Re: [fossil-users] Fossil local repository no more able to push checkin (error 141)

2016-08-09 Thread Étienne Deparis 
Hello there,

It's a long time ago, but I finally found time to identify precisely the
cause of my problem.

This error was caused by the reverse proxy closing the connexion due to
the size of the request. It emits the 413 http error code (413 Payload
Too Large, previously called 413 Request Entity Too Large).

This hard closing seems to occur after fossil has begun to send data,
hence the SIGPIPE error.

After having incruised my server max request size parameter, the problem
is solved.

On the other hand, I don't know much fossil code and this story let me
wondering how is managed exception during the data sending. Does it
seems possible to listen for http error code before, during or after
having send data to display information to the user? If it is already
the case, I'll try to look further why in my case it silently crashes
without saying anything else.

Best regards,

Étienne

Le 11/04/2016 à 10:34, Étienne Deparis a écrit :
> On Sun, 9 Apr 2016, Andy Bradford wrote:
> 
>> Thus said =?ISO-8859-15?Q?=C9tienne_Deparis?= on Fri, 08 Apr 2016
>> 09:36:01 +0200:
>>
>>> Thank you  for your quick  reply. I did  not think about  bash related
>>> problem. I found interesting stuff when searching about bash and error
>>> code 141: it seems it is caused by the SIGPIPE error [1].
>>
>> What version  of Fossil  are you  using? On  which OS?  And how  are you
>> syncing? HTTP, HTTPS, SSH?
>>
> 
> Thank your for your interest. I didn't have time this week-end to
> investigate on my problem. I'd like to test various configuration on my
> reverse proxy as I suspect it is responsible.
> 
> I use fossil version 1.34 [62dcb00e68] 2015-11-02 17:35:44 UTC on
> Archlinux and I'm syncing through an https (letsencrypt) connection and
> a hiawatha webserver as a reverse-proxy to another hiawatha installation
> in a LXC container serving my different repository as cgi scripts.
> 
> Best regards,
> 
> Étienne
> 
>> Thanks,
>>
>> Andy
>>
> ___
> fossil-users mailing list
> fossil-users@lists.fossil-scm.org
> http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

-- 
Étienne Deparis

0x8C9CEE3D
https://etienne.depar.is/
xmpp: etie...@depar.is
twitter: @milouse
06 01 64 22 55



signature.asc
Description: OpenPGP digital signature
___
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 local repository no more able to push checkin (error 141)

2016-04-11 Thread Étienne Deparis 

On Sun, 9 Apr 2016, Andy Bradford wrote:


Thus said =?ISO-8859-15?Q?=C9tienne_Deparis?= on Fri, 08 Apr 2016 09:36:01 
+0200:


Thank you  for your quick  reply. I did  not think about  bash related
problem. I found interesting stuff when searching about bash and error
code 141: it seems it is caused by the SIGPIPE error [1].


What version  of Fossil  are you  using? On  which OS?  And how  are you
syncing? HTTP, HTTPS, SSH?



Thank your for your interest. I didn't have time this week-end to
investigate on my problem. I'd like to test various configuration on my
reverse proxy as I suspect it is responsible.

I use fossil version 1.34 [62dcb00e68] 2015-11-02 17:35:44 UTC on
Archlinux and I'm syncing through an https (letsencrypt) connection and
a hiawatha webserver as a reverse-proxy to another hiawatha installation
in a LXC container serving my different repository as cgi scripts.

Best regards,

Étienne


Thanks,

Andy


___
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 local repository no more able to push checkin (error 141)

2016-04-09 Thread Andy Bradford 
Thus said =?ISO-8859-15?Q?=C9tienne_Deparis?= on Fri, 08 Apr 2016 09:36:01 
+0200:

> Thank you  for your quick  reply. I did  not think about  bash related
> problem. I found interesting stuff when searching about bash and error
> code 141: it seems it is caused by the SIGPIPE error [1].

What version  of Fossil  are you  using? On  which OS?  And how  are you
syncing? HTTP, HTTPS, SSH?

Thanks,

Andy
-- 
TAI64 timestamp: 40005709e296


___
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 local repository no more able to push checkin (error 141)

2016-04-08 Thread Étienne Deparis 

On Thu, 7 Apr 2016, Richard Hipp wrote:


On 4/7/16, Étienne Deparis  wrote:


I encounter no problem commiting (and thus, pushing) upstream, but
for one of them pulling is ok, commiting is ok but the « push phase » of
syncing fails with a return code of 141 (as you can see on screen [3])


That 141 seems to be something your shell is generating.  I don't know
what it means.  I grepped the Fossil sources for 141 and didn't find
anything.  I don't think Fossil ever generates any kind of "141"
error.



Thank you for your quick reply. I did not think about bash related
problem. I found interesting stuff when searching about bash and error
code 141: it seems it is caused by the SIGPIPE error [1]. It may be
caused by something betweek my computer and my server, causing the http
connection to unexpectidly be shut down and thus breaking fossil piping
its data somewhere in its subsystem during push phasis.

I'll try to investigate this way. Thanks for the tip.

Best regards,

Étienne



[1] http://www.linuxmisc.com/12-unix-shell/81a38efe331b23e2.htm

--
Étienne Deparis

0x8C9CEE3D
https://etienne.depar.is/
xmpp: etie...@depar.is
twitter: @milouse
06 01 64 22 55
___
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 local repository no more able to push checkin (error 141)

2016-04-07 Thread Richard Hipp 
On 4/7/16, Étienne Deparis  wrote:
>
> Since yesterday, one of my machine stopped pushing to upstream. On all
> other machines (I had some stuff to do, so several commit helped me to
> test) I encounter no problem commiting (and thus, pushing) upstream, but
> for one of them pulling is ok, commiting is ok but the « push phase » of
> syncing fails with a return code of 141 (as you can see on screen [3])

That 141 seems to be something your shell is generating.  I don't know
what it means.  I grepped the Fossil sources for 141 and didn't find
anything.  I don't think Fossil ever generates any kind of "141"
error.


-- 
D. Richard Hipp
d...@sqlite.org
___
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 local repository no more able to push checkin (error 141)

2016-04-07 Thread Étienne Deparis 

Hello,

First of all, I'm very sorry if this problem has yet been
discussed. I've search on the wiki/doc and quickly on the net, but in
fact I'm not very sure of the cause of my problem, so find an answer was
impossible. I need some help to better understand what is going on my
repository. How to extract log file or debug information from fossil for
instance.

My case :

I've an upstream repository available at [1]

I've cloned this repository on various machines of mine and usually
update it without problem (I use fossil sync by default, thus any commit
made is directly pushed upstream).

This upstream repository is served by a cgi script, as described here
[2] with simple cgi :

#!/usr/bin/fossil
repository: /home/fossil/repo.fossil

I use hiawatha webserver to expose this cgi script inside a chroot on a
custom (non ssl) port, and a second hiawatha webserver elsewhere as
reverse proxy, which expose the https port.

Since yesterday, one of my machine stopped pushing to upstream. On all
other machines (I had some stuff to do, so several commit helped me to
test) I encounter no problem commiting (and thus, pushing) upstream, but
for one of them pulling is ok, commiting is ok but the « push phase » of
syncing fails with a return code of 141 (as you can see on screen [3])

Any other attempt to simply push this repository fails with the same
error code.

Question :

 - what is the signification of this error code ?
 - can I do something to get some more informations on what is
   happening ?
 - could the problem be related to ssl or reverse-proxy configuration
   (even if, for simpler commit (one file modified) it seems to works) ?

Thank you very much for any advice or help on this subject.

Cheers,

Étienne


[1] https://projects.depar.is/divers
[2] http://fossil-scm.org/index.html/doc/trunk/www/server.wiki
[3] http://ovh.to/81uGgtF

--
Étienne Deparis

0x8C9CEE3D
https://etienne.depar.is/
xmpp: etie...@depar.is
twitter: @milouse
06 01 64 22 55
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users