The server I copied to is a debian server with the same version of mariadb installed. Could I point that server at the db that I copied and then dump that, restore, and copy it back to wmflabs?
Just have to figure out how to permanently discard the unfinished transactions.
rsync is done.
copying via rsync in process....
I've updated the LocalSettings.php on the site to point to this ticket. If you have cloud access and want to help with the recovery of the site, I've copied the site to /srv/www/mediawiki/public_html/testing/ and it can be accessed here.
(WikiApiary is on the host named pub2 mentioned here.)
Sun, Feb 10
Fri, Feb 8
Also, thanks for pointing out the escape in what is on master. I created the test based on what posted on the support desk, but copied the KEY_FORMAT_REGEX above from master without noticing the added escape.
It isn't me. Original user (see support desk link above) was on 1.32.1.
Wed, Feb 6
Thank you so much, @osorio-juan-microsoft ! (and you gave me enough hint on how to solve similar issues.)
Fri, Feb 1
Wed, Jan 30
Sun, Jan 27
Sat, Jan 26
@osorio-juan-microsoft I have not had time to spend more time on this myself.
Thu, Jan 24
Dupe of T204800
Tue, Jan 22
Note that the confusion came because the informative message wasn't seen until $wgShowExceptionDetails=true; was added.
Jan 16 2019
Jan 13 2019
Jan 12 2019
Jan 11 2019
Could you upload the modified mssql/tables.sql file?
Jan 5 2019
Jan 3 2019
@Catrope: I did not install EventLogging. I don't think that is related to the timeout issue, though.
Jan 2 2019
Dec 31 2018
Handled with T212635
Dec 28 2018
For reference: I love the support desk and don't use discourse. I find the notification system on MW.o to work wonderfully. I can ignore it when I need to and, when I'm ready to get engaged, I find it easy to use the Flow's notifications.
Dec 27 2018
Dec 23 2018
Dec 21 2018
Example of repeated 404s that should only really be one:
I'm going to guess that this is caused by locks on the db when the bees are busy crawling as they were this afternoon.
looks like this was intermittent?
Note that I just mean that that in that particular run it shouldn't make further requests. Right now, it hits a site several times even if the first one got 500.
Dec 20 2018
I apologize for letting this slip for so long.
CSRF errors should be fixed, but they should be logged as well.
We found that we needed to install librsvg2-bin to get rsvg-convert installed and this process working.
Dec 12 2018
Dec 10 2018
As someone else noted on the support desk, there is already an "appendtext" parameter.
Dec 7 2018
Dec 5 2018
I was about to file another bug but rememberd this one, instead.
Nov 28 2018
Ah! I had to specify
$wgNamespacePermissionLockdown[NS_TIP]['edit'] = [ '*' ];
to get it to work.
Nov 6 2018
I reapplied your patch and checked out the version of CREDITS that ships with the extension.
Nov 5 2018
Moar "fun" with quoting:
C:\php-sdk\phpdev\vc14\x64\php-7.0.27-src>rc /nologo /d "THANKS_GUYS=Thanks to <tim>" win32\build\template.rc win32\build\template.rc(59) : error RC2104 : undefined keyword or key name: Thanks
It looks like your patch works. Thank you.
Nov 3 2018
Oct 5 2018
This isn't really a lockdown problem.
The actual api call that parsoid is sending is
I've finally been able to get the error 500 you described. (I was writing out my configuration and found the problem that was keeping me from seeing it.)
Oct 4 2018
Do you have cookie forwarding set up as the documentation for VE suggests?
I'm not sure that this is caused by lockdown, though. I'm trying to duplicate it and I'm running into the same error even without Lockdown.
Oct 2 2018
I get "apierror-visualeditor-docserver-http: HTTP 500" suggesting that
API interaction is broken in this particular namespace.
Sep 30 2018
Sep 29 2018
Thanks, @Tgr. That said, I found this bug in schema v1 (which is the only version available in MW 1.27).
Also, worth pointing out that T148582#4624990 is where I discovered this.
See this commit for initial discussion.