savannah-hackers-public
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Savannah-hackers-public] Re: Error in triggering updates


From: Yavor Doganov
Subject: [Savannah-hackers-public] Re: Error in triggering updates
Date: Fri, 05 Jun 2009 20:38:52 +0300
User-agent: Wanderlust/2.15.5 (Almost Unreal) SEMI/1.14.6 (Maruoka) FLIM/1.14.9 (Goj┼Ź) APEL/10.7 Emacs/22.3 (i486-pc-linux-gnu) MULE/5.0 (SAKAKI)

Andrea Pescetti wrote:
> As Savannah was down for a few days recently, I had accumulated a small
> queue of changes to upload to the gnu.org website CVS repository.
> 
> When I committed them, yesterday, I got an error on the "Triggering
> webpage updates..." (or very similar) message that appears at every CVS
> commit: I received code 403 and a dump of a generic "Access Denied"
> error message.

That's expected.

The recovory of the www repository (and all WebCVS repositories, FWIW)
was left after the Sources ones are done.  I noticed a few commits to
www today, so I assume it was finally recovered.

> I considered it a warning only and assumed translations would be rebuilt
> on a daily basis anyway,

No, I disabled all GNUN+whatsnew cronjobs shortly after the Savannah
filesystem crash, see
http://lists.gnu.org/archive/html/trans-coord-devel/2009-06/msg00000.html

Sorry I didn't announce it here, I thought it was self-explanatory.

> but I probably assumed wrong, as I still can't see the new
> translations online (example: I submitted an Italian translation for
> http://www.gnu.org/software/reliability.html and I updated the
> English page to point to it, but it doesn't show now).

There was an HTML validity error following your change to
licenses/translations.html, which Brett Smith has fixed a few minutes
ago.  As you know, the entire build halts when it encounters an error,
so software/reliability.html is still unprocessed.  (Meanwhile, I ask
you to revert that commit entirely and remove the translation from the
Cookbook, unless you have talked to rms and he has approved hosting
translations of licenses at Savannah.)

But there's more, that's why I'm CC-ing sv-hackers.
Even if there were no build errors, today's job would fail with:

cvs [commit aborted]: could not find desired version 1.5 in 
/web/www/www/philosophy/po/free-sw.af.po,v

This revision was committed on May 28, so I guess we have to manually
recommit all lost commits, right (assuming www was recovered from the
May 27 backup)?  I vaguely remember a plan to mass-restore them from
the directories on the web-server?  How shall we proceed?  Karl, John,
if manual recommits are necessary, do I have green light?

(Sorry if this was already solved/dealth with, I have a few hundred
unread emails so I'm lagging a bit.)




reply via email to

[Prev in Thread] Current Thread [Next in Thread]