[fossil-users] Building with PortableMingW, OpenSSL

2010-03-31 Thread Rob Powell

Hey folks;

In my last message I was building fossil with PortableMingW.  I've figured out
the steps to add OpenSSL support.

I have compiled executables, but I have not verify functionality.

Building Fossil on Windows - mingw

Using mingw, zlib, OpenSSL

Download portable mingw 5.1.4, available from portableapps.com
  http://portableapps.com/node/18601

Start up the portable build environment by running MSYSPortable.exe
Change to a working directory:
  $ cd /c/dev/fossil/src


zlib

Download the zlib source files from zlib.net:
  http://zlib.net/zlib-1.2.4.tar.gz

Extract zlib source:
  $ cd /c/dev/fossil/src
  $ tar zxvf zlib-1.2.4.tar.gz
This puts the source code in /c/dev/fossil/src/zlib-1.2.4

Build the zlib library:
  $ cd /c/dev/fossil/src/zlib-1.2.4
  $ make -f ./win32/Makefile.gcc
(This should build with no errors.)


OpenSSL
---
Download OpenSSL from http://www.openssl.org/source/
  http://www.openssl.org/source/openssl-1.0.0.tar.gz

Extract OpenSSL source:
  $ cd /c/dev/fossil/src
  $ tar zxvf openssl-1.0.0.tar.gz
This puts the source code in /c/dev/fossil/src/openssl-1.0.0
(There are some errors about symbolic links, ignoring them for now.)

Run the config script, disable asm and shared libs:
  $ cd /c/dev/fossil/src/openssl-1.0.0
  $ ./config no-asm no-shared
  ...
  Configured for mingw.

Build libssl.a and libcrypto.a;
  $ make build_ssl build_crypto
  ...

This should build libcrypto.a and libssl.a with no errors;
  $ ls -la lib*
  -rw-r--r-- 1 rppowell Administrators 2227328 Mar 30 21:47 libcrypto.a
  -rw-r--r-- 1 rppowell Administrators  419516 Mar 30 21:45 libssl.a


fossil
--

Download fossil-src from http://www.fossil-scm.org/download.html
  http://www.fossil-scm.org/download/fossil-src-20100318142033.tar.gz

Extract fossil-src:
  $ cd /c/dev/fossil/src
  $ tar zxvf fossil-src-20100318142033.tar.gz
  $ cd fossil-src-20100318142033

Edit Makefile.w32
Enable SSL by enabling the commented value:
Before:
  # FOSSIL_ENABLE_SSL=1
After:
  FOSSIL_ENABLE_SSL=1

Add zlib and OpenSSL to TCC line:
Before:
  TCC = gcc -Os -Wall -DFOSSIL_I18N=0 -L/mingw/lib -I/mingw/include
After
  TCC = gcc -Os -Wall -DFOSSIL_I18N=0 -L/mingw/lib -I/mingw/include
  TCC += -I../zlib-1.2.4 -L../zlib-1.2.4
  TCC += -I../openssl-1.0.0/include -L../openssl-1.0.0


Edit options to LIB;
Before:
  LIB = -lmingwex -lz -lws2_32
  # OpenSSL:
  ifdef FOSSIL_ENABLE_SSL
  LIB += -lcrypto -lssl
  endif

After:
  # OpenSSL:
  ifdef FOSSIL_ENABLE_SSL
  LIB += -lssl -lcrypto -lgdi32
  endif
  LIB += -lmingwex -lz -lws2_32

Note: The order of the linking libraries is IMPORTANT!

I found that if the the 'ssl' and 'crypto' libs are linked last, there are
linking errors, such as undefined reference to 'create...@16',
undefined reference to 'shutd...@8', undefined reference to 'winm...@16', 
etc.

Build fossil;
  $ make -f Makefile.w32

This generates a 'fossil.exe'.

I have tested the basic function of the generated binary by starting the ui:
  C:\dev\fossil\src\fossil-src-20100318142033fossil.exe new TEST
  project-id: 
  server-id:  
  admin-user: rppowell (initial password is XX)

  C:\dev\fossil\fossiltestfossil-ssl.exe ui SSL
  Listening for HTTP requests on TCP port 8080
  Launch webbrowser: start http://127.0.0.1:8080/
  Type Ctrl-C to stop the HTTP server

Comments/thoughts?
-rppowell

  
_
The New Busy is not the old busy. Search, chat and e-mail from your inbox.
http://www.windowslive.com/campaign/thenewbusy?ocid=PID27925::T:WLMTAGL:ON:WL:en-US:WM_HMP:032010_3___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


Re: [fossil-users] SSL support, prebuilt binaries, Windows

2010-03-31 Thread Bjorn Toft Madsen
Thanks a lot, that fixed it.

I've verified https support and it works; sniffed the traffic too, it
is indeed all encrypted.

There is one outstanding issue, which I'm about to write a separate
email about, but that's more about how the SSL implementation handles
certificates signed by a CA, rather than self-signed.

Bjorn

On 30 March 2010 18:54, Rene de Zwart renew...@xs4all.nl wrote:

 Op Di, 30 maart, 2010 17:52, schreef Bjorn Toft Madsen:
 I managed a successful build of fossil on Windows, through the kind
 help of other list members.

 The only outstanding issue is some undefined references when
 FOSSIL_ENABLE_SSL=1:
 change
  ifdef FOSSIL_ENABLE_SSL
  LIB += -lcrypto -lssl
  endif
 to
  ifdef FOSSIL_ENABLE_SSL
  LIB = -lssl -lcrypto -lgdi32 -lmingwex -lz -lws2_32
  endif

 and it works (for me :-)



 --
 Rene de Zwart

 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


[fossil-users] Fossil GUI for local source tree operations

2010-03-31 Thread Andy Reynolds
I understand the rationale for the command line interface, of course.

I am very comfortable using the command line (I always have at least one 
dos-box open).

But sometimes you get a list of files e.g. fossil changes and for some of those 
files you would like to see the diffs - it's just easier to click.

I envisage a seperate application just shelling out to the fossil binary which 
would do the real work.

The situation is seems to be made worse by e.g. fossil reporting on all files 
in the source tree. 

I expected e.g. fossil changes to give me my current directory changes only.

If I have a lot of files out and there are many changes then fossil diff 
blatts a lot of stuff.

I still love fossil though!

andy___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


[fossil-users] HTTPS implementation doesn't remember certificates signed by CAs

2010-03-31 Thread Bjorn Toft Madsen
Through much help from list-members, I've finally managed to build a
SSL-capable binary of fossil.

On self-signed certificates the process is exactly as described by the
original author (Dmitry, BCCed): Fossil displays certificate info
before prompting to always accept, temporarily accept or reject the
certificate. If always is chosen, the certificate is remembered in
the global fossil DB (AppData/Local/_fossil on Windows, ~/.fossil on
*nix I believe). Next time, fossil doesn't ask but silently accepts.

On certificates signed by a CA, however, the always option doesn't
stick, with symptoms as described in
http://www.mail-archive.com/fossil-users@lists.fossil-scm.org/msg01289.html.

What happens is that, yes, the certificate is stored and retrieved in
the global fossil DB, but the SSL_get_verify_result() call returns 19
(the certificate chain could be built up using the untrusted
certificates but the root could not be found locally., from
http://www.openssl.org/docs/apps/verify.html#item_19).

Not knowing OpenSSL very well, I'm guessing this is because we never
add the CA (and any intermediates) into the X509 store
(X509_STORE_add_cert), but only the end client certificate. As a
result, the verification process fails because it can't find the
trusted CA/intermediate certificate in the X509 store.

If that analysis is correct, I think there are two options for fixing the issue:
1) Accept X509_V_ERR_SELF_SIGNED_CERT_IN_CHAIN == 19 as a good
result. At the minute the code requires X509_V_OK to connect without
prompting the user. I do not know OpenSSL well enough to understand if
X509_V_ERR_SELF_SIGNED_CERT_IN_CHAIN _only_ happens when the
end-certificate is valid, but the CAs/IntermediateCAs aren't known.
If that was indeed the case, accepting
X509_V_ERR_SELF_SIGNED_CERT_IN_CHAIN seems valid enough to me.

2) Use SSL_get_peer_cert_chain rather than SSL_get_peer_certificate
when retrieving the server certificate, then store all certificates
into the global DB. Before connecting, add all certificates belonging
to this server to the X509 store (calling X509_STORE_add_cert multiple
times, in effect).

I don't know of Dmitry is still active on this list, so I've taken the
liberty to BCC him to hear his opinion on either option. I'm happy to
have a whack at either option, but thought I should seek some advice
before doing so.

Thanks,
Bjorn
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


Re: [fossil-users] HTTPS implementation doesn't remember certificates signed by CAs

2010-03-31 Thread Bjorn Toft Madsen
Silly me, I forgot to mention that I've added a public fossil repo at
https://www.toftmadsen.org/public.fossil

The certificate has been signed by an intermediate, in turn signed by
a root CA. Feel free to clone to verify this issue and/or any fix
anybody may be contemplating.

Thanks
Bjorn

On 31 March 2010 14:57, Bjorn Toft Madsen bj...@toftmadsen.org wrote:
 Through much help from list-members, I've finally managed to build a
 SSL-capable binary of fossil.

 On self-signed certificates the process is exactly as described by the
 original author (Dmitry, BCCed): Fossil displays certificate info
 before prompting to always accept, temporarily accept or reject the
 certificate. If always is chosen, the certificate is remembered in
 the global fossil DB (AppData/Local/_fossil on Windows, ~/.fossil on
 *nix I believe). Next time, fossil doesn't ask but silently accepts.

 On certificates signed by a CA, however, the always option doesn't
 stick, with symptoms as described in
 http://www.mail-archive.com/fossil-users@lists.fossil-scm.org/msg01289.html.

 What happens is that, yes, the certificate is stored and retrieved in
 the global fossil DB, but the SSL_get_verify_result() call returns 19
 (the certificate chain could be built up using the untrusted
 certificates but the root could not be found locally., from
 http://www.openssl.org/docs/apps/verify.html#item_19).

 Not knowing OpenSSL very well, I'm guessing this is because we never
 add the CA (and any intermediates) into the X509 store
 (X509_STORE_add_cert), but only the end client certificate. As a
 result, the verification process fails because it can't find the
 trusted CA/intermediate certificate in the X509 store.

 If that analysis is correct, I think there are two options for fixing the 
 issue:
 1) Accept X509_V_ERR_SELF_SIGNED_CERT_IN_CHAIN == 19 as a good
 result. At the minute the code requires X509_V_OK to connect without
 prompting the user. I do not know OpenSSL well enough to understand if
 X509_V_ERR_SELF_SIGNED_CERT_IN_CHAIN _only_ happens when the
 end-certificate is valid, but the CAs/IntermediateCAs aren't known.
 If that was indeed the case, accepting
 X509_V_ERR_SELF_SIGNED_CERT_IN_CHAIN seems valid enough to me.

 2) Use SSL_get_peer_cert_chain rather than SSL_get_peer_certificate
 when retrieving the server certificate, then store all certificates
 into the global DB. Before connecting, add all certificates belonging
 to this server to the X509 store (calling X509_STORE_add_cert multiple
 times, in effect).

 I don't know of Dmitry is still active on this list, so I've taken the
 liberty to BCC him to hear his opinion on either option. I'm happy to
 have a whack at either option, but thought I should seek some advice
 before doing so.

 Thanks,
 Bjorn

___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


[fossil-users] TONS of errors after upgrade, rebuild does not help

2010-03-31 Thread Zed A. Shaw
Hey,

I upgraded to the latest fossil, with mine not being that old at all
(only a few months), and even after several rebuilds I get a ton of
bizarre errors:

-- logging into the server
unable to open database file
UPDATE user SET
cookie='1/02AAB0D7FFB698CD97B848BEC1159977A498AA58CDD03213E9',
ipaddr='127.0.0.1',   cexpire=julianday('now')+31557600/86400.0 WHERE
uid=1

-- fossil up
Error: Database error: unable to open database file
DELETE FROM unclustered WHERE rid IN (SELECT rid FROM private)

-- fossil clone
Error: Database error: unable to open database file
DELETE FROM unclustered WHERE rid IN (SELECT rid FROM private)


So what's going on?  Is it some kind of permissions problem?  The files
are owned by the user, and the web interface works right until it tries
to write like the above UPDATE and DELETE commands.  They're being run
out of inetd with:

4545 stream tcp nowait.1000 root /usr/bin/fossil /usr/bin/fossil http
/home/USER/fossils/source.fossil

Did fossil change this or something?

-- 
Zed A. Shaw
http://zedshaw.com/
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


Re: [fossil-users] TONS of errors after upgrade, rebuild does not help

2010-03-31 Thread Zed A. Shaw
On Wed, Mar 31, 2010 at 07:25:47AM -0700, Zed A. Shaw wrote:
 Hey,
 
 I upgraded to the latest fossil, with mine not being that old at all
 (only a few months), and even after several rebuilds I get a ton of
 bizarre errors:
... 
 4545 stream tcp nowait.1000 root /usr/bin/fossil /usr/bin/fossil http
 /home/USER/fossils/source.fossil
 
 Did fossil change this or something?

Yep, looks like running out of inetd is now read only.  If I run it with
fossil serve it's just fine.

Is there a reason this changed?  What was the change for?

Sending out a changelog to the mailing list when there's a new drop, and
using version numbers that aren't giant hashes with a date in them would
*really* help people keep up.

-- 
Zed A. Shaw
http://zedshaw.com/
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


Re: [fossil-users] TONS of errors after upgrade, rebuild does not help

2010-03-31 Thread D. Richard Hipp

On Mar 31, 2010, at 10:25 AM, Zed A. Shaw wrote:

 Hey,

 I upgraded to the latest fossil, with mine not being that old at all
 (only a few months), and even after several rebuilds I get a ton of
 bizarre errors:

 -- logging into the server
 unable to open database file
 UPDATE user SET
 cookie='1/02AAB0D7FFB698CD97B848BEC1159977A498AA58CDD03213E9',
 ipaddr='127.0.0.1',   cexpire=julianday('now')+31557600/86400.0 WHERE
 uid=1

 -- fossil up
 Error: Database error: unable to open database file
 DELETE FROM unclustered WHERE rid IN (SELECT rid FROM private)

 -- fossil clone
 Error: Database error: unable to open database file
 DELETE FROM unclustered WHERE rid IN (SELECT rid FROM private)


 So what's going on?  Is it some kind of permissions problem?  The  
 files
 are owned by the user, and the web interface works right until it  
 tries
 to write like the above UPDATE and DELETE commands.  They're being run
 out of inetd with:

 4545 stream tcp nowait.1000 root /usr/bin/fossil /usr/bin/fossil http
 /home/USER/fossils/source.fossil

 Did fossil change this or something?

Nothing has changed in Fossil, that I know of, that would cause this.

Are you sure your permissions are right?  Do you have write permission  
on the directory that contains the source.fossil file?



 -- 
 Zed A. Shaw
 http://zedshaw.com/
 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

D. Richard Hipp
d...@hwaci.com



___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


Re: [fossil-users] TONS of errors after upgrade, rebuild does not help

2010-03-31 Thread D. Richard Hipp

On Mar 31, 2010, at 10:33 AM, Zed A. Shaw wrote:

 On Wed, Mar 31, 2010 at 07:25:47AM -0700, Zed A. Shaw wrote:
 Hey,

 I upgraded to the latest fossil, with mine not being that old at all
 (only a few months), and even after several rebuilds I get a ton of
 bizarre errors:
 ...
 4545 stream tcp nowait.1000 root /usr/bin/fossil /usr/bin/fossil http
 /home/USER/fossils/source.fossil

 Did fossil change this or something?

 Yep, looks like running out of inetd is now read only.  If I run it  
 with
 fossil serve it's just fine.

 Is there a reason this changed?  What was the change for?

Perhaps this:  http://www.fossil-scm.org/fossil/info/7ba10f1a6a

But that change (really a bug fix) was in August 2009.  How old was  
your prior version of Fossil, did you say?



 Sending out a changelog to the mailing list when there's a new drop,  
 and
 using version numbers that aren't giant hashes with a date in them  
 would
 *really* help people keep up.

 -- 
 Zed A. Shaw
 http://zedshaw.com/
 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

D. Richard Hipp
d...@hwaci.com



___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


Re: [fossil-users] TONS of errors after upgrade, rebuild does not help

2010-03-31 Thread Zed A. Shaw
On Wed, Mar 31, 2010 at 10:34:16AM -0400, D. Richard Hipp wrote:
 
  4545 stream tcp nowait.1000 root /usr/bin/fossil /usr/bin/fossil http
  /home/USER/fossils/source.fossil
 
  Did fossil change this or something?
 
 Nothing has changed in Fossil, that I know of, that would cause this.
 
 Are you sure your permissions are right?  Do you have write permission  
 on the directory that contains the source.fossil file?

The previous binary was only a couple months old.  In fact, if you go
look at the list archives I had problems back when you changed how logins
work and had to upgrade all my sites.

So, that means I've been running fine, through multiple upgrades since
that change you posted, including with the chroot jail working, and only
just today did it break.

These files are stored in the user's home directory, accessible by that
user, so a chroot to become that user should work.

All signs point to some recent change in the way inetd operation works
that would cause this to happen now.

-- 
Zed A. Shaw
http://zedshaw.com/
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


Re: [fossil-users] TONS of errors after upgrade, rebuild does not help

2010-03-31 Thread D. Richard Hipp

On Mar 31, 2010, at 10:45 AM, Zed A. Shaw wrote:

 All signs point to some recent change in the way inetd operation works
 that would cause this to happen now.


I don't recall changing anything associated with the inetd logic.  Of  
course, that might just mean that I forgot...

What does the /test_env URL to your inetd server tell you?  (example: 
http://www.fossil-scm.org/fossil/test_env) 
   Any clues there?

I just checked in a change - http://www.fossil-scm.org/fossil/ci/99fea6cde4 
  - that logs all SQLite errors to the reply HTML.  If you can  
recompile to the latest, it might give us better clues about what is  
going wrong.

D. Richard Hipp
d...@hwaci.com



___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


Re: [fossil-users] TONS of errors after upgrade, rebuild does not help

2010-03-31 Thread Heinrich Huss
It seems I'm running in the same problem. After recompile I'm not able  
to login.


It looks like the sqite cannot open the database with write access.  
Filepermissions are ok and I did a rebuild.


-rwxrw-rw-1 adminadminist 10977280 Mar 31 18:47 fossil.fossil*


I've got the following error message:
SQLITE_CANTOPEN: cannot open file at source line 26160

SQLITE_CANTOPEN: statement aborts at 31: [UPDATE user SET  
cookie='1/669CBDB7E9AB8F1565A3C3F19FB96E3900C5C9B8F18BE6031C',  
ipaddr='10.0.0.10', cexpire=julianday('now')+31557600/86400.0 WHERE  
uid=1] unable to open database file


Database Error

unable to open database file
UPDATE user SET  
cookie='1/669CBDB7E9AB8F1565A3C3F19FB96E3900C5C9B8F18BE6031C',  
ipaddr='10.0.0.10',   cexpire=julianday('now')+31557600/86400.0 WHERE  
uid=1
If you have recently updated your fossil executable, you might need to  
run fossil all rebuild to bring the repository schemas up to date.




Any ideas?



Thanks

Hein



Am 31.03.2010 um 16:56 schrieb D. Richard Hipp:



On Mar 31, 2010, at 10:45 AM, Zed A. Shaw wrote:


All signs point to some recent change in the way inetd operation  
works

that would cause this to happen now.



I don't recall changing anything associated with the inetd logic.  Of
course, that might just mean that I forgot...

What does the /test_env URL to your inetd server tell you?   
(example: http://www.fossil-scm.org/fossil/test_env)

  Any clues there?

I just checked in a change - http://www.fossil-scm.org/fossil/ci/99fea6cde4
 - that logs all SQLite errors to the reply HTML.  If you can
recompile to the latest, it might give us better clues about what is
going wrong.

D. Richard Hipp
d...@hwaci.com



___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


Re: [fossil-users] TONS of errors after upgrade, rebuild does not help

2010-03-31 Thread D. Richard Hipp

On Mar 31, 2010, at 1:04 PM, Heinrich Huss wrote:

 It seems I'm running in the same problem. After recompile I'm not  
 able to login.

 It looks like the sqite cannot open the database with write access.  
 Filepermissions are ok and I did a rebuild.

 -rwxrw-rw-1 adminadminist 10977280 Mar 31 18:47 fossil.fossil*


 I've got the following error message:
 SQLITE_CANTOPEN: cannot open file at source line 26160

Please recompile with http://www.fossil-scm.org/fossil/ci/a158c4e75d  
and let me know what error message you get then.



 SQLITE_CANTOPEN: statement aborts at 31: [UPDATE user SET  
 cookie='1/669CBDB7E9AB8F1565A3C3F19FB96E3900C5C9B8F18BE6031C',  
 ipaddr='10.0.0.10', cexpire=julianday('now')+31557600/86400.0 WHERE  
 uid=1] unable to open database file

 Database Error
 unable to open database file
 UPDATE user SET  
 cookie='1/669CBDB7E9AB8F1565A3C3F19FB96E3900C5C9B8F18BE6031C',  
 ipaddr='10.0.0.10',   cexpire=julianday('now')+31557600/86400.0  
 WHERE uid=1
 If you have recently updated your fossil executable, you might need  
 to run fossil all rebuild to bring the repository schemas up to  
 date.



 Any ideas?



 Thanks

 Hein



 Am 31.03.2010 um 16:56 schrieb D. Richard Hipp:


 On Mar 31, 2010, at 10:45 AM, Zed A. Shaw wrote:

 All signs point to some recent change in the way inetd operation  
 works
 that would cause this to happen now.


 I don't recall changing anything associated with the inetd logic.  Of
 course, that might just mean that I forgot...

 What does the /test_env URL to your inetd server tell you?   
 (example: http://www.fossil-scm.org/fossil/test_env)
   Any clues there?

 I just checked in a change - http://www.fossil-scm.org/fossil/ci/99fea6cde4
  - that logs all SQLite errors to the reply HTML.  If you can
 recompile to the latest, it might give us better clues about what is
 going wrong.

 D. Richard Hipp
 d...@hwaci.com



 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil- 
 users

 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

D. Richard Hipp
d...@hwaci.com



___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


Re: [fossil-users] TONS of errors after upgrade, rebuild does not help

2010-03-31 Thread Heinrich Huss

I have recompiled. Now I do have the following fossil version:

Fossil version [a158c4e75d] 2010-03-31 17:14:18

This is the error message I got:

SQLITE_CANTOPEN: cannot open file [/share/HDA_DATA/Public/Hein/ 
fossil.fossil-journal]: No such file or directory


SQLITE_CANTOPEN: statement aborts at 31: [UPDATE user SET  
cookie='1/0F420A8378AC596FC324F1ED6E647662CF99B3BE8144E2DA70',  
ipaddr='10.0.0.10', cexpire=julianday('now')+31557600/86400.0 WHERE  
uid=1] unable to open database file


Database Error

unable to open database file
UPDATE user SET  
cookie='1/0F420A8378AC596FC324F1ED6E647662CF99B3BE8144E2DA70',  
ipaddr='10.0.0.10',   cexpire=julianday('now')+31557600/86400.0 WHERE  
uid=1
If you have recently updated your fossil executable, you might need to  
run fossil all rebuild to bring the repository schemas up to date.




Thanks for your help.

Hein





Am 31.03.2010 um 19:15 schrieb D. Richard Hipp:



On Mar 31, 2010, at 1:04 PM, Heinrich Huss wrote:


It seems I'm running in the same problem. After recompile I'm not
able to login.

It looks like the sqite cannot open the database with write access.
Filepermissions are ok and I did a rebuild.

-rwxrw-rw-1 adminadminist 10977280 Mar 31 18:47  
fossil.fossil*



I've got the following error message:
SQLITE_CANTOPEN: cannot open file at source line 26160


Please recompile with http://www.fossil-scm.org/fossil/ci/a158c4e75d
and let me know what error message you get then.




SQLITE_CANTOPEN: statement aborts at 31: [UPDATE user SET
cookie='1/669CBDB7E9AB8F1565A3C3F19FB96E3900C5C9B8F18BE6031C',
ipaddr='10.0.0.10', cexpire=julianday('now')+31557600/86400.0 WHERE
uid=1] unable to open database file

Database Error
unable to open database file
UPDATE user SET
cookie='1/669CBDB7E9AB8F1565A3C3F19FB96E3900C5C9B8F18BE6031C',
ipaddr='10.0.0.10',   cexpire=julianday('now')+31557600/86400.0
WHERE uid=1
If you have recently updated your fossil executable, you might need
to run fossil all rebuild to bring the repository schemas up to
date.



Any ideas?



Thanks

Hein



Am 31.03.2010 um 16:56 schrieb D. Richard Hipp:



On Mar 31, 2010, at 10:45 AM, Zed A. Shaw wrote:


All signs point to some recent change in the way inetd operation
works
that would cause this to happen now.



I don't recall changing anything associated with the inetd logic.   
Of

course, that might just mean that I forgot...

What does the /test_env URL to your inetd server tell you?
(example: http://www.fossil-scm.org/fossil/test_env)
 Any clues there?

I just checked in a change - http://www.fossil-scm.org/fossil/ci/99fea6cde4
- that logs all SQLite errors to the reply HTML.  If you can
recompile to the latest, it might give us better clues about what is
going wrong.

D. Richard Hipp
d...@hwaci.com



___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-
users


___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil- 
users


D. Richard Hipp
d...@hwaci.com



___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


Re: [fossil-users] TONS of errors after upgrade, rebuild does not help

2010-03-31 Thread D. Richard Hipp

On Mar 31, 2010, at 1:42 PM, Heinrich Huss wrote:

 I have recompiled. Now I do have the following fossil version:

 Fossil version [a158c4e75d] 2010-03-31 17:14:18

 This is the error message I got:

 SQLITE_CANTOPEN: cannot open file [/share/HDA_DATA/Public/Hein/ 
 fossil.fossil-journal]: No such file or directory

SQLite is unable to create and open a new rollback journal.  Does your  
process have write permission on the directory?



 SQLITE_CANTOPEN: statement aborts at 31: [UPDATE user SET  
 cookie='1/0F420A8378AC596FC324F1ED6E647662CF99B3BE8144E2DA70',  
 ipaddr='10.0.0.10', cexpire=julianday('now')+31557600/86400.0 WHERE  
 uid=1] unable to open database file

 Database Error
 unable to open database file
 UPDATE user SET  
 cookie='1/0F420A8378AC596FC324F1ED6E647662CF99B3BE8144E2DA70',  
 ipaddr='10.0.0.10',   cexpire=julianday('now')+31557600/86400.0  
 WHERE uid=1
 If you have recently updated your fossil executable, you might need  
 to run fossil all rebuild to bring the repository schemas up to  
 date.



 Thanks for your help.

 Hein





 Am 31.03.2010 um 19:15 schrieb D. Richard Hipp:


 On Mar 31, 2010, at 1:04 PM, Heinrich Huss wrote:

 It seems I'm running in the same problem. After recompile I'm not
 able to login.

 It looks like the sqite cannot open the database with write access.
 Filepermissions are ok and I did a rebuild.

 -rwxrw-rw-1 adminadminist 10977280 Mar 31 18:47  
 fossil.fossil*


 I've got the following error message:
 SQLITE_CANTOPEN: cannot open file at source line 26160

 Please recompile with http://www.fossil-scm.org/fossil/ci/a158c4e75d
 and let me know what error message you get then.



 SQLITE_CANTOPEN: statement aborts at 31: [UPDATE user SET
 cookie='1/669CBDB7E9AB8F1565A3C3F19FB96E3900C5C9B8F18BE6031C',
 ipaddr='10.0.0.10', cexpire=julianday('now')+31557600/86400.0 WHERE
 uid=1] unable to open database file

 Database Error
 unable to open database file
 UPDATE user SET
 cookie='1/669CBDB7E9AB8F1565A3C3F19FB96E3900C5C9B8F18BE6031C',
 ipaddr='10.0.0.10',   cexpire=julianday('now')+31557600/86400.0
 WHERE uid=1
 If you have recently updated your fossil executable, you might need
 to run fossil all rebuild to bring the repository schemas up to
 date.



 Any ideas?



 Thanks

 Hein



 Am 31.03.2010 um 16:56 schrieb D. Richard Hipp:


 On Mar 31, 2010, at 10:45 AM, Zed A. Shaw wrote:

 All signs point to some recent change in the way inetd operation
 works
 that would cause this to happen now.


 I don't recall changing anything associated with the inetd  
 logic.  Of
 course, that might just mean that I forgot...

 What does the /test_env URL to your inetd server tell you?
 (example: http://www.fossil-scm.org/fossil/test_env)
  Any clues there?

 I just checked in a change - http://www.fossil-scm.org/fossil/ci/99fea6cde4
 - that logs all SQLite errors to the reply HTML.  If you can
 recompile to the latest, it might give us better clues about what  
 is
 going wrong.

 D. Richard Hipp
 d...@hwaci.com



 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-
 users

 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

 D. Richard Hipp
 d...@hwaci.com



 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil- 
 users

 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

D. Richard Hipp
d...@hwaci.com



___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


[fossil-users] Character entity references

2010-03-31 Thread Higham, Paul
Is it possible in Fossil wiki markup to add special characters from the
W3C character entity references?  It would be quite useful to have these
in documentation.  For example, in html, a rightwards double arrow is
specified as either rArr or #8658, but the wiki page displays both of
these as just the text code with no rendering.

 

Is there wiki magic for this?

 

Paul Higham

Senior Software Engineer

Cardiac Rhythm Management Division

 

St. Jude Medical

709 East Evelyn Avenue

Sunnyvale, CA 95086 USA

Tel +1 408 522 6225

phig...@sjm.com

 

This communication, including any attachments, may contain information that is 
proprietary, privileged, confidential or legally exempt from disclosure.  If 
you are not a named addressee, you are hereby notified that you are not 
authorized to read, print, retain a copy of or disseminate any portion of this 
communication without the consent of the sender and that doing so may be 
unlawful.  If you have received this communication in error, please immediately 
notify the sender via return e-mail and delete it from your system.
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


Re: [fossil-users] TONS of errors after upgrade, rebuild does not help

2010-03-31 Thread D. Richard Hipp

On Mar 31, 2010, at 1:56 PM, D. Richard Hipp wrote:


 On Mar 31, 2010, at 1:42 PM, Heinrich Huss wrote:

 I have recompiled. Now I do have the following fossil version:

 Fossil version [a158c4e75d] 2010-03-31 17:14:18

 This is the error message I got:

 SQLITE_CANTOPEN: cannot open file [/share/HDA_DATA/Public/Hein/
 fossil.fossil-journal]: No such file or directory

 I think I know the problem.  Fossil now opens the database file
 *before* it enters the chroot jail.  But SQLite remembers the original
 path to the database, not the chroot-ed path.

 I guess what I need to do is close the database before entering the
 chroot jail then reopen the database under the new name

OK.  This problem should be fixed now (as of 
http://www.fossil-scm.org/fossil/info/42ba7b97aa) 
.  Please let me know if you have any other problems.



 SQLITE_CANTOPEN: statement aborts at 31: [UPDATE user SET
 cookie='1/0F420A8378AC596FC324F1ED6E647662CF99B3BE8144E2DA70',
 ipaddr='10.0.0.10', cexpire=julianday('now')+31557600/86400.0 WHERE
 uid=1] unable to open database file

 Database Error
 unable to open database file
 UPDATE user SET
 cookie='1/0F420A8378AC596FC324F1ED6E647662CF99B3BE8144E2DA70',
 ipaddr='10.0.0.10',   cexpire=julianday('now')+31557600/86400.0
 WHERE uid=1
 If you have recently updated your fossil executable, you might need
 to run fossil all rebuild to bring the repository schemas up to
 date.



 Thanks for your help.

 Hein





 Am 31.03.2010 um 19:15 schrieb D. Richard Hipp:


 On Mar 31, 2010, at 1:04 PM, Heinrich Huss wrote:

 It seems I'm running in the same problem. After recompile I'm not
 able to login.

 It looks like the sqite cannot open the database with write access.
 Filepermissions are ok and I did a rebuild.

 -rwxrw-rw-1 adminadminist 10977280 Mar 31 18:47
 fossil.fossil*


 I've got the following error message:
 SQLITE_CANTOPEN: cannot open file at source line 26160

 Please recompile with http://www.fossil-scm.org/fossil/ci/a158c4e75d
 and let me know what error message you get then.



 SQLITE_CANTOPEN: statement aborts at 31: [UPDATE user SET
 cookie='1/669CBDB7E9AB8F1565A3C3F19FB96E3900C5C9B8F18BE6031C',
 ipaddr='10.0.0.10', cexpire=julianday('now')+31557600/86400.0 WHERE
 uid=1] unable to open database file

 Database Error
 unable to open database file
 UPDATE user SET
 cookie='1/669CBDB7E9AB8F1565A3C3F19FB96E3900C5C9B8F18BE6031C',
 ipaddr='10.0.0.10',   cexpire=julianday('now')+31557600/86400.0
 WHERE uid=1
 If you have recently updated your fossil executable, you might need
 to run fossil all rebuild to bring the repository schemas up to
 date.



 Any ideas?



 Thanks

 Hein



 Am 31.03.2010 um 16:56 schrieb D. Richard Hipp:


 On Mar 31, 2010, at 10:45 AM, Zed A. Shaw wrote:

 All signs point to some recent change in the way inetd operation
 works
 that would cause this to happen now.


 I don't recall changing anything associated with the inetd
 logic.  Of
 course, that might just mean that I forgot...

 What does the /test_env URL to your inetd server tell you?
 (example: http://www.fossil-scm.org/fossil/test_env)
 Any clues there?

 I just checked in a change - 
 http://www.fossil-scm.org/fossil/ci/99fea6cde4
 - that logs all SQLite errors to the reply HTML.  If you can
 recompile to the latest, it might give us better clues about what
 is
 going wrong.

 D. Richard Hipp
 d...@hwaci.com



 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-
 users

 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

 D. Richard Hipp
 d...@hwaci.com



 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-
 users

 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil- 
 users

 D. Richard Hipp
 d...@hwaci.com



 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

D. Richard Hipp
d...@hwaci.com



___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


Re: [fossil-users] Character entity references

2010-03-31 Thread Ross Berteig
At 11:30 AM 3/31/2010, you wrote:
 Is it possible in Fossil wiki markup to add special characters
 from the W3C character entity references?  It would be quite
 useful to have these in documentation.  For example, in html, a
 rightwards double arrow is specified as either rArr or #8658,
 but the wiki page displays both of these as just the text code
 with no rendering.

I just checked in the Sandbox page of the wiki, and both codes
work just fine, as long as they include both the ampersand at the
front and the semicolon at the end... without either punctuation
mark, they are not a proper HTML entity reference.

Ross Berteig   r...@cheshireeng.com
Cheshire Engineering Corp.   http://www.CheshireEng.com/

___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


Re: [fossil-users] TONS of errors after upgrade, rebuild does not help

2010-03-31 Thread Heinrich Huss
Hi Richard,

thanks, That really helps!!

It's working again.

Cheers
Hein

Am 31.03.2010 um 20:43 schrieb D. Richard Hipp:


 On Mar 31, 2010, at 1:56 PM, D. Richard Hipp wrote:


 On Mar 31, 2010, at 1:42 PM, Heinrich Huss wrote:

 I have recompiled. Now I do have the following fossil version:

 Fossil version [a158c4e75d] 2010-03-31 17:14:18

 This is the error message I got:

 SQLITE_CANTOPEN: cannot open file [/share/HDA_DATA/Public/Hein/
 fossil.fossil-journal]: No such file or directory

 I think I know the problem.  Fossil now opens the database file
 *before* it enters the chroot jail.  But SQLite remembers the  
 original
 path to the database, not the chroot-ed path.

 I guess what I need to do is close the database before entering the
 chroot jail then reopen the database under the new name

 OK.  This problem should be fixed now (as of 
 http://www.fossil-scm.org/fossil/info/42ba7b97aa)
 .  Please let me know if you have any other problems.



 SQLITE_CANTOPEN: statement aborts at 31: [UPDATE user SET
 cookie='1/0F420A8378AC596FC324F1ED6E647662CF99B3BE8144E2DA70',
 ipaddr='10.0.0.10', cexpire=julianday('now')+31557600/86400.0 WHERE
 uid=1] unable to open database file

 Database Error
 unable to open database file
 UPDATE user SET
 cookie='1/0F420A8378AC596FC324F1ED6E647662CF99B3BE8144E2DA70',
 ipaddr='10.0.0.10',   cexpire=julianday('now')+31557600/86400.0
 WHERE uid=1
 If you have recently updated your fossil executable, you might need
 to run fossil all rebuild to bring the repository schemas up to
 date.



 Thanks for your help.

 Hein





 Am 31.03.2010 um 19:15 schrieb D. Richard Hipp:


 On Mar 31, 2010, at 1:04 PM, Heinrich Huss wrote:

 It seems I'm running in the same problem. After recompile I'm not
 able to login.

 It looks like the sqite cannot open the database with write  
 access.
 Filepermissions are ok and I did a rebuild.

 -rwxrw-rw-1 adminadminist 10977280 Mar 31 18:47
 fossil.fossil*


 I've got the following error message:
 SQLITE_CANTOPEN: cannot open file at source line 26160

 Please recompile with http://www.fossil-scm.org/fossil/ci/ 
 a158c4e75d
 and let me know what error message you get then.



 SQLITE_CANTOPEN: statement aborts at 31: [UPDATE user SET
 cookie='1/669CBDB7E9AB8F1565A3C3F19FB96E3900C5C9B8F18BE6031C',
 ipaddr='10.0.0.10', cexpire=julianday('now')+31557600/86400.0  
 WHERE
 uid=1] unable to open database file

 Database Error
 unable to open database file
 UPDATE user SET
 cookie='1/669CBDB7E9AB8F1565A3C3F19FB96E3900C5C9B8F18BE6031C',
 ipaddr='10.0.0.10',   cexpire=julianday('now')+31557600/86400.0
 WHERE uid=1
 If you have recently updated your fossil executable, you might  
 need
 to run fossil all rebuild to bring the repository schemas up to
 date.



 Any ideas?



 Thanks

 Hein



 Am 31.03.2010 um 16:56 schrieb D. Richard Hipp:


 On Mar 31, 2010, at 10:45 AM, Zed A. Shaw wrote:

 All signs point to some recent change in the way inetd operation
 works
 that would cause this to happen now.


 I don't recall changing anything associated with the inetd
 logic.  Of
 course, that might just mean that I forgot...

 What does the /test_env URL to your inetd server tell you?
 (example: http://www.fossil-scm.org/fossil/test_env)
 Any clues there?

 I just checked in a change - 
 http://www.fossil-scm.org/fossil/ci/99fea6cde4
 - that logs all SQLite errors to the reply HTML.  If you can
 recompile to the latest, it might give us better clues about what
 is
 going wrong.

 D. Richard Hipp
 d...@hwaci.com



 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-
 users

 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

 D. Richard Hipp
 d...@hwaci.com



 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-
 users

 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-
 users

 D. Richard Hipp
 d...@hwaci.com



 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil- 
 users

 D. Richard Hipp
 d...@hwaci.com



 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


Re: [fossil-users] Fossil GUI for local source tree operations

2010-03-31 Thread Steve Landers
Folks,

Re a Fossi GUI, please note I've suggested a Tk Fossil GUI in the Google Summer 
of Code (GSOC) Tcl/TK projects. Search for Fossil in http://wiki.tcl.tk/23186

If you know of any students who would like to earn $5k over the summer whilst 
learning about GUI design and development (not to mention DVCS), please 
encourage them to visit the Tclers Wiki at http://wiki.tcl.tk/_/recent and note 
the links to the GSOC at the top of the Recent Changes page.

Steve
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users