic wrote:
> On Fri, 14 Nov 2008 19:39:45 +, Matthew Toseland
>  wrote:
> 
> > Freenet 0.7 build 1178 is now available.
> 
> Wow! What a difference. 1178 is like a totally different program, with
> respect to both cpu and memory usage. THANK YOU! It's no longer peaking
> for minutes at a time. It's consistently low. My 15minute load average
> is below 1 now! Hurray. Memory use also seems much lower.
> 
> Regarding the "not acked packets" problem, I have alrady been alerted
> about 5 peers with "severe" problems, one of which is a darknet peer. 2
> of the others are listed as CONNECTED. And the other 2 are listed as
> BACKED OFF. All of them are running 1178.
> 
> Also, I should repeat ;), One of my peers (ip addresses) is CONNECTED
> to me both in opennet and darknet.

That's to be expected. We kill the connection, and purge the trackers etc, but 
we can reconnect afterwards. Also, it's quite possible that those peers were 
running an earlier version when they actually broke. However, it's also 
possible that we still have the underlying bug... I've not made much headway 
with finding it, next week we should know whether it's still active, since 
1178 is mandatory on Sunday...
> 
> All in all, though, 1178 is a HUGE improvement!
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1175

[Matthew Toseland]

On Friday 14 November 2008 17:40, Dennis Nezic wrote:
> On Thu, 13 Nov 2008 18:16:55 +, Matthew Toseland
>  wrote:
> 
> > Freenet 0.7 build 1175 is now available. Please upgrade. Changes:
> > - Handle the "Packet X sent Yms ago and still not acked" bug better.
> > Don't complain in the log until 10 minutes, then disconnect from the
> > peer and log the fact. We can then reconnect, but we will do so
> > cleanly, so hopefully not get it again for a while. It seems that
> > most of the disruption caused by this bug was caused by the logging
> > on every resend and not the bug itself... Of course we will need to
> > fix the bug too.
> 
> State Of The (My) 1175... It "feels" like it's running smoother,
> although it's cpu profile here hasn't changed substantially. Perhaps
> just a bit, due to better logging. The log files are finally usable!
> (Usually under 100K per hour now :)

It still has 100% CPU most of the time then? Apparently resulting from Freenet 
itself rather than downloads etc?

Try 1178 ... and then wait until say Wednesday ... hopefully once everyone has 
1178, it will be much better.
> 
> So, in one random 1 hour sample, here are the main ones that I get:
> 
> 4 "Forcing disconnect" PacketSender ERRORs, both from opennet and
> darknet peers, "because packets not acked after 10 minutes!". Does this
> mean that NO packets have been received from these peers, or just a big
> bunch of them? (I'm pretty sure my ISP doesn't kill connections, but
> just drops a fraction of the packets.) For some reason, these error
> messages come in pairs, first by FNPPacketMangler, then by (Dark|Open)
> netPeerNode.

This is caused by the bugs we're chasing at the moment. Hopefully next week it 
will have gone away, or be very rare. It means that we are still connected to 
the node, it's still sending us packets, but it apparently doesn't understand 
some of the packets we send it.
> 
> 40 MessageCore ERRORs "from RequestSender  waitFor _unclaimed iteration
> took 3428ms with unclaimedFIFOSize of 533 for ret of packetTransmit" or
> "from UdpSocketHandler checkFilters took 6260ms with unclaimedFIFOSize
> of 596 for matched: true"
> 
> 24 freenet.io.xfer.BlockTransmitter.sendAsync(), "Terminated send
> -123413232423 ... as we haven't heard from receiver in 1m54s."
> 
> 4 freenet.io.xferPacketThrottle, Blocktransmitter: "Unable to send
> throttled message, waited 60100ms"
> 
> 6 freenet.io.comm.MessageCore Scheduled job ERRORS
> "removeTimedOutFilters took 4000ms"

Hopefully these are related to the current bugs. If so they should greatly 
reduce. On the other hand, they may not be ...
> 
> 4 freenet.node.KeyTracker PacketSender "Packet 1234 sent over 600600ms
> ago and still not acked". (THEY STILL EXIST!?! ;)

Yes, we log them at the same time as the disconnections.
> 
> 19 freenet.io.comm.UdpSocketHandler ERROR messages (from both darknet
> and opennet), "processing packet took 4000ms" .. with times ranging
> from 3000ms to 6600ms.
> 
> 8 freenet.node.FNPPacketMangler "Message[1..4] timeout error:
> Processing packet for freenetnode.tld:1234"
> 
> Also, in this log, I have noticed that the port number, in the
> @ip:port@ portion of the error message does NOT match the port number
> listed on the /friends/ web page. Why is that? 

It can be different, usually because of NATs, occasionally because you are 
connected both to the darknet and opennet nodes on that IP (which is a bug, 
shouldn't happen unless you have some wierd configs set).

> (What happens, by the 
> way, if a darknet peer does change his port, by the way--we'd have to
> re-add his node reference?)

Probably yes.
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1178 (and 1176, 1177)

[Matthew Toseland]

Freenet 0.7 build 1178 is now available. Please upgrade. Sorry I accidentally 
released 1176, then had some bugfixes in 1177 and 1178. Anyway, the main 
changes relative to 1175:
- Try to prevent both sides of a connection attempting to rekey at the same 
time, by having the receiver start 2.5 minutes early. This seems to have been 
the root cause of many of the recent problems. It may not be completely 
fixed, but should happen much less often.
- When we disconnect from a peer because of it not acknowledging our packets, 
tell the user via a useralert as well as logging an error.
- Get rid of WouldBlockException errors as well: If we cannot send a packet to 
a peer because there are too many packets in flight and none are being 
acknowledged, don't busy-loop and spam the logs, wait for it to become 
un-blocked again. If it doesn't become un-blocked in 10 minutes, log an error 
and tell the user via a useralert.
- Change the default log rotation period to 10 minutes rather than 1 hour to 
save disk space.
- Clarify the config option for packet padding: Turning off packet padding is 
BAD. Packet padding is essential against a passive attacker.

In the next few days you may see the "Probably a bug" errors on the web 
interface. Hopefully once this build is mandatory (on Monday) these will go 
away. If you still see these from Tuesday onwards, let us know.

Please update, and let us know how it works out. Hopefully this build will fix 
most of the recent network problems, oversized log files and high CPU usage 
when not doing anything. If you find any bugs, please report them on the bug 
tracker: https://bugs.freenetproject.org/ . Thanks.
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: not available
URL: 

[freenet-support] Freenet 0.7 build 1178 (and 1176, 1177)

[Dennis Nezic]

On Fri, 14 Nov 2008 19:39:45 +, Matthew Toseland
 wrote:

> Freenet 0.7 build 1178 is now available.

Wow! What a difference. 1178 is like a totally different program, with
respect to both cpu and memory usage. THANK YOU! It's no longer peaking
for minutes at a time. It's consistently low. My 15minute load average
is below 1 now! Hurray. Memory use also seems much lower.

Regarding the "not acked packets" problem, I have alrady been alerted
about 5 peers with "severe" problems, one of which is a darknet peer. 2
of the others are listed as CONNECTED. And the other 2 are listed as
BACKED OFF. All of them are running 1178.

Also, I should repeat ;), One of my peers (ip addresses) is CONNECTED
to me both in opennet and darknet.

All in all, though, 1178 is a HUGE improvement!

[freenet-support] Freenet 0.7 build 1175

[Dennis Nezic]

On Fri, 14 Nov 2008 19:52:46 +, Matthew Toseland
 wrote:

> It still has 100% CPU most of the time then? Apparently resulting
> from Freenet itself rather than downloads etc?

Well, it's not /so/ serious that it stays at 100% all the time. It
jumps up for a few minutes, then calms down to almost nothing for a
few minutes. (The valleys are longer than the peaks.) And keeps
repeating this. I'm running another process that steadily uses on
average 25% of the (1.2GHz) cpu, and my 15minute load time is
consistently around 3. Maybe the threads are all triggered to run at
the same time, and should be staggered? 


> > Also, in this log, I have noticed that the port number, in the
> > @ip:port@ portion of the error message does NOT match the port
> > number listed on the /friends/ web page. Why is that? 
> 
> It can be different, usually because of NATs, occasionally because
> you are connected both to the darknet and opennet nodes on that IP
> (which is a bug, shouldn't happen unless you have some wierd configs
> set).

Actually, the logs do say that I was connected to this same ip address
both in opennet and darknet :|. (And the opennet port is properly
listed in the strangers page. (It seems unlikely that his darknet port
would be routed in a NAT, but not his opennet port :|.)

[freenet-support] Freenet 0.7 build 1175

[Dennis Nezic]

On Thu, 13 Nov 2008 18:16:55 +, Matthew Toseland
 wrote:

> Freenet 0.7 build 1175 is now available. Please upgrade. Changes:
> - Handle the "Packet X sent Yms ago and still not acked" bug better.
> Don't complain in the log until 10 minutes, then disconnect from the
> peer and log the fact. We can then reconnect, but we will do so
> cleanly, so hopefully not get it again for a while. It seems that
> most of the disruption caused by this bug was caused by the logging
> on every resend and not the bug itself... Of course we will need to
> fix the bug too.

State Of The (My) 1175... It "feels" like it's running smoother,
although it's cpu profile here hasn't changed substantially. Perhaps
just a bit, due to better logging. The log files are finally usable!
(Usually under 100K per hour now :)

So, in one random 1 hour sample, here are the main ones that I get:

4 "Forcing disconnect" PacketSender ERRORs, both from opennet and
darknet peers, "because packets not acked after 10 minutes!". Does this
mean that NO packets have been received from these peers, or just a big
bunch of them? (I'm pretty sure my ISP doesn't kill connections, but
just drops a fraction of the packets.) For some reason, these error
messages come in pairs, first by FNPPacketMangler, then by (Dark|Open)
netPeerNode.

40 MessageCore ERRORs "from RequestSender  waitFor _unclaimed iteration
took 3428ms with unclaimedFIFOSize of 533 for ret of packetTransmit" or
"from UdpSocketHandler checkFilters took 6260ms with unclaimedFIFOSize
of 596 for matched: true"

24 freenet.io.xfer.BlockTransmitter.sendAsync(), "Terminated send
-123413232423 ... as we haven't heard from receiver in 1m54s."

4 freenet.io.xferPacketThrottle, Blocktransmitter: "Unable to send
throttled message, waited 60100ms"

6 freenet.io.comm.MessageCore Scheduled job ERRORS
"removeTimedOutFilters took 4000ms"

4 freenet.node.KeyTracker PacketSender "Packet 1234 sent over 600600ms
ago and still not acked". (THEY STILL EXIST!?! ;)

19 freenet.io.comm.UdpSocketHandler ERROR messages (from both darknet
and opennet), "processing packet took 4000ms" .. with times ranging
from 3000ms to 6600ms.

8 freenet.node.FNPPacketMangler "Message[1..4] timeout error:
Processing packet for freenetnode.tld:1234"

Also, in this log, I have noticed that the port number, in the
@ip:port@ portion of the error message does NOT match the port number
listed on the /friends/ web page. Why is that? (What happens, by the
way, if a darknet peer does change his port, by the way--we'd have to
re-add his node reference?)

Re: [freenet-support] Freenet 0.7 build 1175

[Dennis Nezic]

On Thu, 13 Nov 2008 18:16:55 +, Matthew Toseland
[EMAIL PROTECTED] wrote:

 Freenet 0.7 build 1175 is now available. Please upgrade. Changes:
 - Handle the Packet X sent Yms ago and still not acked bug better.
 Don't complain in the log until 10 minutes, then disconnect from the
 peer and log the fact. We can then reconnect, but we will do so
 cleanly, so hopefully not get it again for a while. It seems that
 most of the disruption caused by this bug was caused by the logging
 on every resend and not the bug itself... Of course we will need to
 fix the bug too.

State Of The (My) 1175... It feels like it's running smoother,
although it's cpu profile here hasn't changed substantially. Perhaps
just a bit, due to better logging. The log files are finally usable!
(Usually under 100K per hour now :)

So, in one random 1 hour sample, here are the main ones that I get:

4 Forcing disconnect PacketSender ERRORs, both from opennet and
darknet peers, because packets not acked after 10 minutes!. Does this
mean that NO packets have been received from these peers, or just a big
bunch of them? (I'm pretty sure my ISP doesn't kill connections, but
just drops a fraction of the packets.) For some reason, these error
messages come in pairs, first by FNPPacketMangler, then by (Dark|Open)
netPeerNode.

40 MessageCore ERRORs from RequestSender  waitFor _unclaimed iteration
took 3428ms with unclaimedFIFOSize of 533 for ret of packetTransmit or
from UdpSocketHandler checkFilters took 6260ms with unclaimedFIFOSize
of 596 for matched: true

24 freenet.io.xfer.BlockTransmitter.sendAsync(), Terminated send
-123413232423 ... as we haven't heard from receiver in 1m54s.

4 freenet.io.xferPacketThrottle, Blocktransmitter: Unable to send
throttled message, waited 60100ms

6 freenet.io.comm.MessageCore Scheduled job ERRORS
removeTimedOutFilters took 4000ms

4 freenet.node.KeyTracker PacketSender Packet 1234 sent over 600600ms
ago and still not acked. (THEY STILL EXIST!?! ;)

19 freenet.io.comm.UdpSocketHandler ERROR messages (from both darknet
and opennet), processing packet took 4000ms .. with times ranging
from 3000ms to 6600ms.

8 freenet.node.FNPPacketMangler Message[1..4] timeout error:
Processing packet for freenetnode.tld:1234

Also, in this log, I have noticed that the port number, in the
@ip:port@ portion of the error message does NOT match the port number
listed on the /friends/ web page. Why is that? (What happens, by the
way, if a darknet peer does change his port, by the way--we'd have to
re-add his node reference?)
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[EMAIL PROTECTED]

[freenet-support] Freenet 0.7 build 1178 (and 1176, 1177)

[Matthew Toseland]

Freenet 0.7 build 1178 is now available. Please upgrade. Sorry I accidentally 
released 1176, then had some bugfixes in 1177 and 1178. Anyway, the main 
changes relative to 1175:
- Try to prevent both sides of a connection attempting to rekey at the same 
time, by having the receiver start 2.5 minutes early. This seems to have been 
the root cause of many of the recent problems. It may not be completely 
fixed, but should happen much less often.
- When we disconnect from a peer because of it not acknowledging our packets, 
tell the user via a useralert as well as logging an error.
- Get rid of WouldBlockException errors as well: If we cannot send a packet to 
a peer because there are too many packets in flight and none are being 
acknowledged, don't busy-loop and spam the logs, wait for it to become 
un-blocked again. If it doesn't become un-blocked in 10 minutes, log an error 
and tell the user via a useralert.
- Change the default log rotation period to 10 minutes rather than 1 hour to 
save disk space.
- Clarify the config option for packet padding: Turning off packet padding is 
BAD. Packet padding is essential against a passive attacker.

In the next few days you may see the Probably a bug errors on the web 
interface. Hopefully once this build is mandatory (on Monday) these will go 
away. If you still see these from Tuesday onwards, let us know.

Please update, and let us know how it works out. Hopefully this build will fix 
most of the recent network problems, oversized log files and high CPU usage 
when not doing anything. If you find any bugs, please report them on the bug 
tracker: https://bugs.freenetproject.org/ . Thanks.


pgpREOBv5UTIk.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[EMAIL PROTECTED]

Re: [freenet-support] Freenet 0.7 build 1175

[Matthew Toseland]

On Friday 14 November 2008 17:40, Dennis Nezic wrote:
 On Thu, 13 Nov 2008 18:16:55 +, Matthew Toseland
 [EMAIL PROTECTED] wrote:
 
  Freenet 0.7 build 1175 is now available. Please upgrade. Changes:
  - Handle the Packet X sent Yms ago and still not acked bug better.
  Don't complain in the log until 10 minutes, then disconnect from the
  peer and log the fact. We can then reconnect, but we will do so
  cleanly, so hopefully not get it again for a while. It seems that
  most of the disruption caused by this bug was caused by the logging
  on every resend and not the bug itself... Of course we will need to
  fix the bug too.
 
 State Of The (My) 1175... It feels like it's running smoother,
 although it's cpu profile here hasn't changed substantially. Perhaps
 just a bit, due to better logging. The log files are finally usable!
 (Usually under 100K per hour now :)

It still has 100% CPU most of the time then? Apparently resulting from Freenet 
itself rather than downloads etc?

Try 1178 ... and then wait until say Wednesday ... hopefully once everyone has 
1178, it will be much better.
 
 So, in one random 1 hour sample, here are the main ones that I get:
 
 4 Forcing disconnect PacketSender ERRORs, both from opennet and
 darknet peers, because packets not acked after 10 minutes!. Does this
 mean that NO packets have been received from these peers, or just a big
 bunch of them? (I'm pretty sure my ISP doesn't kill connections, but
 just drops a fraction of the packets.) For some reason, these error
 messages come in pairs, first by FNPPacketMangler, then by (Dark|Open)
 netPeerNode.

This is caused by the bugs we're chasing at the moment. Hopefully next week it 
will have gone away, or be very rare. It means that we are still connected to 
the node, it's still sending us packets, but it apparently doesn't understand 
some of the packets we send it.
 
 40 MessageCore ERRORs from RequestSender  waitFor _unclaimed iteration
 took 3428ms with unclaimedFIFOSize of 533 for ret of packetTransmit or
 from UdpSocketHandler checkFilters took 6260ms with unclaimedFIFOSize
 of 596 for matched: true
 
 24 freenet.io.xfer.BlockTransmitter.sendAsync(), Terminated send
 -123413232423 ... as we haven't heard from receiver in 1m54s.
 
 4 freenet.io.xferPacketThrottle, Blocktransmitter: Unable to send
 throttled message, waited 60100ms
 
 6 freenet.io.comm.MessageCore Scheduled job ERRORS
 removeTimedOutFilters took 4000ms

Hopefully these are related to the current bugs. If so they should greatly 
reduce. On the other hand, they may not be ...
 
 4 freenet.node.KeyTracker PacketSender Packet 1234 sent over 600600ms
 ago and still not acked. (THEY STILL EXIST!?! ;)

Yes, we log them at the same time as the disconnections.
 
 19 freenet.io.comm.UdpSocketHandler ERROR messages (from both darknet
 and opennet), processing packet took 4000ms .. with times ranging
 from 3000ms to 6600ms.
 
 8 freenet.node.FNPPacketMangler Message[1..4] timeout error:
 Processing packet for freenetnode.tld:1234
 
 Also, in this log, I have noticed that the port number, in the
 @ip:port@ portion of the error message does NOT match the port number
 listed on the /friends/ web page. Why is that? 

It can be different, usually because of NATs, occasionally because you are 
connected both to the darknet and opennet nodes on that IP (which is a bug, 
shouldn't happen unless you have some wierd configs set).

 (What happens, by the 
 way, if a darknet peer does change his port, by the way--we'd have to
 re-add his node reference?)

Probably yes.


pgpSwGcvJVVsB.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[EMAIL PROTECTED]

Re: [freenet-support] Freenet 0.7 build 1175

[Matthew Toseland]

Please tell me, one way or the other, whether the problem is solved or whether 
Freenet continues to use lots of CPU etc etc. Thanks.

On Friday 14 November 2008 19:52, Matthew Toseland wrote:
 On Friday 14 November 2008 17:40, Dennis Nezic wrote:
  On Thu, 13 Nov 2008 18:16:55 +, Matthew Toseland
  [EMAIL PROTECTED] wrote:
  
   Freenet 0.7 build 1175 is now available. Please upgrade. Changes:
   - Handle the Packet X sent Yms ago and still not acked bug better.
   Don't complain in the log until 10 minutes, then disconnect from the
   peer and log the fact. We can then reconnect, but we will do so
   cleanly, so hopefully not get it again for a while. It seems that
   most of the disruption caused by this bug was caused by the logging
   on every resend and not the bug itself... Of course we will need to
   fix the bug too.
  
  State Of The (My) 1175... It feels like it's running smoother,
  although it's cpu profile here hasn't changed substantially. Perhaps
  just a bit, due to better logging. The log files are finally usable!
  (Usually under 100K per hour now :)
 
 It still has 100% CPU most of the time then? Apparently resulting from 
Freenet 
 itself rather than downloads etc?
 
 Try 1178 ... and then wait until say Wednesday ... hopefully once everyone 
has 
 1178, it will be much better.
  
  So, in one random 1 hour sample, here are the main ones that I get:
  
  4 Forcing disconnect PacketSender ERRORs, both from opennet and
  darknet peers, because packets not acked after 10 minutes!. Does this
  mean that NO packets have been received from these peers, or just a big
  bunch of them? (I'm pretty sure my ISP doesn't kill connections, but
  just drops a fraction of the packets.) For some reason, these error
  messages come in pairs, first by FNPPacketMangler, then by (Dark|Open)
  netPeerNode.
 
 This is caused by the bugs we're chasing at the moment. Hopefully next week 
it 
 will have gone away, or be very rare. It means that we are still connected 
to 
 the node, it's still sending us packets, but it apparently doesn't 
understand 
 some of the packets we send it.
  
  40 MessageCore ERRORs from RequestSender  waitFor _unclaimed iteration
  took 3428ms with unclaimedFIFOSize of 533 for ret of packetTransmit or
  from UdpSocketHandler checkFilters took 6260ms with unclaimedFIFOSize
  of 596 for matched: true
  
  24 freenet.io.xfer.BlockTransmitter.sendAsync(), Terminated send
  -123413232423 ... as we haven't heard from receiver in 1m54s.
  
  4 freenet.io.xferPacketThrottle, Blocktransmitter: Unable to send
  throttled message, waited 60100ms
  
  6 freenet.io.comm.MessageCore Scheduled job ERRORS
  removeTimedOutFilters took 4000ms
 
 Hopefully these are related to the current bugs. If so they should greatly 
 reduce. On the other hand, they may not be ...
  
  4 freenet.node.KeyTracker PacketSender Packet 1234 sent over 600600ms
  ago and still not acked. (THEY STILL EXIST!?! ;)
 
 Yes, we log them at the same time as the disconnections.
  
  19 freenet.io.comm.UdpSocketHandler ERROR messages (from both darknet
  and opennet), processing packet took 4000ms .. with times ranging
  from 3000ms to 6600ms.
  
  8 freenet.node.FNPPacketMangler Message[1..4] timeout error:
  Processing packet for freenetnode.tld:1234
  
  Also, in this log, I have noticed that the port number, in the
  @ip:port@ portion of the error message does NOT match the port number
  listed on the /friends/ web page. Why is that? 
 
 It can be different, usually because of NATs, occasionally because you are 
 connected both to the darknet and opennet nodes on that IP (which is a bug, 
 shouldn't happen unless you have some wierd configs set).
 
  (What happens, by the 
  way, if a darknet peer does change his port, by the way--we'd have to
  re-add his node reference?)
 
 Probably yes.
 


pgpJKAp7pRaAF.pgp
Description: PGP signature
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[EMAIL PROTECTED]

Re: [freenet-support] Freenet 0.7 build 1175

[Dennis Nezic]

On Fri, 14 Nov 2008 19:52:46 +, Matthew Toseland
[EMAIL PROTECTED] wrote:

 It still has 100% CPU most of the time then? Apparently resulting
 from Freenet itself rather than downloads etc?

Well, it's not /so/ serious that it stays at 100% all the time. It
jumps up for a few minutes, then calms down to almost nothing for a
few minutes. (The valleys are longer than the peaks.) And keeps
repeating this. I'm running another process that steadily uses on
average 25% of the (1.2GHz) cpu, and my 15minute load time is
consistently around 3. Maybe the threads are all triggered to run at
the same time, and should be staggered? 


  Also, in this log, I have noticed that the port number, in the
  @ip:port@ portion of the error message does NOT match the port
  number listed on the /friends/ web page. Why is that? 
 
 It can be different, usually because of NATs, occasionally because
 you are connected both to the darknet and opennet nodes on that IP
 (which is a bug, shouldn't happen unless you have some wierd configs
 set).

Actually, the logs do say that I was connected to this same ip address
both in opennet and darknet :|. (And the opennet port is properly
listed in the strangers page. (It seems unlikely that his darknet port
would be routed in a NAT, but not his opennet port :|.)
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[EMAIL PROTECTED]