Re: [Savannah-hackers-public] [gnu.org #912039] Migrate gnu.org/nongnu.org from CVS
Here's a first working Vagrant+Puppet environment for testing and improving :) Webpages creation update work, cf. README. Symlinks generation is in-progress, I'm not sure I have enough data (/srv/data/www-mirror??). Comments welcome! Sylvain On Mon, May 05, 2014 at 04:48:24PM -0400, Lisa Maginnis via RT wrote: Staying on (obsolete) (unmaintained) CVS is also a bad idea :) I don't have read access so fencepost:/srv/data/www-config, can you check? I've dumped a tarball of the Apache configs in your home directory: /home/b/beuc/gnu.www-config.tar.gz @sysadmin: still willing to see the latest Apache configuration and as well as update scripts including http://www.gnu.org/new-savannah- project/new.py Please see attached for a copy of new.py Let me know if you need anything else! Thank you for contributing time to look at this :) Happy Hacking, wwwgnu-nongnu.tar.gz Description: Binary data
Re: [Savannah-hackers-public] [gnu.org #912039] Migrate gnu.org/nongnu.org from CVS
On Mon, May 05, 2014 at 04:57:22PM -0400, John Sullivan via RT wrote: [karl - Thu May 01 17:51:59 2014]: 3. regardless of all that, it would be a new and significant source of complication on the savannah side. we already cannot come close to keeping up with support requests. I believe the CVS checkout-by-cron system has been identified as a major source of load on Savannah in the past. With git's hook system (and better efficiency), this could help address that current problem -- changes would be pushed when they are made rather than doing polling. This was fixed a few years ago, all updates are currently done following a notification from a CVS hook (hence the triggering webpages update output on commit). -- Sylvain
Re: [Savannah-hackers-public] [gnu.org #912039] Migrate gnu.org/nongnu.org from CVS
Hi, On Mon, May 05, 2014 at 05:22:34PM -0400, Karl Berry via RT wrote: Second of all, I would be vehemently opposed to anything that changes all web repos over to git, *forcing* people to use git. An option to use git would be one thing. Forcing it, quite another. I doubt Sylvain had a forced switch in mind. The idea appeals to me by its simplicity, but I expect this will cause enough people as grumpy as you to track me down and roast me over a celebration bonfire ;) More seriously it should ease the migration *not* to change all projects at once. I also heard good feedback from github's static pages system, so I'll have a look at it too. Third of all, I cannot envision any changes along these lines happening any time soon, since it's already been umpteen years since the original wishlist item, and the sysadmins and savannah volunteers have no more time than they ever did. Hacking an update script is only a small part of the job. I was about to say such a demotivator would turn into a self-fulfilling prophecy, when I realized you gave everybody to motivation to prove you wrong ;) Cheers! Sylvain
Re: [Savannah-hackers-public] [gnu.org #912039] Migrate gnu.org/nongnu.org from CVS
Hi, On Fri, May 02, 2014 at 05:07:04PM -0400, Karl Berry via RT wrote: @sysadmin: can you send me the gnu.org/nongnu.org update scripts and Apache config so I can see what we can do? I think the proposed functionality is a bad idea, for the reasons stated, but I nevertheless feel compelled to pass along that the Apache config is mirrored to fencepost:/srv/data/www-config (and the full live www.gnu.org web site in sibling dir www-mirror). Staying on (obsolete) (unmaintained) CVS is also a bad idea :) I don't have read access so fencepost:/srv/data/www-config, can you check? @sysadmin: still willing to see the latest Apache configuration and as well as update scripts including http://www.gnu.org/new-savannah-project/new.py Cheers! Sylvain
[Savannah-hackers-public] Anonymous Git access troubles
Hi, Looking at the sv-hack-pub, this hasn't been reported yet: anonymous git access seems to have a problem: $ git pull fatal: The remote end hung up unexpectedly $ git fetch fatal: The remote end hung up unexpectedly $ cat .git/config [...] url = git://git.sv.gnu.org/gnulib.git [...] -- Sylvain
[Savannah-hackers-public] Source IP for VCS commit notifications
Hi,
At Gna! the -commits mailing lists
(in particular, savane-comm...@gna.org)
use an IP-based restriction to prevent SPAM:
only known sender IP addresses are accepted
(in particular, Gna! et Savannah).
Today I received the message below which indicated a change in
configuration: the source IP appears to be 140.186.70.72
(vcs.savannah.gnu.org) rather than the previous 140.186.70.51.
Do you confirm?
Incidentally, I'm curious on why you moved away from the mail
smarthost on 'internal' :) It used to centralize all the mail
aliasing.
- Sylvain
- Forwarded message from Mail Delivery System
mailer-dae...@vcs.savannah.gnu.org -
Date: Sun, 31 Jul 2011 20:29:35 +
From: Mail Delivery System mailer-dae...@vcs.savannah.gnu.org
To: b...@gnu.org
Subject: Mail delivery failed: returning message to sender
This message was created automatically by mail delivery software.
A message that you sent could not be delivered to one or more of its
recipients. This is a permanent error. The following address(es) failed:
savane-comm...@gna.org
SMTP error from remote mail server after RCPT TO:savane-comm...@gna.org:
host mail.gna.org [78.40.125.82]: 550 Administrative prohibition
-- This is a copy of the message, including all the headers. --
Return-path: b...@gnu.org
Received: from Beuc by vcs.savannah.gnu.org with local (Exim 4.72)
(envelope-from b...@gnu.org)
id 1QncdM-0003Uh-S9
for savane-comm...@gna.org; Sun, 31 Jul 2011 20:29:13 +
Date: Sun, 31 Jul 2011 20:29:12 +
Message-Id: e1qncdm-0003uh...@vcs.savannah.gnu.org
To: savane-comm...@gna.org
Subject: [SCM] Savane framework branch, master, updated.
cba498bc6372c0357f5006cd08ae342ba69fbe19
X-Git-Refname: refs/heads/master
X-Git-Reftype: branch
X-Git-Oldrev: 3fa889d285dbf29d5941739051a662a3a6d8073c
X-Git-Newrev: cba498bc6372c0357f5006cd08ae342ba69fbe19
From: Sylvain Beucler b...@gnu.org
This is an automated email from the git hooks/post-receive script. It was
generated because a ref change was pushed to the repository containing
the project Savane framework.
The branch, master has been updated
via cba498bc6372c0357f5006cd08ae342ba69fbe19 (commit)
from 3fa889d285dbf29d5941739051a662a3a6d8073c (commit)
Those revisions listed above that are new to this repository have
not appeared on any other notification email; so we list those
revisions in full, below.
- Log -
http://git.savannah.gnu.org/cgit/savane.git/commit/?id=cba498bc6372c0357f5006cd08ae342ba69fbe19
commit cba498bc6372c0357f5006cd08ae342ba69fbe19
Author: Sylvain Beucler b...@beuc.net
Date: Sun Jul 31 22:27:15 2011 +0200
Django 1.3 new blocktrans syntax (1.3 has incomplete 1.2 syntax support)
diff --git a/templates/svmain/homepage.html b/templates/svmain/homepage.html
index 6263252..80e8385 100644
--- a/templates/svmain/homepage.html
+++ b/templates/svmain/homepage.html
@@ -13,10 +13,10 @@
{% blocktrans %}{{site_name}} statistics{% endblocktrans %}/a/div
div class=smaller
div class={% cycle 'boxitemalt' 'boxitem' as rowcolor %}
-{% blocktrans count nb_users as count and
'strong'|add:nb_users|add:'/strong'|safe as html %}{{html}} registered
user{% plural %}{{html}} registered users{% endblocktrans %}
+{% blocktrans with html='strong'|add:nb_users|add:'/strong'|safe
count count=nb_users %}{{html}} registered user{% plural %}{{html}} registered
users{% endblocktrans %}
/div
div class={% cycle rowcolor %}
-{% blocktrans count nb_groups as count and
'strong'|add:nb_groups|add:'/strong'|safe as html %}{{html}} hosted
project{% plural %}{{html}} hosted projects{% endblocktrans %}
+{% blocktrans with html='strong'|add:nb_groups|add:'/strong'|safe
count count=nb_groups %}{{html}} hosted project{% plural %}{{html}} hosted
projects{% endblocktrans %}
/div
{% for conf in group_confs %}
div class={% cycle rowcolor %}
---
Summary of changes:
templates/svmain/homepage.html |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
hooks/post-receive
--
Savane framework
- End forwarded message -
[Savannah-hackers-public] savane-cleanup and commits specific to Savannah
Hi Karl, I notice that you committed a notice about hotmail to 'savane-cleanup'. I suppose that this is related to the mail.gnu.org setup. The git repository is shared by all Savane instances, e.g. Gna!, so Savannah-specific changes need to be placed somewhere else (typically in project 'administration' CVS web repository). Can you fix this? Cheers! -- Sylvain
Re: [Savannah-hackers-public] savane-cleanup and commits specific to Savannah
On Tue, Jul 26, 2011 at 09:51:34PM +, Karl Berry wrote: so Savannah-specific changes need to be placed somewhere else (typically in project 'administration' CVS web repository). That file (register.php) doesn't exist in administration CVS as far as I can see (it was the first place I looked). Which leaves me no place to commit the change. Am I missing somewhere else to look? Certainly possible ... otherwise savannah scripts can't be changed without affecting gna? That doesn't seem good. Normally you don't modify the code, but you define various bits of text that are included. If you really want to modify the script, you need to fork, but at the same time we never really needed it. -- Sylvain
[Savannah-hackers-public] Re: CVS question
Hi, On Wed, Feb 23, 2011 at 12:48:53PM -0500, Michael J. Flickinger wrote: One thing I'm slightly confused about is how exactly permissions in CVS now work. As far as I can tell, the documentation doesn't clear state what exactly is going on and how exactly permissions of cvs repositories are handled. Currently, if a file in a cvs repository let's say FOO,v, isn't even owned by the group of the repository it's in, upon committing a revision of the file the group and owner change to the correct values. What exactly is going on here? CVS removes the files (it can do so thanks to the directory permissions) and recreates it with new ownership and permissions. Similar to VI's :w! command. -- Sylvain
Re: [Savannah-hackers-public] Stepping Down of Savannah Maintenance
Hi Michael, On Tue, Feb 22, 2011 at 04:04:44PM -0500, Michael J. Flickinger wrote: In the case that you actually want to stay away from the project, I'll devote more time to sysadmin'ing Savannah and step in where you left off. If you intend to pick up where I left, I support you wholeheartedly :) Cheers! -- Sylvain
[Savannah-hackers-public] Stepping Down of Savannah Maintenance
Hi, I hereby step down from my unofficial position of Savannah maintainer. I've been contributing for 7 years and now I feel the need to look for new challenges. The system architecture is documented at: http://savannah.gnu.org/maintenance/SavannahArchitecture and the up-to-date core config files can be found at: bzr branch bzr://bzr.savannah.gnu.org/administration I'll be reasonably available to answer questions people have on the system when the above documentation is not enough. Cheers! -- Sylvain
Re: [Savannah-hackers-public] Re: [gnu.org #670138] colonialone.fsf.org Dom0 upgrade
Hi, On Wed, Feb 16, 2011 at 04:19:23PM -0500, Bernardo Innocenti via RT wrote: SSH is visible but Debian 5 is still supported for at least a year, so no impact on security. SSH is also not accessible from the public internet on most of our Dom0s... Colonialone seems to be the only exception. For improved security, we could limit access to the IPs of people how need to have access? Regardless of which version of Debian we use, this would protect us from 0-day exploits and compromised keys. That would be quite inconvenient. This is also an extremely risky way to consider security, because AFAICS it makes you think running a 1000-days-old kernel (with at least 2 root privilege escalation kernel exploits around) is safe. Whenever you choose to go ahead, I could assist you any day from 10am to 4pm. Does that include going at the colo? As long as we don't make the machine unbootable, we should be able to recover it remotely from the serial console. And it's actually the 'make the machine unbootable' case that I want to cover :) That, and your expertise on possible coreboot-related Xen issues. Let us know when you have tested recent Xen some more :) -- Sylvain
[Savannah-hackers-public] Re: [gnu.org #670138] colonialone.fsf.org Dom0 upgrade
Hi, On Mon, Feb 14, 2011 at 06:08:58PM -0500, Bernie Innocenti via RT wrote: [beuc - Sun Feb 13 13:49:19 2011]: Debian Squeeze 6.0 was recently released and we want to upgrade the base Xen system. Questions: - Are you happy with the Dom0 being installed under Debian Squeeze, or would you rather another distro? Only I want to avoid running an obsolete version. We don't have any Squeeze Dom0s in production yet. A few weeks ago, I upgraded a test Dom0 from Lenny to Squeeze. It required some fixes to the scripts for configuring the bridge, but otherwise it seems to be working. There's just one DomU running on this box, so it was tested very lightly. OK. What do you generally use for Dom0? I'd recommend waiting a little longer for production use. The attack surface of Dom0s is very limited because it doesn't run any externally visible services. SSH is visible but Debian 5 is still supported for at least a year, so no impact on security. It's more a matter of avoiding last minute upgrades, and leveraging newer features (iptables TARPIT comes to mind :)). - Can we plan a date to do it together, so as to have a hacker in front of the physical keyboard in case something goes wrong? I'm never confident with Xen in that regard. Whenever you choose to go ahead, I could assist you any day from 10am to 4pm. Does that include going at the colo? -- Sylvain
Re: [Savannah-hackers-public] bzr/index.txt update
Hi, On Tue, Feb 15, 2011 at 12:39:14AM +, Karl Berry wrote: I edited administration/content/gnu-content/bzr/index.txt to change the text for the bzr pages to use the same name (my-working-copy) in the example and the text. The change hasn't been reflected on, e.g., https://savannah.gnu.org/bzr/?group=emacs yet. Still requires manual updating of the live checkout? Yes. I'm pretty sure I sent you the procedure the other day, it's probably on the wiki. -- Sylvain
Re: [Savannah-hackers-public] jobs for a package?
On Sun, Feb 13, 2011 at 10:02:50PM +, Karl Berry wrote: Apparently there isn't a page that filters jobs per-project. Well, shouldn't there be? The point of having a jobs feature is to find volunteers. If a package can't put a useful link on its home page for its open jobs, it seems to me the likelihood of finding volunteers is greatly reduced. Many people want to help a given project, not just cast around for something to do, regardless of project. If I should submit it as an enhancement somewhere, please tell me where. Unfortunately there's zero chance I will ever program it myself. Possibly there should be. We can work on it in the Savane rewrite. -- Sylvain
[Savannah-hackers-public] Re: German Translations of Savane and Savane clean-up
Hi, On Tue, Feb 08, 2011 at 01:49:28PM +0100, Joerg Kohne wrote: Hi Sylvain, after you've committedthe German language files I sentyou via email, I'vecommitted, too (because of extended ASCII characters have been replaced by ???). Unfortunately, the changes are stillnot visible in the demo of SV4, gna.org, or savannah.gnu.org. In the German's 2005/2007? translation of savannah.gnu.org, among other things, contained a serious error. For example, the new creation of mailing lists is almost impossible, because some designations reversed and misleading. How do you update a French language filewith new strings? What am I doing wrong?Can you send me a brief instruction (only execute steps)? I updated the version at Savannah and rebuilt the locales. There isn't an automated update and I overlooked that we needed to take into account your new translation. Is it ok now? -- Sylvain
Re: [Savannah-hackers-public] jobs for a package?
Hi, On Thu, Feb 10, 2011 at 06:00:37PM +, Karl Berry wrote: Is there a direct url for users to view all jobs associated with a package? Poking around at https://savannah.gnu.org/people/ I didn't see it, yet it seems like it would be a very useful thing to link to from package web pages, etc. And obviously the information is there ... There's Edit Jobs for the package maintainers, but I didn't see anything per-package for users. I feel like I'm missing something obvious ... Apparently there isn't a page that filters jobs per-project. -- Sylvain
[Savannah-hackers-public] Convert ZWiki to a decentralized wiki
Hi, I just upgraded sv.gnu.org (the frontend VM) to Debian Squeeze 6.0. I see that Zope2 is not supported anymore, and ZWiki (which runs the http://savannah.gnu.org/maintenance/ wiki) depends on it. I think it's a good opportunity to convert the maintenance wiki to a decentralized wiki, which would better fit our philosophy. Questions: - what wiki to use? suggestion: ikiwiki - how to convert suggestion: raw export+import (aka history is lost) -- Sylvain
Re: [Savannah-hackers-public] I will be unavailable
OK Mario, I hope everything is alright, hear from you soon :) - Sylvain On Sat, Jan 29, 2011 at 11:39:48AM -0600, Mario Castelan Castro wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA256 2011-01-29 in savannah-hackers-public@gnu.org thread I will be unavailable. Hello. This message is to inform I will be busy with other personal projects and hence I won't have as much time as usual to evaluate submissions in Savannah. My normal activity as project evaluator will be restored in 2010-02-28. Regards and thanks for your work. -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (GNU/Linux) iEYEAREIAAYFAk1ET4sACgkQZ4DA0TLic4gotwCdFpd0al6SIz5s/5CYmfInCkEa rMQAniKi2w6RIeSXI0V901gxp6z9s7Zb =H//D -END PGP SIGNATURE-
[Savannah-hackers-public] Re: [PATCH] notify the user by email on change of ssh keys
On Wed, Dec 08, 2010 at 03:35:54PM +0100, Jim Meyering wrote: Brian Gough wrote: This patch notifies the user by email when any new ssh keys are added to their account through the web interface. This would make it more difficult for any intruders to get access to the repositories without being discovered. From 087f9988959dd50c29de10a9e9be3896ce8189dd Mon Sep 17 00:00:00 2001 From: Brian Gough b...@gnu.org Date: Wed, 8 Dec 2010 10:43:36 + Subject: [PATCH] notify the user by email on change of ssh keys Good idea. Thanks! I applied the patch, and did a similar notification at the system level -- if the web interface is compromised, the attacker may just block the notifications. http://git.savannah.gnu.org/cgit/savane-cleanup.git/commit/?id=33d227c2c6d90c77a57f42d2bb6d00d2f4e4e420 http://git.savannah.gnu.org/cgit/savane-cleanup.git/commit/?id=91a0400e1fc5efe1b22843de871ab5e5459c2b7a -- Sylvain
[Savannah-hackers-public] Changed SSH key for backup homepage
Hi, A server with a DRBD backup of my dedicated box (which we used to store the Savannah downtime homepages) was compromised last week through exim4's remote root vulnerability. As a result I regenerated the SSH keys for the dedicated box. You should now see: RSA key fingerprint is b9:b9:60:18:57:71:67:ea:04:be:ed:d5:a9:c5:ef:b0. -- Sylvain
[Savannah-hackers-public] Re: Savannah is back up with a synchonized RAID
Peabo, we're backing up the system RIGHT NOW, don't do anything without communicating on #savannah, especially if I shut down all the VMs - Sylvain On Sun, Dec 19, 2010 at 11:48:59AM -0500, pe...@fsf.org wrote: The resynchronization of /dev/md3 is complete. I have restarted the VMs (and placed them back in the /etc/xen/auto directory). Next backup starts at 2010-12-20T07:20:00-0500.
[Savannah-hackers-public] Re: Savannah is back up with a synchonized RAID
On Sun, Dec 19, 2010 at 12:06:31PM -0500, pe...@fsf.org wrote: Peabo, we're backing up the system RIGHT NOW, don't do anything without communicating on #savannah, especially if I shut down all the VMs - Sylvain I apologize. I'm done with my part of it. We'll talk soon about what to do about installing the missing RAID components. Great. For the record (maybe you already saw in the savannah ChangeLog), the resync failed a few hours ago at 81.7%, possibly because of the VMs+resync+backup. Now the disks are sync'd, and the backup is updated, and the VMs are restarted (in that order). -- Sylvain
Re: [Savannah-hackers-public] Savannah down for several hours December 18
On Sat, Dec 18, 2010 at 11:43:13AM -0500, pe...@fsf.org wrote: The Xen host running the Savannah virtual machines failed this morning due to a RAID array failure. We have removed the bad drive from the array and rebooted the machine. The Savannah virtual machines are now back up, with one RAID array still resynchronizing. Thanks! Btw, any ETA for replacing the disks? -- Sylvain
[Savannah-hackers-public] Xen domU time drift
Hi, There's a time drift in the colonialone.fsf.org domUs: https://savannah.gnu.org/support/index.php?107551 I don't think this happened in the previous install, but maybe we just weren't aware of it. I expect domUs to sync with the dom0. Do you have a recommendation to fix it? (Do you have this issue on other Xen installs?) -- Sylvain
[Savannah-hackers-public] Replace 2 hard disks on colonialone.fsf.org
This ticket to officially remember that 2 disks need replacement on colonialone.fsf.org. -- Sylvain
Re: [Savannah-hackers-public] Re: Subversion web interface URL change
On Mon, Dec 13, 2010 at 10:09:42PM +0200, Alexander Shulgin wrote: On Fri, Dec 10, 2010 at 13:56, Alexander Shulgin alex.shul...@gmail.com wrote: Hello, This is to notify you that the Subversion web interface is no longer accessible with the URL of the following form: http://svn.sv.nongnu.org/project/ Anyone? It never was accessible that way as far as I remember. /viewvc is missing. -- Sylvain
[Savannah-hackers-public] Re: Access to GNU projects' webpages by GNU webmasters
Hi, On Wed, Dec 08, 2010 at 01:50:02AM +, Karl Berry wrote: I saw a communication from a GNU webmaster Any idea who? Matt Lee? Rob Myers? John Sullivan? Jason Self? I believe it was from Yavor. I don't think it was any of the above. Yavor has been essentially incommunicado for a while now, so I'm not sure we'll get an answer from him. It is not completely unreasonable to kill webmasters' access to the sv project /web pages. As it is, I've had to remind the other webmasters a zillion times not to edit them without talking to the maintainers. However, the webmasters could potentially help maintainers with the web pages, so it's not unreasonable for the access to remain, either. Mainly, I was surprised at the change since I didn't recall ever seeing any discussion about it. (As for me, if we decide to kill access, I suppose I could always temporarily make myself a member of the project when I need to edit the web pages (typically to say, this package is looking for a maintainer). It'd be a time-consuming pain for me, but probably better than preserving the whole group access infrastructure stuff when it's not needed for anyone else.) Well, anyway, we've reasserted that this permissions setup is necessary, I re-enabled it, cf. /etc/init.d/cvs-permissions. -- Sylvain
[Savannah-hackers-public] Access to GNU projects' webpages by GNU webmasters
Yavor, I clearly remember that at a point, I saw a communication from a GNU webmaster stating that it was no longer necessary to provide a way for 'www' members to access to per-projects webpages (www.gnu.org/s/*). I don't mind re-enabling this, but I'd like to know whether my memory is failing me. - Sylvain On Mon, Dec 06, 2010 at 05:44:44PM -0800, Karl Berry wrote: I thought that webmasters decided a little while ago that they didn't need direct privileges on project repositories? What? Not that I ever heard of. I never saw any suggestion of or discussion about this whatsoever. At any rate, *I* need it, in my maintainers capacity more than my webmasters capacity. All the time. k
[Savannah-hackers-public] Re: [Savannah-hackers-private] frontend and vcs-noshell down
Ward replaced the disk. Michael: consider using savannah-hackers-public ? - Sylvain On Sun, Dec 05, 2010 at 10:44:07AM -0500, Michael J. Flickinger wrote: Investigating issue, apparently the frontend and xen guests are down. colonialone:~# ssh frontend.in.sv.gnu.org ssh_exchange_identification: Connection closed by remote host colonialone:~# ssh vcs-noshell.in.sv.gnu.org ssh_exchange_identification: Connection closed by remote host I checked the console of the internal xen instance and this is what was spewing out: [366400.520192] === [366400.520197] INFO: task exim4:11450 blocked for more than 120 seconds. [366400.520204] echo 0 /proc/sys/kernel/hung_task_timeout_secs disables this message. [366400.520211] exim4 D 13a5e000 0 11450 11448 [366400.520220]e8270a40 0282 f57f8000 13a5e000 0001 e8270bc8 c115db40 [366400.520236]e8122040 c1a47120 0548538a 0001 c1a47120 c0112c2b c1a47140 0001 [366400.520250]c0152edc f57f8000 0002 001280d2 ed052338 ed052340 [366400.520267] Call Trace: [366400.520274] [c0112c2b] kunmap_atomic+0x35/0x4a [366400.520282] [c0152edc] get_page_from_freelist+0x387/0x466 [366400.520290] [c02cb376] __mutex_lock_slowpath+0x67/0xa9 [366400.520298] [c02cb1a2] mutex_lock+0xa/0xb [366400.520307] [c014fc17] generic_file_aio_write+0x41/0xa9 [366400.520317] [ee072015] ext3_file_write+0x19/0x83 [ext3] [366400.520330] [c0170052] do_sync_write+0xbf/0x100 [366400.520338] [c012eddc] autoremove_wake_function+0x0/0x2d [366400.520348] [c010eaf3] do_page_fault+0x605/0xb2e [366400.520357] [c010eb59] do_page_fault+0x66b/0xb2e [366400.520364] [c01b9731] security_file_permission+0xc/0xd [366400.520374] [c016ff93] do_sync_write+0x0/0x100 [366400.520383] [c01707c4] vfs_write+0x83/0x120 [366400.520390] [c0170d96] sys_write+0x3c/0x63 [366400.520399] [c0103f76] syscall_call+0x7/0xb [366400.520406] === [366419.819681] INFO: task syslogd:1066 blocked for more than 120 seconds. [366419.819700] echo 0 /proc/sys/kernel/hung_task_timeout_secs disables this message. [366419.819709] syslogd D 84505170 0 1066 1 [366419.819722]ecc13ac0 0282 c115db40 84505170 00014279 ecc13c48 c115db40 [366419.819738]ed743580 c03f3220 c03f3220 2488 0005 00137334 0001 [366419.819753] c0116913 4a4a ed79dedc ec82dc50 c012eeeb ec82dc00 ed79dedc [366419.819769] Call Trace: [366419.819780] [c0116913] __wake_up+0x29/0x39 [366419.819793] [c012eeeb] prepare_to_wait+0x12/0x4d [366419.819801] [ee0436e9] log_wait_commit+0x9f/0xfc [jbd] [366419.819817] [c012eddc] autoremove_wake_function+0x0/0x2d [366419.819825] [ee03f49a] journal_stop+0x15d/0x1ad [jbd] [366419.819838] [c018a1c0] __writeback_single_inode+0x193/0x28a [366419.819847] [c0153dc1] generic_writepages+0x1a/0x21 [366419.819859] [c0153df1] do_writepages+0x29/0x30 [366419.819866] [c014ed5f] __filemap_fdatawrite_range+0x65/0x70 [366419.819876] [c018aa7a] sync_inode+0x19/0x43 [366419.819883] [ee072164] ext3_sync_file+0x88/0x9c [ext3] [366419.819900] [c018ccdd] do_fsync+0x41/0x83 [366419.819907] [c018cd3c] __do_fsync+0x1d/0x2b [366419.819916] [c0103f76] syscall_call+0x7/0xb [366419.819924] === [366424.120090] [c0103f76] syscall_call+0x7/0xb [366424.120100] === [366471.320257] INFO: task mysqld:11181 blocked for more than 120 seconds. [366471.320277] echo 0 /proc/sys/kernel/hung_task_timeout_secs disables this message. [366471.320285] mysqldD e827abac 0 11181 6470 [366471.320294]e823c180 0286 c1ae5444 e827abac c010621f e823c308 c115db40 [366471.320312]ecb78900 32f5f402 05479c85 32f3ce08 0001761b e827abac 022d535a c1ae5444 [366471.320329]e827abac c0133348 022d535a e827abac 022d535a c1ae5444 c115db40 00d9e000 [366471.320344] Call Trace: [366471.320352] [c010621f] xen_clocksource_read+0xc/0x164 [366471.320366] [c0133348] getnstimeofday+0x37/0xbc [366471.320377] [c014fdf0] sync_page_killable+0x0/0x2a [366471.320386] [c02caf1e] io_schedule+0x49/0x80 [366471.320395] [c014e627] sync_page+0x33/0x36 [366471.320403] [c014fdf5] sync_page_killable+0x5/0x2a [366471.320410] [c02cb04a] __wait_on_bit_lock+0x2a/0x52 [366471.320419] [c014e592] __lock_page_killable+0x51/0x57 [366471.320426] [c012ee09] wake_bit_function+0x0/0x3c [366471.320434] [c01502be] generic_file_aio_read+0x331/0x4e1 [366471.320442] [c0170152] do_sync_read+0xbf/0xfe [366471.320453] [c012eddc] autoremove_wake_function+0x0/0x2d [366471.320463] [c01b9731] security_file_permission+0xc/0xd [366471.320473] [c0170093] do_sync_read+0x0/0xfe [366471.320480] [c01708e2] vfs_read+0x81/0x11e [366471.320487] [c0170d33] sys_read+0x3c/0x63 [366471.320496] [c0103f76] syscall_call+0x7/0xb [366471.320506]
[Savannah-hackers-public] Re: sv.gnu.org DNS alias
On Thu, Dec 02, 2010 at 02:29:00PM +0100, Ole Tange wrote: Hi Sylvain. Thanks for your hard work on Savannah. It seems there are a few minor things remaining: http://sv.gnu.org - should show Savannah. https://sv.gnu.org - should show Savannah. I think there's a replication delay for these. We'll have to wait for a bit. I added a work-around meanwhile. -- Sylvain
Re: [Savannah-hackers-public] Savannah and www.gnu.org downtime
Savannah.gnu and non-gnu are hosted on the same facility. - Sylvain On Thu, Dec 02, 2010 at 10:10:41AM -0800, blobnor-lis...@yahoo.com.br wrote: Hello. I have a question. Since the attack was on Savannah Non-GNU, why was GNU website affected? I mean, GNU isn't Non-GNU, so why one affected the other? Regards. Seff --- Em ter, 30/11/10, John Sullivan i...@fsf.org escreveu: De: John Sullivan i...@fsf.org Assunto: [FSF] Free Software Supporter -- Issue 32, November 2010 Para: info-...@fsf.org Data: Terça-feira, 30 de Novembro de 2010, 21:41 # Free Software Supporter Issue 32, November 2010 ### Savannah and www.gnu.org downtime Over the weekend of the 26th, Savannah was compromised, which led to vandalism on www.gnu.org. We've restored www.gnu.org, but Savannah restoration is still in progress. Here's a chronological account of the events: * http://www.fsf.org/blogs/sysadmin/savannah-and-www.gnu.org-downtime
Re: [Savannah-hackers-public] [task #10344] Deletion of PAPO
On Thu, Dec 02, 2010 at 10:51:57PM +, Alex Fernandez wrote: Update of task #10344 (project administration): Status: In Progress = Done Open/Closed:Open = Closed ___ Follow-up Comment #2: As explained in: [http://www.mail-archive.com/savannah-hackers-public@gnu.org/msg03436.html] one of the authors granted us permission to delete the project. Done now. Do you remember that removing a project from the web interface does not remove it from the system (repos, mailing lists, etc.)? -- Sylvain
Re: [Savannah-hackers-public] [task #10344] Deletion of PAPO
On Fri, Dec 03, 2010 at 12:07:20AM +0100, Alex Fernandez wrote: On Fri, Dec 3, 2010 at 12:03 AM, Sylvain Beucler b...@gnu.org wrote: As explained in: [http://www.mail-archive.com/savannah-hackers-public@gnu.org/msg03436.html] one of the authors granted us permission to delete the project. Done now. Do you remember that removing a project from the web interface does not remove it from the system (repos, mailing lists, etc.)? Oops, not really. At least the project homepage: http://savannah.nongnu.org/projects/papo seems to have stopped working. How can I remove the rest? A bit lengthy to explain, and I need sleep :/ - please ping me later. -- Sylvain
Re: [Savannah-hackers-public] www cvs history warning?
On Thu, Dec 02, 2010 at 05:09:27PM -0800, Karl Berry wrote: When I (user karl) commit to a web repository, there is a warning about the CVS history file. It doesn't happen when I commit to the www repo itself, only other project /web repos. ... Copy/paste please? -- Sylvain
[Savannah-hackers-public] Please enable webpages sync-on-commit
Hi Ward, Bernie, Peter, Can you enable back sync-on-commit for CVS webpages? Cheers! -- Sylvain
Re: [Savannah-hackers-public] Anyone have any updates on Savannah?
On Mon, Nov 29, 2010 at 12:52:37PM -0500, Paul Smith wrote: I've been watching the identi.ca site but there's really nothing informative there, except that people are aware that it's down. But no word on why it's down (unscheduled maintenance) or any indication of how long it might be down for. It's a bit painful to not have access to our source repositories, etc. Any info? Are we waiting for hardware to arrive? Someone to travel on-site? Backups to load? Hi, I'm setting up a redirection page. - Sylvain
Re: [Savannah-hackers-public] email addresses in atom feed and news entries
Hi,
Good catch!
I also made it so it also works for the frontpage.
- Sylvain
On Sat, Nov 20, 2010 at 03:05:59PM +, Brian Gough wrote:
Hello everyone.
I think it would be good if the news pages and atom feeds could be
considered public information, so that email addresses are not
obfuscated there(*).
A simplistic patch to selectively display the addreses is below.
I wasn't sure of the best way to implement it --- passing a parameter
from atom.php - markup_full - _markup_inline - utils_email_basic
would add a lot of parameters.
Therefore it seemed cleaner to use a (global) constant. Instead of
using CONTEXT it could be better to introduce an additional constant,
like ACCESS == 'public' or 'private' of course. However, I'm not
familiar enough with the savannah coding conventions to know what is
best.
Brian
(*) For example on the front page it currently says
Help with installing loggerhead, the bzr web-based browser
...
If you can do one of the above, we'd like to hear from you at -unavailable-
!
The same problem occurs when projects make public news annoucements
with email addresses included.
diff --git a/frontend/php/include/utils.php b/frontend/php/include/utils.php
index 68b778c..a4f231e 100644
--- a/frontend/php/include/utils.php
+++ b/frontend/php/include/utils.php
@@ -178,7 +178,7 @@ function utils_email ($address, $nohtml=0)
# like the previous but does no extended search, just print as it comes
function utils_email_basic ($address, $nohtml=0)
{
- if (user_isloggedin())
+ if (user_isloggedin() || CONTEXT == 'forum' || CONTEXT == 'news')
{
if ($nohtml)
{ return htmlspecialchars($address); }
Re: [Savannah-hackers-public] Issue with email notification of Savannah bugs
On Wed, Nov 10, 2010 at 10:35:41AM -0500, Paul Smith wrote: There's a problem with the email notification code in Savannah. Recently a bug was filed (http://savannah.gnu.org/bugs/?31614) in which the verbatim tag was used to provide some code examples. In the bug, the example looks correct: $ echo 'a\ b:; echo $@' | make -f - a b echo a b a b However, I have configured Savannah to forward bug notifications and in the email I got the backslash was removed; I've noticed this in other bugs as well and it's supremely annoying as it changes the entire meaning of the issue to lose the backslashes; in the email I get: $ echo 'a b:; echo $@' | make -f - a b echo a b a b Note the missing backslash which completely changes the example. Is there somewhere to file a bug about this, and/or can it be fixed? Discussing it here is fine. Would you consider writing a patch? http://git.savannah.gnu.org/cgit/savane-cleanup.git/tree/frontend/php I recommend use 'make' there: http://git.savannah.gnu.org/cgit/savane-cleanup.git/tree/tests/minimal_configs then use http://localhost:50080/ and check /tmp/savane-mini/mailbox -- Sylvain
Re: [Savannah-hackers-public] savannah sources
On Sun, Oct 31, 2010 at 10:17:01PM +, Karl Berry wrote: In sr #107497, Sylvain remarked: (Karl: savane-cleanup.git is the current live version, savane.git is the next version) Oh. So, just to be 100% sure, the CVS administration tree is dead now? I believe you are confusing the *web* CVS administration tree, which contains little bits of texts loaded by the web interface, and the complete source code (in the Git repo). Until now you modified the bits of texts, but you never had to hack the source code, as was needed yesterday to fix the hardcoded text about requests for inclusion. -- Sylvain
Re: [Savannah-hackers-public] Possibly inappropriate usage of project elisp-es by SuSo
Hi, Technically the only concern as far as Savannah is concerned, strictly speaking, is the absence of source code of the GFDL'd HTML manual, and the link to such HTML manual. I suppose that the manual is copyrighted by FSF, so you may want to talk with the FSF legal team to enforce the GFDL, if you can't find an agreement with SuSo. Whether to keep SuSo as a project member is for the project admins to decide. That's MHO :) I disagree with what you have to say but will fight to the death to protect your right to say it. (Voltaire, 1694-1778, writer, philosopher) — Voltaire SuSo if free to express what he wants on Savannah, as long as he doesn't imply Savannah promotes open source, and doesn't misrepresent Linux as a whole operating system rather than a kernel :P - Sylvain On Mon, Nov 01, 2010 at 02:18:03PM -0600, Mario Castelan Castro wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA256 2010-11-01 in savannah-hackers-public@gnu.org copy to David Vázquez dav...@es.gnu.org, Antonio Barrones antonio@gmail.com thread Possibly inappropriate usage of project elisp-es by SuSo. Hello. I have been informed and I have checked myself that user SuSo (susomc in GNU Savannah) is using the elisp-es project hosted here [1] to advertise his translation of GNU emacs documentation. However, he don't cooperate with the other members of the elisp-es project by arguing (In spanish language) it's impossible to meet the required industrial standards and requirements with the conventional free software development process, as used in GNU Savannah [2]. Apart from untrue as proven by hundreds of free projects this is _also_ a fallacy, by implying there is only one development methodology in the free software mundillo/worldish which we're forced to follow. The translated documentation has been actually published outside GNU Savannah in [2], in HTML format. The source (In texinfo format) hasn't been published as of this date, even after request by David and Antonio (Members of the elisp-es project, they are on the CC list). This behavior is inaceptable, don't just the SuSo statements spurn the value of the conventional free software development methodology, but the denial of source code difficults very much the essential right to modify the work. Basically SuSo is using GNU Savannah as a news aggregator and advertising platform to promote his sourceless translation. Please correct me if I'm wrong. I think this is a violation of the GNU Savannah hosting requirements [3] because: * GNU Savannah is a free software and documentation development platform. SuSo don't develop his translation in GNU Savannah, merely links to an external site. * The absence of source code impose an obstacle to the freedom to modify it. Savannah is meant to be used in a way which promotes free software and free documentation and hence the user freedoms. To deny the source code do the opposite. I'm going to ask SuSo to reflect about the issue and reconsider his stance. I consider inmoral -and hope you agree- the fact SuSo spurn our community and deny us the source code on one hand, and on the another he is using Savannah to promote that. Let me know your thoughts. Regards. [1]: https://savannah.nongnu.org/p/elisp-es [2]: http://gnu.manticore.es/manual-introduccion-emacs-lisp [3]: https://savannah.gnu.org/register/requirements.php -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (GNU/Linux) iEYEAREIAAYFAkzPIFsACgkQZ4DA0TLic4jjqACeIU63SFDaUTEsA+BS3s4mKx9J AhkAnjgblJyX+W0loMO2x8NIz8wsW589 =qKCI -END PGP SIGNATURE-
[Savannah-hackers-public] Re: [sr #107505] Please reduce gnash git repo size
On Mon, Oct 25, 2010 at 10:44:00PM +0200, Jim Meyering wrote: Sylvain Beucler wrote: On Mon, Oct 25, 2010 at 10:30:13PM +0200, Jim Meyering wrote: Sylvain Beucler wrote: Do you know about concurrency? I.e. need we make the repo read-only for developers when we're doing such a repack? From what I recall, that's not necessary. I think (probably a gross simplification) it creates a big pack of whatever's on hand, and then flips a ref (atomic rename) to make the new pack's objects live, and then removes the packed (and now logically unlinked) objects at its leisure. Any new objects that came in while packing are simply not packed. I guess I'll try on a copy of the Gnash repo. It seems there's some kind of special condition in that project though, I don't get nearly the same ratio on my projects (admittedly not nearly as active either :)). Those numbers are typical of a large project that's been converted from e.g., cvs, and whose repository has never been properly compressed. I committed during various phases of the 'gc' (counting, compressing, writing), all the commits were kept, just not packed. I'll optimize the real gnash.git right now - lots of b/w saved :) Speaking of git, there are messages from cvs2git piling up in the mail system, due to no alias defined for cvs2git, which is now done. I think you disabled the emacs cvs-git repo but not the cron. -- Sylvain
[Savannah-hackers-public] Re: [sr #107505] Please reduce gnash git repo size
Jim, Can you have a look at this? The gnash repo can be packed from 760M (gc) to 60M (gc --aggressive). Does that sound normal? Is there a risk to lose data with it? If not, should we plan to do it on other repo too? - Sylvain On Sun, Oct 24, 2010 at 08:14:03PM +, Sylvain Beucler wrote: Follow-up Comment #2, sr #107505 (project administration): Wow, it's down to 60M after 20mn. I'll check with Jim to have an opinion on doing it on the real repo. ___ Reply to this item at: http://savannah.gnu.org/support/?107505 ___ Message sent via/by Savannah http://savannah.gnu.org/
[Savannah-hackers-public] Re: commit issues with bzr / bzr whoami
Hi,
It's a GECOS field. The field is encoded in UTF-8 and made available
to the system through libnss-mysql-bg
(http://packages.debian.org/lenny/libnss-mysql-bg).
Reinhard: I don't think bzr should fail in such case, or to the least
it should be more explicit (the username _was_ here, after all).
- Sylvain
On Mon, Oct 11, 2010 at 09:20:36PM +0200, Jan Djärv wrote:
Sylvain Beucler skrev 2010-10-11 20.45:
(The right contact is savannah-help-pub...@gnu.org)
I suspect a bug wrt encoding.
Can you try to edit your Savannah account, get rid of accents in the
Savannah real name setting:
https://savannah.gnu.org/my/admin/change.php?item=realname
and if that works, report the bug to bzr?
That did the trick! Thanks. Just for the info in the bug report,
exactly where on the ssh/bzr pipeline on the server does the real
name I put in in savannah account ifo show up? As a user name
(GECOS-field?) or as a ssh item?
Jan D.
- Sylvain
On Mon, Oct 11, 2010 at 05:58:48AM +0200, Eli Zaretskii wrote:
Let's try this address...
Sylvain, could you please help Jan regain his write access to the
Emacs repository? See the details below.
TIA
Date: Sun, 10 Oct 2010 20:30:22 +0200
From: Jan Djärvjan@swipnet.se
CC: Sylvain Beucler via RTsysad...@gnu.org
Just to clarify, I have done bzr whoami:
% bzr whoami
Jan D.jan.h.d#swipnet.se
I also tried (jhd is my savannah login):
jan.h.d#swipnet.se
jhdjan.h.d#swipnet.se
jhd
Jan D.
Jan D.jhdgoo#gmail.com
No success. This is bzr version 2.2.1. I also tried on another machine
which
has bzr version 2.1.2, and tried two different emacs branches (trunk and
emacs-23), and a totally fresh checkout of the emacs-23 branch. No
difference, bzr update works fine, but no luck on commit. Any ideas?
Thanks,
Jan D.
Eli Zaretskii skrev 2010-10-10 20.15:
Sylvain, could you please help Jan? Since the Emacs repository
switched to bzr+ssh, he cannot commit, although he can update from the
repo. The error message is described below; I will forward the full
backtrace from his .bzr.log in a moment.
Everybody else seems to get along just fine with the new protocol,
including myself.
TIA
--- Start of forwarded message ---
X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00
autolearn=unavailable version=3.3.1
Date: Sun, 10 Oct 2010 14:54:57 +0200
From: Jan Djärvjan@swipnet.se
To: Eli Zaretskiie...@gnu.org
Cc: Andreas Schwabsch...@linux-m68k.org, emacs-de...@gnu.org
Subject: Re: Something wrong with the repo?
It is still something wrong:
bzr: ERROR: Server sent an unexpected error: ('error', 'Unable to
determine
your name. Use bzr whoami to set it.')
I tried 5 or 6 different settings to bzr whoami, no permutation worked.
Anybody know what should go in here that wasn't required previously?
Jan D.
Eli Zaretskii skrev 2010-10-10 14.40:
From: Andreas Schwabsch...@linux-m68k.org
Date: Sun, 10 Oct 2010 14:04:59 +0200
Lars Magne Ingebrigtsenla...@gnus.orgwrites:
I get the following when I try to do a bzr update:
[la...@quimbies ~/src/emacs/trunk]$ bzr update
Warning: Permanently added the RSA host key for IP address
'140.186.70.72' to the list of known hosts.
subsystem request failed on channel 0
bzr: ERROR: Unable to connect to target of bound branch
BzrBranch7('file:///home/larsi/src/emacs/trunk/') =
sftp://la...@bzr.savannah.gnu.org/srv/bzr/emacs/trunk/: Unable to
connect to SSH host bzr.savannah.gnu.org; EOF during negotiation
Use bzr[+ssh]://... instead. Apparently there is suddenly a bzr server
at savannah now. Would have been nice to get an announcement though.
I've just forwarded it, it was sent to project admins less than an
hour ago.
--- End of forwarded message ---
[Savannah-hackers-public] Re: [gnu.org #622071] colonialone: disk 'sdd' failed
Hi, On Thu, Oct 07, 2010 at 02:53:22PM -0400, Peter Olson via RT wrote: [beuc - Wed Oct 06 15:21:47 2010]: Hi, On Wed, Oct 06, 2010 at 03:05:04PM -0400, Peter Olson via RT wrote: [beuc - Wed Oct 06 14:46:46 2010]: Hi, Disk 'sdd' is not available anymore at colonialone. Smartmontools detected an issue, and mdadm removed it from the RAID array. Can you investigate and possibly replace the failed disk? Btw, did you receive the failure notifications? Thanks, We took the failed disk out of the RAID array because it appears to be a hard failure rather than a glitch (all partitions containing the disk degraded at the same time). The array contained 4 members and now contains 3 members, all in service. We expect to replace it when we next make a trip to the colo. colonialone:~# cat /proc/mdstat Personalities : [raid1] md3 : active raid1 sda6[0] sdb6[2] sdc6[1] 955128384 blocks [3/3] [UUU] md2 : active raid1 sda5[0] sdb5[2] sdc5[1] 19534976 blocks [3/3] [UUU] md1 : active raid1 sda2[0] sdb2[2] sdc2[1] 200 blocks [3/3] [UUU] md0 : active raid1 sda1[0] sdb1[2] sdc1[1] 96256 blocks [3/3] [UUU] unused devices: none I'm worried that 'dmesg' shows lots of ext3 errors. How can a failed disk in a RAID1x4 array cause *filesystem*-level errors? Do we need a fsck or something? Here are some of the errors from dmesg: [20930306.805714] ext3_orphan_cleanup: deleting unreferenced inode 86646 [20930306.805714] ext3_orphan_cleanup: deleting unreferenced inode 85820 [20930306.822520] ext3_orphan_cleanup: deleting unreferenced inode 86643 [20930306.829335] ext3_orphan_cleanup: deleting unreferenced inode 86645 [20930306.840398] EXT3-fs: dm-5: 30 orphan inodes deleted [20930306.840542] EXT3-fs: recovery complete. [20930307.015205] EXT3-fs: mounted filesystem with ordered data mode. I found some discussion on the Net that says these messages are a normal byproduct of making an LVM snapshot. Are you doing this as part of your backup procedure? Yes (cf. remote_backup.sh). Good to know it's not a disk error, thanks. I wrote a script to convert dmesg timestamps to wall clock. These messages are issued every morning between 07:58 and 08:15 (or sometimes as late as 08:27). Yes, the backup from savannah-backup.gnu.org runs at 12:00 GMT. Also, LVM is still looking for /dev/sdd7 colonialone:~# lvs /dev/sdd7: read failed after 0 of 2048 at 0: Input/output error [...] I suggest we plan a reboot this week. -- Sylvain
[Savannah-hackers-public] Re: [gnu.org #622071] colonialone: disk 'sdd' failed
Hi, On Thu, Oct 07, 2010 at 02:53:22PM -0400, Peter Olson via RT wrote: [beuc - Wed Oct 06 15:21:47 2010]: Hi, On Wed, Oct 06, 2010 at 03:05:04PM -0400, Peter Olson via RT wrote: [beuc - Wed Oct 06 14:46:46 2010]: Hi, Disk 'sdd' is not available anymore at colonialone. Smartmontools detected an issue, and mdadm removed it from the RAID array. Can you investigate and possibly replace the failed disk? Btw, did you receive the failure notifications? Thanks, We took the failed disk out of the RAID array because it appears to be a hard failure rather than a glitch (all partitions containing the disk degraded at the same time). The array contained 4 members and now contains 3 members, all in service. We expect to replace it when we next make a trip to the colo. colonialone:~# cat /proc/mdstat Personalities : [raid1] md3 : active raid1 sda6[0] sdb6[2] sdc6[1] 955128384 blocks [3/3] [UUU] md2 : active raid1 sda5[0] sdb5[2] sdc5[1] 19534976 blocks [3/3] [UUU] md1 : active raid1 sda2[0] sdb2[2] sdc2[1] 200 blocks [3/3] [UUU] md0 : active raid1 sda1[0] sdb1[2] sdc1[1] 96256 blocks [3/3] [UUU] unused devices: none I'm worried that 'dmesg' shows lots of ext3 errors. How can a failed disk in a RAID1x4 array cause *filesystem*-level errors? Do we need a fsck or something? Here are some of the errors from dmesg: [20930306.805714] ext3_orphan_cleanup: deleting unreferenced inode 86646 [20930306.805714] ext3_orphan_cleanup: deleting unreferenced inode 85820 [20930306.822520] ext3_orphan_cleanup: deleting unreferenced inode 86643 [20930306.829335] ext3_orphan_cleanup: deleting unreferenced inode 86645 [20930306.840398] EXT3-fs: dm-5: 30 orphan inodes deleted [20930306.840542] EXT3-fs: recovery complete. [20930307.015205] EXT3-fs: mounted filesystem with ordered data mode. I found some discussion on the Net that says these messages are a normal byproduct of making an LVM snapshot. Are you doing this as part of your backup procedure? Yes (cf. remote_backup.sh). Good to know it's not a disk error, thanks. I wrote a script to convert dmesg timestamps to wall clock. These messages are issued every morning between 07:58 and 08:15 (or sometimes as late as 08:27). Yes, the backup from savannah-backup.gnu.org runs at 12:00 GMT. Also, LVM is still looking for /dev/sdd7 colonialone:~# lvs /dev/sdd7: read failed after 0 of 2048 at 0: Input/output error [...] I suggest we plan a reboot this week. -- Sylvain
[Savannah-hackers-public] bzr move from sftp:// to bzr+ssh://
Hi, (This message is sent to project admins who have a bzr repository) We tried to find contributors to move bzr repositories from sftp:// to bzr+ssh:// , however this proved unsuccessful. This is becoming a problem, with some of you switching away from bzr for that reason. Consequently, I'm going to migrate the repositories myself in rush mode, that is, I'm going to do that within the next few hours, and during the process your repositories won't be accessible. I'll send another e-mail when the migration is done. After that, you'll have to update your working copies accordingly (with 'bzr switch'). -- Sylvain
Re: [Savannah-hackers-public] Setting up bzr+ssh on Savannah.
Hi Toshio and Paul, I marked you as MIA and disabled your access accordingly. I did the sftp-bzr+ssh migration this afternoon along the way. Let us know if you plan to contribute again to Savannah. I hope we'll still be able to contact you for bzr advice! -- Sylvain On Wed, Aug 04, 2010 at 10:26:15AM -0700, Toshio Kuratomi wrote: On Fri, Jul 30, 2010 at 6:57 AM, Sylvain Beucler b...@gnu.org wrote: Hi! On Thu, Jul 29, 2010 at 01:05:43AM +0200, Sylvain Beucler wrote: On Wed, Jul 28, 2010 at 04:58:34PM -0600, Paul Hummer wrote: So I'm not entirely sure what's entailed by this still. It seems like there's an existing installation and some technical decisions made that I'm not up to date on, and am feeling quite lost by just reading some wiki pages. Is there anyone who has a clear idea of what the next actions are? Check http://savannah.gnu.org/maintenance/Bzr if not already and ask me what you don't understand. Did you read it? Do you understand where to go next? Hi guys, I just got back from travelling last week and as usual, found that there was a whole ton of work related business I had to take care of waiting for me. I've plowed through the biggest task there so I can start looking into this. As for a plan of action -- Paul, since you're going to be maintaining this long term, I propose that you have most of the control wrt how this thing is setup and use me as a resource for actually writing scripts, asking questions of wrt how we could design it, etc. The important thing in my mind is that you understand and feel comfortable working with what we have in the end.
Re: [Savannah-hackers-public] Setting up bzr+ssh on Savannah.
Hi, Eventually I had to do the migration myself. Cheers, - Sylvain On Fri, Aug 20, 2010 at 11:30:10PM +0200, Sylvain Beucler wrote: On Sat, Aug 21, 2010 at 09:02:34AM +1200, Robert Collins wrote: On Sat, Aug 21, 2010 at 8:56 AM, Sylvain Beucler b...@gnu.org wrote: On Fri, Aug 20, 2010 at 03:41:02PM -0400, Karl Fogel wrote: Sylvain Beucler wrote: Robert Collins wrote: I pinged the two volunteers late last week - they haven't had an ACK about their logins to progress it further. Karl has root access and I already gave me quite some information about how Savannah works, so you can add the SSH keys and pass on the information. I said a long time ago saying that I was off this -- I don't have any time to work on it. All progress I made is publicly posted. See: http://lists.gnu.org/archive/html/savannah-hackers-public/2010-05/msg00024.html http://lists.gnu.org/archive/html/savannah-hackers-public/2010-06/msg00015.html http://lists.gnu.org/archive/html/savannah-hackers-public/2010-05/threads.html#1 Thanks, FYI the 2 volunteers have root access (for 1 month), and I gave them directions and pings, but nothing has progressed so far. I'm considering shutting down that access. Toshio said very clearly that he was going to be without internet on holiday for a month or so; Both you and Karl seem to be one month late :) Toshio said he would be that way during July and back at the beginning of _this_ month - and he did send a mail at that time stating that he's not interesting in maintaining, but in helping Paul setting up. Paul has been unwell + travelling too. Sadly we didn't receive any info about this (my last ping on July 30th, Toshio's last ping on August 4th). -- Sylvain
[Savannah-hackers-public] Re: [gnu.org #622071] colonialone: disk 'sdd' failed
Hi, On Wed, Oct 06, 2010 at 03:05:04PM -0400, Peter Olson via RT wrote: [beuc - Wed Oct 06 14:46:46 2010]: Hi, Disk 'sdd' is not available anymore at colonialone. Smartmontools detected an issue, and mdadm removed it from the RAID array. Can you investigate and possibly replace the failed disk? Btw, did you receive the failure notifications? Thanks, We took the failed disk out of the RAID array because it appears to be a hard failure rather than a glitch (all partitions containing the disk degraded at the same time). The array contained 4 members and now contains 3 members, all in service. We expect to replace it when we next make a trip to the colo. colonialone:~# cat /proc/mdstat Personalities : [raid1] md3 : active raid1 sda6[0] sdb6[2] sdc6[1] 955128384 blocks [3/3] [UUU] md2 : active raid1 sda5[0] sdb5[2] sdc5[1] 19534976 blocks [3/3] [UUU] md1 : active raid1 sda2[0] sdb2[2] sdc2[1] 200 blocks [3/3] [UUU] md0 : active raid1 sda1[0] sdb1[2] sdc1[1] 96256 blocks [3/3] [UUU] unused devices: none I'm worried that 'dmesg' shows lots of ext3 errors. How can a failed disk in a RAID1x4 array cause *filesystem*-level errors? Do we need a fsck or something? -- Sylvain
[Savannah-hackers-public] Re: [gnu.org #622071] colonialone: disk 'sdd' failed
Hi, On Wed, Oct 06, 2010 at 03:05:04PM -0400, Peter Olson via RT wrote: [beuc - Wed Oct 06 14:46:46 2010]: Hi, Disk 'sdd' is not available anymore at colonialone. Smartmontools detected an issue, and mdadm removed it from the RAID array. Can you investigate and possibly replace the failed disk? Btw, did you receive the failure notifications? Thanks, We took the failed disk out of the RAID array because it appears to be a hard failure rather than a glitch (all partitions containing the disk degraded at the same time). The array contained 4 members and now contains 3 members, all in service. We expect to replace it when we next make a trip to the colo. colonialone:~# cat /proc/mdstat Personalities : [raid1] md3 : active raid1 sda6[0] sdb6[2] sdc6[1] 955128384 blocks [3/3] [UUU] md2 : active raid1 sda5[0] sdb5[2] sdc5[1] 19534976 blocks [3/3] [UUU] md1 : active raid1 sda2[0] sdb2[2] sdc2[1] 200 blocks [3/3] [UUU] md0 : active raid1 sda1[0] sdb1[2] sdc1[1] 96256 blocks [3/3] [UUU] unused devices: none I'm worried that 'dmesg' shows lots of ext3 errors. How can a failed disk in a RAID1x4 array cause *filesystem*-level errors? Do we need a fsck or something? -- Sylvain
Re: [Savannah-hackers-public] gug-la group
On Thu, Sep 30, 2010 at 09:52:09PM +, Karl Berry wrote: FYI: Chris Bryant (user chrisbryant_ucla) wrote maintain...@gnu.org about resuming his work on the gug-la nongnu group (https://savannah.nongnu.org/projects/gug-la/). Although it's a savannah question and not a GNU question, with my savannah hat on, I saw no reason for him not to resume work on it since it was his project in the first place, and there were zero other members (admins or otherwise) of the project, so I approved his membership request in the normal way. (I also know Chris personally and am sure he is not doing anything questionable.) Ok, I'm a bit surprised because the same Chris recently retracted his offer to contribute to Savane due to unexpected time constraints. -- Sylvain
Re: [Savannah-hackers-public] Re: [gnu-advisory] gnu-group-projects
On Mon, Sep 20, 2010 at 10:03:58PM +, Karl Berry wrote: gnu-games-group, gnu-network-group would be more natural to me. Could someone remind me if there is a reason that it needs to be a prefix - if we need to identify the groups, could we not match on /^gnu-\w+-group$/ instead? There is no reason, technical or otherwise, it's just a convention. I have no objection to gnu-*-group. Consider that the new standard :). I'll write to rms next. As far as the Savannah Hackers are concerned I guess there's no objection. The next version of Savane could include a way to link projects together, so that games could be linked to such a group project. That being said, the group project needs resources for itself (repos, webpages, etc.) so it need to be a regular project too. -- Sylvain
Re: [Savannah-hackers-public] User account request
On Fri, Sep 17, 2010 at 03:04:43PM +0200, Ludovic Courtès wrote: Hi Sylvain, The account is now active https://savannah.gnu.org/users/jeffk, but I can’t find it through the “Search users” button in “Administration Summary: Manage Members” (I searched for ‘jeffk’, ‘Jeffrey’, ‘Kingston’, etc.) It's still marked as 'pending'. Have Jeff click on the link in his confirmation mail. -- Sylvain
Re: [Savannah-hackers-public] User account request
On Fri, Sep 17, 2010 at 09:39:11PM +0200, Sylvain Beucler wrote: On Fri, Sep 17, 2010 at 03:04:43PM +0200, Ludovic Courtès wrote: Hi Sylvain, The account is now active https://savannah.gnu.org/users/jeffk, but I can’t find it through the “Search users” button in “Administration Summary: Manage Members” (I searched for ‘jeffk’, ‘Jeffrey’, ‘Kingston’, etc.) It's still marked as 'pending'. Have Jeff click on the link in his confirmation mail. Also, registering the account with 'j...@it.usyd.edua.u' as the mail addy will cause troubles at a point or another... ;) I removed the pending account, please register it right this time. -- Sylvain
Re: [Savannah-hackers-public] User account request
On Thu, Sep 16, 2010 at 06:45:18PM +0200, Ludovic Courtès wrote: Hello! The ‘jeffk’ account[*] was requested several days ago and apparently wasn’t created. Jeff reports that the user page first read “Account Pending” and now reads “Invalid User Name”, which isn’t confidence-inspiring. ;-) Could you please let us know if something went wrong of if we should just be more patient? Well if the account was pending, it was waiting for Jeff to reply to the e-mail that was sent to him, containing a confirmation hash to validate his e-mail. After a few days, pending accounts are removed, explaining the Invalid User Name message. So now Jeff can re-register his account, and carefully check his mailbox :) -- Sylvain
Re: [Savannah-hackers-public] Savannah user posting spam
On Sun, Sep 05, 2010 at 10:49:03AM -0400, Paul Smith wrote: Hi all; This morning at about 1am EDT this user account: https://savannah.gnu.org/users/sevanath posted spam to 21 of the GNU make bug reports. Please disable this account; thanks! Ok. -- Sylvain
Re: [Savannah-hackers-public] new mirror redirection
On Thu, Aug 26, 2010 at 01:17:41PM +0100, Brian Gough wrote: The new mirror redirection on download.savannah.gnu.org is now live. Directories are served locally, requests for files are redirected. There's a small footer on each directory listing: Downloads will redirect to your nearest mirror site. Files on mirrors may be subject to a replication delay of up to 24 hours. In case of problems use http://download-mirror.savannah.gnu.org/releases/ Thanks! I updated the mirmon configuration to point to the new mirror URL. I noticed that redirection still happens when you don't add the trailing '/' - not sure if we should consider this a bug or a feature. This is worth a news item - are you OK with being cited on the Savannah frontpage? :) -- Sylvain
Re: [Savannah-hackers-public] new mirror redirection
Hi, On Thu, Aug 26, 2010 at 07:21:43PM +0100, Brian Gough wrote: At Thu, 26 Aug 2010 18:42:38 +0200, Sylvain Beucler wrote: I noticed that redirection still happens when you don't add the trailing '/' - not sure if we should consider this a bug or a feature. It's a small bug, but for simplicity I decided to match on a trailing slash for directories. There is a way to test for a directory in the filesystem with mod_rewrite but it's more complicated. This is worth a news item - are you OK with being cited on the Savannah frontpage? :) Yes. I noticed a few things on dl.sv.gnu.org when I was working on it: - the apache log files aren't being rotated and are HUGE (e.g. /var/log/apache2/bzr/access.log is 6.2GB). Do you want me to fix that? It just looks like /etc/logrotate.d/apache2 doesn't take into account the subdirectories. Hmm, looking at /etc/logrotate.d/, the config was overwritten during the last Debian upgrade, and apparently I forgot to fix it :/ I just put the rotation back, you don't need to do it. - /var/www/download/robots.txt is commented out so robots are downloading everything -- using quite a lot of resources. Is there a reason to leave the site open? We can handle it (especially if it's mirrored), and it seems to me that blocking robots would be detrimental to search results ranking (e.g. like http://lists.gnu.org/robots.txt). -- Sylvain
[Savannah-hackers-public] Re: Bazaar and Savannah
Hi, On Mon, Aug 23, 2010 at 09:54:59PM -0600, Paul Hummer wrote: Hi Sylvain- So here's my take on the task list: Great, so you're back and ready? :) * Update backport.org's bzr package to 2.1 - I don't think this is necessary, at least not to switch to bzr+ssh. In other words, we shouldn't block on this. We can do this later. I'd go as far as getting bzr 2.2 in, but it has some rather annoying bugs, so maybe wait for 2.2.1. * Warn all Savannah project admins who use bzr about the planned migration from sftp:// to bzr+ssh://, with a date for the switch - The switch here is what's confusing me. I think I don't know enough about the infrastructure to know the difference. I've got a login to colonialone, but it looks like bzr.savannah.gnu.org and bzr2.savannah.gnu.org are different servers. I don't think the switch is really going to require too much, but I think I need a bit more background than this. Please have a look at: http://savannah.gnu.org/maintenance/SavannahArchitecture and let me know if you still have questions. (I just update the diagram to remove an obsolete VM) * Install email notifications - I don't think this will be difficult. How do you plan to implement it? (you'll have /etc/cron.hourly/bzr_commit_mail_notification to migrate btw) * Adapt bzr repository instanciation in Savane - I'm not sure what this means. What are we adapting here? Well, Savane :) http://git.savannah.gnu.org/cgit/savane-cleanup.git/tree/lib/Savane/Bzr.pm * Move Apache2 virtual host from sftp to vcs-noshell - I think the explanation of the second task should help me understand this. * Update documentation in project administration's bzr repository - I'll read up on this and see what needs to be updated. * Sync repositories from sftp to vcs-noshell - I assuming this is syncing repositories over servers. An rsync should suffice, correct? I think rsync is enough indeed, but you tell me. Usually to avoid any inconsistency I do: - hot sync - downtime - cold re-sync (quick) - uptime * Disable bzr at sftp - This is pretty self explanatory. Can you fill in the blanks for me? I'd like to get through as much of this as I can this week. Hopefully this is clearer now. Please update the wiki page so that that blanks are filled :) -- Sylvain
Re: [Savannah-hackers-public] mirror redirection
This sounds great, thanks! Feel free to drop by #savan...@oftc when you do the change, I should be around :) -- Sylvain On Fri, Aug 20, 2010 at 09:10:35AM +0100, Brian Gough wrote: Hi This is my proposed change to redirect download links for files only, and serve directory listings locally (this was to avoid unwanted ads in directory listings on mirror sites.) I plan to put it online at the weekend. Brian Gough *** /etc/apache2/sites-available/download Fri May 7 07:04:18 2010 --- /etc/apache2/sites-available/download.no_redirect_dirs Fri Aug 20 07:54:59 2010 *** *** 68,74 Alias /mirmon/icons/ /usr/share/mirmon/icons/ ScriptAlias /awstats.pl /var/www/download/awstats.pl ! Location /releases SetHandler perl-script PerlSetVar GeoIPDBFile /usr/local/share/GeoIP/GeoIP.dat PerlSetVar GeoIPMirror /usr/local/share/GeoIP/download.txt --- 68,81 Alias /mirmon/icons/ /usr/share/mirmon/icons/ ScriptAlias /awstats.pl /var/www/download/awstats.pl ! # Send requests for files to the Apache2::Geo::Mirror redirector ! RewriteEngine On ! RewriteRule ^/releases/(.*[^/])$ /releases-redirect/$1 [NS,PT] ! ! # Serve everything else locally (i.e. directory listings) ! Alias /releases/ /srv/download/ ! ! Location /releases-redirect SetHandler perl-script PerlSetVar GeoIPDBFile /usr/local/share/GeoIP/GeoIP.dat PerlSetVar GeoIPMirror /usr/local/share/GeoIP/download.txt
Re: [Savannah-hackers-public] Setting up bzr+ssh on Savannah.
On Fri, Aug 20, 2010 at 03:41:02PM -0400, Karl Fogel wrote: Sylvain Beucler wrote: Robert Collins wrote: I pinged the two volunteers late last week - they haven't had an ACK about their logins to progress it further. Karl has root access and I already gave me quite some information about how Savannah works, so you can add the SSH keys and pass on the information. I said a long time ago saying that I was off this -- I don't have any time to work on it. All progress I made is publicly posted. See: http://lists.gnu.org/archive/html/savannah-hackers-public/2010-05/msg00024.html http://lists.gnu.org/archive/html/savannah-hackers-public/2010-06/msg00015.html http://lists.gnu.org/archive/html/savannah-hackers-public/2010-05/threads.html#1 Thanks, FYI the 2 volunteers have root access (for 1 month), and I gave them directions and pings, but nothing has progressed so far. I'm considering shutting down that access. -- Sylvain
[Savannah-hackers-public] Connect !fsfstatus to #savannah
Hi, #savannah recently moved to OFTC. Could you move gbot0 there so we get the identi.ca !fsfstatus notices about downtimes? Thanks, -- Sylvain
Re: [Savannah-hackers-public] git commits to email
On Mon, Aug 09, 2010 at 09:57:34AM -0600, Rob Savoye wrote: Could somebody enable sending the usual email messages for commits via git (or bzr for now) to our gnash-com...@gnu.org list ? We often use the commit list for code review discussions. Done :) -- Sylvain
Re: [Savannah-hackers-public] git commits to email
On Mon, Aug 09, 2010 at 12:20:07PM -0600, Rob Savoye wrote: On 08/09/10 10:33, Sylvain Beucler wrote: Done :) Thanks once again! I now get the email, but would like to replace UNNAMED PROJECT with Gnash ? I'm not sure if this is something I have to do on my end, or on the other... This is an automated email from the git hooks/post-receive script. It was generated because a ref change was pushed to the repository containing the project UNNAMED PROJECT. Done too :) (.git/description) -- Sylvain
[Savannah-hackers-public] lists.gnu.org down
Hi sysadmins, I guess you already know, but lists.gnu.org is down. (No route to host) -- Sylvain
[Savannah-hackers-public] Re: Sebastian Gerhardt quitted the project Savannah Administration
OK thanks and see you soon :) - Sylvain On Sun, Aug 08, 2010 at 08:59:06PM +0200, Sebastian Gerhardt wrote: I announced it, but it is still in the queue of mailman. :) Sebastian On Sun, 2010-08-08 at 20:50 +0200, Sylvain Beucler wrote: Goodbye? - Sylvain On Sun, Aug 08, 2010 at 06:27:52PM +, invalid.nore...@gnu.org wrote: This message is being sent to notify the administrator(s) of project Savannah Administration that Sebastian Gerhardt sgerhardt has chosen to remove him/herself from the project.
[Savannah-hackers-public] Re: Sebastian Gerhardt quitted the project Savannah Administration
Goodbye? - Sylvain On Sun, Aug 08, 2010 at 06:27:52PM +, invalid.nore...@gnu.org wrote: This message is being sent to notify the administrator(s) of project Savannah Administration that Sebastian Gerhardt sgerhardt has chosen to remove him/herself from the project. ___ Message sent via/by Savannah http://savannah.gnu.org/
[Savannah-hackers-public] Re: ciabot.py fixes
Hello, On Sun, Aug 01, 2010 at 12:38:25AM +0200, Sylvain Beucler wrote: On Sat, Jul 31, 2010 at 04:52:47PM -0400, Eric Raymond wrote: Sylvain Beucler b...@gnu.org: Hello, contrib/ciabot/ciabot.py is outdated, and http://www.catb.org/~esr/ciabot/ciabot.py is 500. What should I do? :) Nevertheless, I got a patch from the mailing list archive, and the current result has 2 bugs: - it looks for 'xmlrpc' instead of 'ciabot.xmlrpc' in the config - it does not strip the mail headers before sending XML to the XML-RPC service I can send patches, but it's probably better that I do a diff on your latest release rather than what I currently have. -- Sylvain Latest version enclosed. Sorry about the 500; it's some kind of config error at ibiblio. Thanks, and here's the patch. I suspect that renaming to 'ciabot.py.txt' would help working around the ibiblio config issue. Here's another patch that escapes the XML, otherwise commits than include '' characters get rejected. --- ciabot.py-esr 2010-07-31 22:32:47.0 + +++ ciabot.py 2010-08-08 21:54:57.0 + @@ -48,6 +48,7 @@ # import os, sys, commands, socket, urllib +from xml.sax.saxutils import escape # Changeset URL prefix for your repo: when the commit ID is appended # to this, it should point at a CGI that will display the commit @@ -134,6 +136,8 @@ files=do(git diff-tree -r --name-only '+ merged +' | sed -e '1d' -e 's-.*-file/file-') metainfo = do(git log -1 '--pretty=format:%an %ae%n%at%n%s' + merged) (author, ts, logmsg) = metainfo.split(\n) +logmsg = escape(logmsg) +author = escape(author) # This discards the part of the authors addrsss after @. # Might be be nice to ship the full email address, if not Regards, -- Sylvain
[Savannah-hackers-public] Re: ssh key
Hi Brian, I added your key. Please log to r...@colonialone.fsf.org and edit contact information in authorized_keys. Please also subscribe to (at least) savannah-hackers-priv...@gnu.org and maybe another list: http://savannah.gnu.org/maintenance/SavannahHackersCommunication Then you'll want to check r...@dl.sv.gnu.org:/etc/apache2/sites-available/ :) -- Sylvain
Re: [Savannah-hackers-public] gmorph/xmorph web co fails; grabcomics gnu
On Wed, Aug 04, 2010 at 05:57:56PM +, Karl Berry wrote: 1. FYI: I did NonGnuToGnuSysadmin for grabcomics (it was dubbed a GNU package long ago), and informed the maintainer. 2. Help needed: When I try to do a web checkout of the gmorph project (also GNU), it is somehow still referring to its previous project name, xmorph. (Sorry if I was the one who messed up renaming it; I don't remember if I did it.) Thus: $ cvs -d:ext:k...@cvs.sv.gnu.org:/web/gmorph co gmorph cvs checkout: Updating gmorph cvs [checkout aborted]: cannot stat /var/lock/cvs/web/xmorph: No such file or directory And indeed, there is no xmorph in /var/lock/cvs/web on vcs-noshell, only gmorph. Which is presumably as it should be. Sylvain, can you fix the reference to xmorph, wherever it is? Done! (/srv/cvs/web/gmorph/CVSROOT/config) -- Sylvain
Re: [Savannah-hackers-public] delete old git repo
On Thu, Aug 05, 2010 at 08:58:51AM -0600, Rob Savoye wrote: It turns out that many years ago, somebody created a git repository on savannah for Gnash. I checked, and there is nothing in there that we need, but it's keeping me from pushing my current git tree than was exported from bzr. Could somebody just delete the existing git repo so I can work around merge hell ? Done! -- Sylvain
Re: [Savannah-hackers-public] git features on Savannah
Hi, On Thu, Jul 29, 2010 at 11:54:07PM +, Karl Berry wrote: Also, in CVS I had my CVSROOT/loginfo trigger set up to send email on checkin to a mailing list; is this possible with git on Savannah? Yes, we currently do it manually. Is the manual process written down? I couldn't find it. Yes, it's in infra/git.txt. -- Sylvain
Re: [Savannah-hackers-public] Setting up bzr+ssh on Savannah.
Hi! On Thu, Jul 29, 2010 at 01:05:43AM +0200, Sylvain Beucler wrote: On Wed, Jul 28, 2010 at 04:58:34PM -0600, Paul Hummer wrote: On Wed, 28 Jul 2010 14:16:45 +0200 Sylvain Beucler b...@gnu.org wrote: On Wed, Jul 14, 2010 at 09:05:08AM -0600, Rob Savoye wrote: On 07/14/10 08:45, Sylvain Beucler wrote: Can you make a test on this temporary copy of the repository? bzr clone bzr+ssh://rsav...@bzr2.savannah.gnu.org/gnash It appears to work. After 11 minutes, I got this as the only file, which was README: This is the wrong repository. Please get current Gnash sources from: http://bzr.savannah.gnu.org/r/gnash/trunk; I got 2 volunteers to work on this but they don't answer my pings now. I'm contacting them again. So I'm not entirely sure what's entailed by this still. It seems like there's an existing installation and some technical decisions made that I'm not up to date on, and am feeling quite lost by just reading some wiki pages. Is there anyone who has a clear idea of what the next actions are? Check http://savannah.gnu.org/maintenance/Bzr if not already and ask me what you don't understand. Did you read it? Do you understand where to go next? -- Sylvain
[Savannah-hackers-public] Fw: tatchai quitted the project Thai Translation Team for the gnu's web
Hi, Is it normal that www-th does not have a project admin? - Sylvain - Forwarded message from invalid.nore...@gnu.org - Date: Fri, 02 Jul 2010 16:44:39 + From: invalid.nore...@gnu.org To: savannah-reports-priv...@gnu.org Subject: [Savannah-reports-private] tatchai quitted the project Thai Translation Team for the gnu's web User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.2.6) Gecko/20100625 Firefox/3.6.6 X-Apparently-From: 58.8.26.159 (Savane authenticated user chaitat) This message is being sent to notify the site administrator(s) that the last administrator of the project Thai Translation Team for the gnu's web (tatchai chaitat) has chosen to remove him/herself from the project. As result, the project is administrator-orphan. ___ Message sent via/by Savannah http://savannah.gnu.org/ - End forwarded message -
Re: [Savannah-hackers-public] Moving #savannah from FreeNode to OFTC
Hi, On Thu, Jul 22, 2010 at 10:09:24PM +, Karl Berry wrote: Consequently, consistent with promoting privacy (and Tor), I propose to move #savannah from FreeNode to OFTC. IMHO: 1. I would find it regrettable that savannah not use the official GNU IRC. 2. But I find it even more regrettable that rms did not see fit to consider your experience sufficient to take some kind of action. (I wonder if he realized the privacy issues.) 3. So I don't oppose the change, speaking personally. I'll be interested to see what other savannah folk think. Thanks for your point. I got an OK from Mario and Jonathan via IRC too, and no objections, so let's do that. -- Sylvain
Re: [Savannah-hackers-public] creating git repository for www-ml project
There was a configuration problem for group type www.gnu.org translation team, this is now fixed. - Sylvain On Sun, Jul 18, 2010 at 01:00:04PM +0200, Sylvain Beucler wrote: I started investigating but I didn't find the problem yet. I'll keep you informed. - Sylvain On Fri, Jul 16, 2010 at 08:19:02PM +0530, Shyam | ശ്യാം കാരനാട്ട് | Karanatt wrote: Hi, I have ticked for git repository on the select features for www-ml project on last 13th and retried it on 15th. But still not able to do clone/initial push. This is what i am getting: [sh...@shyam www-git-ml]$ git clone git://git.savannah.gnu.org/www-ml.git Cloning into www-ml... fatal: The remote end hung up unexpectedly OR [sh...@shyam www-ml]$ git push --all aeshya...@git.sv.gnu.org:/srv/git/www-ml.git Enter passphrase for key '/home/shyam/.ssh/id_rsa': fatal: '/srv/git/www-ml.git' does not appear to be a git repository fatal: The remote end hung up unexpectedly [sh...@shyam www-ml]$ the web frontend shows 404 http://git.savannah.gnu.org/gitweb/?p=www-ml.git;a=summary How to make it work? Thanks Shyam k gpg public keyserver: keys.gnupg.net
Re: [Savannah-hackers-public] Re: New Savannah volunteers (bzr support)
On Thu, Jul 15, 2010 at 09:22:06PM +0200, Sylvain Beucler wrote: On Wed, Jul 14, 2010 at 04:41:18PM +0200, Sylvain Beucler wrote: - Before we give you access to Savannah, you need to agree with the rule: What you break, you fix. Let us know if you accept the rule :) I'm sad to know that you actually have to ASK people to agree with this. I wholeheartedly agree. OK, key installed. Please log at r...@colonialone.fsf.org and fill in contact information at ~/.ssh/authorized_keys. I added some info at: http://savannah.gnu.org/maintenance/Bzr Is everything ok? Hello? -- Sylvain
Re: [Savannah-hackers-public] creating git repository for www-ml project
I started investigating but I didn't find the problem yet. I'll keep you informed. - Sylvain On Fri, Jul 16, 2010 at 08:19:02PM +0530, Shyam | ശ്യാം കാരനാട്ട് | Karanatt wrote: Hi, I have ticked for git repository on the select features for www-ml project on last 13th and retried it on 15th. But still not able to do clone/initial push. This is what i am getting: [sh...@shyam www-git-ml]$ git clone git://git.savannah.gnu.org/www-ml.git Cloning into www-ml... fatal: The remote end hung up unexpectedly OR [sh...@shyam www-ml]$ git push --all aeshya...@git.sv.gnu.org:/srv/git/www-ml.git Enter passphrase for key '/home/shyam/.ssh/id_rsa': fatal: '/srv/git/www-ml.git' does not appear to be a git repository fatal: The remote end hung up unexpectedly [sh...@shyam www-ml]$ the web frontend shows 404 http://git.savannah.gnu.org/gitweb/?p=www-ml.git;a=summary How to make it work? Thanks Shyam k gpg public keyserver: keys.gnupg.net
Re: [Savannah-hackers-public] Re: New Savannah volunteers (bzr support)
Hi, On Sat, Jul 17, 2010 at 05:17:31AM -0400, Toshio Kuratomi wrote: Will do. Note that I'm currently in the middle of two weeks of travel to open source conferences in South America. So I may be a little slow to reply to things until the end of the month. I'm a bit busy for the next few days too. - Before we give you access to Savannah, you need to agree with the rule: What you break, you fix. Let us know if you accept the rule :) I agree! :-) I added your key. Please fill in the contact information in .ssh/authorized_keys. -- Sylvain
Re: [Savannah-hackers-public] gcondra savannah mirror
On Mon, Jul 12, 2010 at 03:38:51PM +, Karl Berry wrote: FYI, I added another US mirror of savannah, ftp://gcondra.cs.washington.edu/nongnu/. Neat :)
Re: [Savannah-hackers-public] Setting up bzr+ssh on Savannah.
On Mon, May 17, 2010 at 02:59:41PM +1200, Robert Collins wrote: Newer bzr is in backports, for Lenny. I'd try that to start with. At that time, Bazaar 2.1.0 was out, but backports is still 2.0.3 - are you sure it newer bzr was backported? -- Sylvain
[Savannah-hackers-public] New Savannah volunteers (bzr support)
Hello, Tashio and Paul volunteered to work on bzr at Savannah, even though they lacked direction until now so I'm doing that right now. Tashio and Paul were introduced to me this way: Toshio is the administrator for the Fedora bzr hosting service, and has offered to help with the initial setup; Paul is one of the developers of the Launchpad code hosting service; so both are well versed in bzr deployments and of good character and reputation. Tashio, Paul: - Can you anwer this mail and write a couple sentence about your motivations? - Read http://savannah.gnu.org/maintenance/HowToBecomeASavannahHacker (more aimed at project moderators) and http://savannah.gnu.org/maintenance/SavannahHackersCommunication to get an overview of how we work, and http://savannah.gnu.org/maintenance/SavannahArchitecture to get an idea of how we're installed. - Tashio send a rsa key on the list, but not Paul, please send your keys. - Before we give you access to Savannah, you need to agree with the rule: What you break, you fix. Let us know if you accept the rule :) -- Sylvain
Re: [Savannah-hackers-public] Re: New Savannah volunteers (bzr support)
- Before we give you access to Savannah, you need to agree with the rule: What you break, you fix. Let us know if you accept the rule :) I'm sad to know that you actually have to ASK people to agree with this. I wholeheartedly agree. OK, key installed. Please log at r...@colonialone.fsf.org and fill in contact information at ~/.ssh/authorized_keys. I added some info at: http://savannah.gnu.org/maintenance/Bzr -- Sylvain
Re: [Savannah-hackers-public] Setting up bzr+ssh on Savannah.
On Wed, Jul 14, 2010 at 03:00:20PM +0200, Sylvain Beucler wrote: On Sat, Jul 10, 2010 at 09:20:07AM -0600, Rob Savoye wrote: How is this work going ? Most of the Gnash developers are abandoning savannah for launchpad because the performance of bzr on savannah is unacceptably slow due to sftp. If this can't be fixed reasonably soon, I'm going to move the entire Gnash project off of savannah and switch to Launchpad for our next release, which I don't really want to have to do We just can't handle the often 20-30 minutes it takes to checkin single line changes with sftp. It's so bad it's effecting our ability to get work done. Bzr is a GNU project, we want to use it instead of GIT, but the performance is so bad on savannah that would be our only other option. Can we please get this task completed in the near term ? Our release is scheduled for late August, btw... I'll come up with something. Can you make a test on this temporary copy of the repository? bzr clone bzr+ssh://rsav...@bzr2.savannah.gnu.org/gnash If that works we'll see about integrating it with the rest of the Savannah infrastructure. -- Sylvain
Re: [Savannah-hackers-public] empty dataexplorer files removed
Hi, On Tue, Jul 06, 2010 at 12:38:32AM +, Karl Berry wrote: FYI, at the project owner's request, I removed the files CVS_tree.md5 and CVS_tree.md5 in /srv/download/dataexplorer on sftp. He was able to overwrite them with empty files, but not rsync --delete them. This shouldn't happen, any error message? -- Sylvain
[Savannah-hackers-public] [task #10482] bugtracker manager can't changed bug submitted by field from anon to user
Follow-up Comment #2, task #10482 (project administration): Nope, because Barry is not a member of group 'make'. Better resubmit or simply mark it as public. ___ Reply to this item at: http://savannah.gnu.org/task/?10482 ___ Message posté via/par Savannah http://savannah.gnu.org/
Re: [Savannah-hackers-public] importing cvs repo
On Wed, Jun 23, 2010 at 10:31:46PM +, Karl Berry wrote: They work as in they are run with root privileges and may destroy data, Yeah, of course :). Happily, it seemed to go ok. As always, what you break, you fix. Importing a CVS repository is difficult because there's no standard way to distribute a CVS repository I understand that issue, but nevertheless, http://savannah.gnu.org/maintenance/ImportingCVSRepository right now gives obsolete clues. It still talks about vserver. Pretty unsure what host/directory is relevant now. I'm pretty sure you can figure it out. -- Sylvain
Re: [Savannah-hackers-public] importing cvs repo
On Wed, Jun 23, 2010 at 12:41:09AM +, Karl Berry wrote: Sylvain, http://savannah.gnu.org/maintenance/ImportingCVSRepository looks stale, still talks about vserver. I know someone is going to ask us to do this soon (Winfriend B for dataexplorer) and was hoping someone other than you would be able to help. On a similar front, is http://savannah.gnu.org/maintenance/RenamingProjectGroup (ie, run ~/infra/maintenance/rename_project.sh + mlists + web) ok for us mortals to use? The script looks like it's been updated, but thought I should check. Winfried is going to ask us to do that too (gde - dataexplorer). They work as in they are run with root privileges and may destroy data, and you should run them line by line if you feel unsafe. I have yet to find a way to write such scripts so they run more safely - the plan is to add the rename and delete operations in Savane as first-class operations. Importing a CVS repository is difficult because there's no standard way to distribute a CVS repository (unlike SVN that has dumps, for instance) - the script tries to deal with some corner cases but usually this requires checking if everything went right after it's done. -- Sylvain
Re: [Savannah-hackers-public] Setting up bzr+ssh on Savannah.
On Tue, Jun 15, 2010 at 09:05:11AM -0600, Paul Hummer wrote: On Tue, 15 Jun 2010 13:28:46 +0200 Sylvain Beucler b...@gnu.org wrote: On Mon, Jun 14, 2010 at 09:10:15AM +1200, Robert Collins wrote: On Wed, May 26, 2010 at 6:47 AM, Karl Fogel kfo...@red-bean.com wrote: Still waiting for answers on the question(s) quoted below. Also, I have some upcoming responsiblities that will prevent me from making much progress on this for the forseeable future. If Robert Collins is willing, Sylvain, would you be okay giving him the same access to the Savannah machines that I have? He's just as trustworthy, and frankly he's better qualified to do this anyway. No pressure, Robert. If you can't take this on, we can ask around in the Bazaar community. But you'd be a logical choice, if you're willing. -Karl Generally I would be, but I've been in the middle of moving country, which takes away pretty much all volunteer time :) I've asked in the Bazaar community if there are any volunteers to help with this, and two folk have stepped up. Both are well respected: Toshio is the administrator for the Fedora bzr hosting service, and has offered to help with the initial setup; Paul is one of the developers of the Launchpad code hosting service; so both are well versed in bzr deployments and of good character and reputation. Sylvain, Karl asked about giving me the same access he had, and I hope we can get enough access for Toshio and Paul to get the bzr+ssh service up and migrated too. Sure. Can you precise what they are willing to do? You mention that Toshio wants to help with the inital setup, but that's the easy part - what we're looking for here is somebody willing to _maintain_ the service, including handling user requests. I was under the impression that maintenance was part of what I was volunteering for. I can help with whatever you need. Nice. So about sending your SSH public keys? Also, Karl, did you recap what you did and planned to do with them? -- Sylvain
Re: [Savannah-hackers-public] Setting up bzr+ssh on Savannah.
On Mon, Jun 14, 2010 at 09:10:15AM +1200, Robert Collins wrote: On Wed, May 26, 2010 at 6:47 AM, Karl Fogel kfo...@red-bean.com wrote: Still waiting for answers on the question(s) quoted below. Also, I have some upcoming responsiblities that will prevent me from making much progress on this for the forseeable future. If Robert Collins is willing, Sylvain, would you be okay giving him the same access to the Savannah machines that I have? He's just as trustworthy, and frankly he's better qualified to do this anyway. No pressure, Robert. If you can't take this on, we can ask around in the Bazaar community. But you'd be a logical choice, if you're willing. -Karl Generally I would be, but I've been in the middle of moving country, which takes away pretty much all volunteer time :) I've asked in the Bazaar community if there are any volunteers to help with this, and two folk have stepped up. Both are well respected: Toshio is the administrator for the Fedora bzr hosting service, and has offered to help with the initial setup; Paul is one of the developers of the Launchpad code hosting service; so both are well versed in bzr deployments and of good character and reputation. Sylvain, Karl asked about giving me the same access he had, and I hope we can get enough access for Toshio and Paul to get the bzr+ssh service up and migrated too. Sure. Can you precise what they are willing to do? You mention that Toshio wants to help with the inital setup, but that's the easy part - what we're looking for here is somebody willing to _maintain_ the service, including handling user requests. -- Sylvain
[Savannah-hackers-public] Re: DSA Keys on Fencepost?
On Fri, Jun 11, 2010 at 10:19:03PM +, Karl Berry wrote:
https://savannah.gnu.org/my/admin/editsshkeys.php
I changed the text (and the example key).
Please use only RSA keys, not DSA keys.
Would you mind justifying this (aka {{citation needed}}) ?
The education goal of Savannah means we don't give orders, but explain
advices :)
--
Sylvain
[Savannah-hackers-public] Re: bzr vc pages
Hi, On Sat, Jun 05, 2010 at 10:16:50PM +, Karl Berry wrote: Hi Sylvain -- on another front, you can somehow push the new bzr/index.txt that I committed to the live savannah? https://savannah.gnu.org/support/?107395 I may have made typos or something, since I can't realistically test it. It is just text changes. Done. If there are explicit instructions about how other people can do this, please let me know. I couldn't find any. Here they are: ssh r...@savannah.gnu.org cd /etc/savane/content/ cvs update -- Sylvain
Re: [Savannah-hackers-public] Re: question about Dina Project (Aleix?)
Hi, On Fri, May 14, 2010 at 10:11:10PM +, Karl Berry wrote: i have question about Dina and savannah , is it possible to have an other repository along side of dina main repository ? we need a repository for dina official applications. but we don't want to have another home page or another issue manager . I would suggest simply making a subdirectory applications for yourself in the dina repository, and treating that as its own repository. Simpler than another project. As far as I know there is no practical way in Savane/Savannah to support multiple independent repositories (of the same type) for a single project. Sameer: I see you're using Mercurial. Karl: Git and Mercuria don't have the same 'modules' facilities than CVS and SVN, making the need for different repositories more important. It should be possible to add another Mercurial repositories. I believe Aleix did it for GNU Octave. Aleix? :) -- Sylvain
Re: [Savannah-hackers-public] renaming GNU speedx to gnukart
Hey, On Sat, Apr 24, 2010 at 12:58:46AM +, Karl Berry wrote: Hello savannah hackers, Eric Hutchins (cc'd) is the maintainer of the GNU speedx package. We'd like to rename that package to gnukart. I will update the maintainers file and contact the sysadmins about the ftp.gnu.org directory. There are no mailing lists. Other issues, though: 1. Since http://savannah.gnu.org/maintenance/RenamingProjectGroup has dire warnings about having to understand how savannah works before running the rename_project.sh script, I do not dare run it myself. Sylvain (or anyone else who understands, if there is anyone), could you please do this? I just remembered I needed to do this - don't hesitate to ping me if that happens again. And indeed, I had to fix a quoting issue in the script, so the warning is not unjustified yet ;) You're also raising a valid concern that currently, in spite of the past work to make the system square and get rid of non-standard setups, I'm the only one available to do this kind of task - probably because I'm the only permanent system hacker. We gotta find some new volunteers for the sysadmin side. 2. What, if anything, will Eric need to do from his side after the script has been run? I think you'll need to setup a redirect for gnu.org/s/speedx - gnu.org/s/gnukart -- Sylvain
Re: [Savannah-hackers-public] Setting up bzr+ssh on Savannah.
On Wed, May 05, 2010 at 10:10:20PM +, Karl Berry wrote: Ah, you mentioned that before, and then I forgot about the .in part Me too. Could we create host aliases (say, in /etc/hosts) such that ssh sftp, etc., work? I don't like it: - too much magic (no reference to internal), - in the long run, this means we'll want to maintain /etc/hosts in all subsystems, which sounds like a desync waiting to happen. If you confuse sftp.sv.gnu.org and sftp.in.sv.gnu.org, this is probably a good thing that you get an error :P -- Sylvain
[Savannah-hackers-public] Hostname changed at sftp.in.sv.gnu.org
Hi, Apparently somebody changed the hostname from 'sftp' to 'sftp.in.sv.gnu.org', which triggered an Apache warning during the logs rotation. I think it could be me, Karl, or Karl (Fogel :)). Hint: the culprit is a user of nslookup. Unless I didn't understand something, it's probably a use of 'hostname' instead of 'host': 'host' looks up a hostname, 'hostname' redefines it. I set it back to 'sftp'. -- Sylvain
Re: [Savannah-hackers-public] Code of conduct
On Thu, Apr 22, 2010 at 07:56:09PM -0500, Mario Castelán Castro wrote: Thile I don't have any objection, I also don't see any particular need to technically lock it, any more than we lock any of the other pages. Sylvain is notified of every change to the wiki (and I'm sure anyone else who wants to be notified could also be), Yes, at savannah-...@gnu.org. so spurious changes wouldn't last long. I don't mean to put the page under a technical lock, but rather to add a notice as Don't modify this page! or so. I think we can think about it when there is a change. Despite its open nature, this wiki is seldom modified by other people. -- Sylvain
Re: [Savannah-hackers-public] [gnu.org #568301] [mailer-dae...@gnu.org: Mail delivery failed: returning message to sender]
On Fri, Apr 23, 2010 at 10:22:13AM -0400, Ward Vandewege via RT wrote: [js...@gnu.org - Fri Apr 23 10:15:18 2010]: I tried sending test messages to all 3 GNUtrition mailing lists. Only bug-gnutrition worked. Were these lists created via Savannah? The 2 lists that do not work are marked as disabled, so that's no big surprise. Savannah hackers - why were the help-gnutrition info-gnutrition mailing lists disabled by renaming domains/gnu.org to domains/gnu.org-disabled? Apparently they were created, deleted, and re-created a week after (but that second time, incompletely, I guess recreation is not properly handled). -- Sylvain
Re: [Savannah-hackers-public] Setting up bzr+ssh on Savannah.
On Wed, Apr 21, 2010 at 12:17:19AM -0400, Karl Fogel wrote: Sylvain Beucler b...@gnu.org writes: I have no problem with turning sftp off. I'm glad you're working to get an official way to disable user-controlled plugins in bzr, server-side. What you also need to do is: - coordinate with other bzr users - move the bzr service to the 'vcs-noshell' VM (instead of the current 'sftp' VM) - so you don't disable SFTP access to the download area. I gave you root access to 'colonialone.fsf.org' (the rsa key only). Check: http://savannah.gnu.org/maintenance/SavannahArchitecture to access other areas of Savannah. Hey, Sylvain. I've logged into colonialone and thence to bzr.sv now, and familiarized myself with the lay of the land. Regarding this: move the bzr service to the 'vcs-noshell' VM (instead of the current 'sftp' VM) - so you don't disable SFTP access to the download area. I didn't quite understand what you mean, sorry. Can you expand it a little? What exactly is the download area and how would it disable SFTP access to it if bzr didn't move to vcs-noshell? I realize the answers are probably obvious to you, but I'm still learning how this system is set up... Do you understand that all sftp-based services are managed by the 'sftp' VM? -- Sylvain
