Change 427656 had a related patch set uploaded (by Esanders; owner: Esanders):
[mediawiki/extensions/VisualEditor@master] Sortable table header: Account for null entries in table matrix

https://www.wikidata.org/wiki/Wikidata:Project_chat#Phoenician_language: according to the IANA lang tag list, the default script for this language is "Phnx" Phoenician but the example there is given in Latin transcription so shoul de denoted like "gadir"@phn-Latn.

• To capture this, I believe you should register a lang tag phn-Latn with name eg "Phoenician (Latin transcription)".
• The tag phn should be used only if you install Phoenician script (alphabet) in the language tools; and somehow direct the user to select "Phoenician (Latin transcription)" if

In the meantime, I’ll start preparing for this, because the internal challenges of implementing such a counting mode are independent of how it’s modeled as a constraint statement. (As far as I can tell, this pretty much kills the current getSnakGroup abstraction, since now the composition of the “snak group” varies…)

Thanks both for comments.

Change 427655 had a related patch set uploaded (by Addshore; owner: Addshore):
[mediawiki/extensions/Wikibase@master] Factor out NonExistingEntityIdHtmlFormatter

I contacted the WikiProject about this: https://www.wikidata.org/wiki/Wikidata_talk:WikiProject_property_constraints

Or we could upgrade the Ganeti cluster to stretch? It provides qemu 2.8 out of the box.

Completed auto-reimage of hosts:

['wdqs2003.codfw.wmnet']

Change 427654 had a related patch set uploaded (by Filippo Giunchedi; owner: Filippo Giunchedi):
[operations/puppet@production] base: alert on SMART health failure

Minimal test case:

{| class="wikitable sortable"
!Foo
|-
|-
|Bar
|-
|}

Gonna ping @JAllemandou on this one ^ :)

After a quick talk to @Lea_WMDE we came to the conclusion, that this should best be treated as custom error because it should be recoverable like for example the duplicated title error. Therefore the validation will happen early in the title validation steps.

Looking at the content of the emails @fgiunchedi forwarded to me, the last offense right when the next cron kicks in after the restart and keeps going for the rest of the hour on the hourly "all" log, and for the rest of the day on the daily "all" log.

Mentioned in SAL (#wikimedia-operations) [2018-04-19T13:14:55Z] <reedy@tin> Synchronized wmf-config/InitialiseSettings.php: T192427 T189277 (duration: 01m 17s)

Mentioned in SAL (#wikimedia-operations) [2018-04-19T13:14:55Z] <reedy@tin> Synchronized wmf-config/InitialiseSettings.php: T192427 T189277 (duration: 01m 17s)

Change 427653 had a related patch set uploaded (by Xqt; owner: Xqt):
[pywikibot/core@master] [test] use python 3.4.1 instead of 3.4.0 for tests

Change 418070 merged by jenkins-bot:
[operations/mediawiki-config@master] Change NS aliases on ruwiki

Change 427390 merged by jenkins-bot:
[operations/mediawiki-config@master] Enable edit patrol in hiwikiversity

In T188620#4142566, @deryckchan wrote:

@Roy17 has done a few tests in his userspace and saw that most websites failed to generate, including ones like BBC which usually work on en.wp. Have we missed some configurations?

Current status: stuff moved around in /var/www/, /etc/apache2/sites-available/*.conf updated, /var/www/w/LocalSettings.php edited in Emacs but not yet saved (in lucas-T189864 tmux session). Currently waiting to receive the password so that I can run sudo systemctl reload apache2 and then save LocalSettings.php. And then hope that the new config works…

The root cause is broken table syntax (defining the row twice). Ideally VE should be able to handle that.

Screenshot taken with the latest patch:

Triggering the right hooks here seems to be quite easy. But now I come across the question how to handle the error.

All this code is compatible with MediaWiki 1.29, right?

Script wmf-auto-reimage was launched by gehel on neodymium.eqiad.wmnet for hosts:

['wdqs2003.codfw.wmnet']

The log can be found in /var/log/wmf-auto-reimage/201804191254_gehel_24890.log.

In T187909#4142562, @kaldari wrote:

I imagine they will at least need to create a "cite web" template in order for Citoid to be practically useful (besides their existing cite book and cite journal). @Elitre, maybe you could ping Guliolopez and see if they would be willing to create that. Also, pinging via Phabricator: @Guliolopez.

I've created an initial Citoid mapping here: https://ga.wikipedia.org/wiki/MediaWiki:Citoid-template-type-map.json.

Membership Applications:

Change 427651 had a related patch set uploaded (by Xqt; owner: Xqt):
[pywikibot/core@master] [bugfix] Re-enable TestCategoryFromWikibase tests

@Roy17 has done a few tests in his userspace and saw that most websites failed to generate, including ones like BBC which usually work on en.wp. Have we missed some configurations?

@Lydia_Pintscher, why should an error message that can clearly be identified as an error message say "an error has occurred"? I just saw this in the patch. I think this is not only redundant but counter-productive.

I imagine they will at least need to create a "cite web" template in order for Citoid to be practically useful (besides their existing cite book and cite journal). @Elitre, maybe you could ping Guliolopez and see if they would be willing to create that. Also, pinging via Phabricator: @Guliolopez.

In T191996#4139205, @Marostegui wrote:

For the record, the irq for eno1 is balanced across CPUs, so I don't think it is the bottleneck here:

root@db1114:/srv/tmp# for i in `cat /proc/interrupts | grep eno1 | awk -F ":" '{print $1}'`; do cat /proc/irq/$i/smp_affinity_list; done
0,2,4,6,8,10,12,14
0,2,4,6,8,10,12,14
0,2,4,6,8,10,12,14
0,2,4,6,8,10,12,14
0,2,4,6,8,10,12,14

@mmodell for sure, I was reading the log and I wonder why architecture changed from all to any?

Change 427648 had a related patch set uploaded (by Tonina Zhelyazkova; owner: Tonina Zhelyazkova):
[mediawiki/extensions/AdvancedSearch@master] Mock deepcategory on test wikis

As specified above, I should be the initial bureaucrat.

Is the Apr 18 occurence on that screenshot after the change was deployed?

Change 427625 merged by jenkins-bot:
[integration/config@master] Bump quibble-stretch to 0.0.8-2

Change 427394 had a related patch set uploaded (by Niedzielski; owner: Stephen Niedzielski):
[mediawiki/extensions/MobileFrontend@master] Hygiene: trim unused PageListItem template tag

Somewhat frequent but not a lot, I don't have exact numbers tho

Please check your web browser's "Developer Tools" in the "network" console for errors when loading that page.
https://www.mediawiki.org/wiki/Help:Locating_broken_scripts is about scripts, but it explains how to use the developer tools.

@Mvolz, @Elitre: It looks like ga.wikipedia.org does have at least one citation template in use: https://ga.wikipedia.org/wiki/Teimpl%C3%A9ad:Cite_journal, which is also aliased to "Template:Cite book". I'm very skeptical of importing Template:Citation to wikis that don't have technically-proficient editors. That template has 96 other template dependencies and a dozen module dependencies, all of which are very complicated and have to be maintained by someone on that wiki. If they aren't maintained, they will gradually produce errors as MediaWiki's citation handling, module capabilities, and Wikitext parsing evolve. Wikitext parsing is especially likely to evolve over the next few years due to the Platform Evolution project and parser unification. Localizing those 100+ templates and modules is also not a trivial task (especially since many of them are explicitly designed to support English grammar). Currently ga.wikipedia has one simple, functional, easy to maintain and update citation template, which probably makes sense for them.

Sideffect: That list might get too long and become unmaintainable?

With cache=none being set in all cluster for unrelated reasons, this is now unblocked. In the meantime jessie-backports has upgrade to 2.8. Fortunately the changelog[1] does not have any worrying items in it. The upgrade will require a round of VM reboots, but otherwise looks ok. I 'll empty an eqiad ganeti host, upgrade to 2.8 and move a few VMs to it for testing.

You can comment who should get a bureaucrat account and under what
username. We then run a script to create them.

As it seems to have something to do with logging in and sessions perhaps @Anomie might be able to spot something in the logs.

All VMs have been migrated to using cache=none. I 'll resolve this successfully, hopefully we will not meet this issue again

I see the wiki is now available. What is the process for having accounts on that wiki and what other steps are needed? Thanks

Change 427646 had a related patch set uploaded (by Jayprakash12345; owner: Jayprakash12345):
[mediawiki/extensions/Translate@master] Convert Special:ExportTranslations to OOUI

Change 427407 abandoned by Cparle:
Add haswbstatement feature

Suggestion after discussion with Andreas. Note that the port for HTTPS has been added.

Yes of course ! I would be happy to work on that also its really an easy small task to do !. To follow up with deployment date, will be there an announcement on ai-request@lists.wikimedia.org on the exact time is known?

Going through several old e-mails I found the final specs for the old Windows server (note that these are higher than needed as per our investigation of why the wiki was going down).

It's also still in puppet, BTW:

@Ramesh2744 Are your images or thumbnails perhaps not on the exact same host as the rest of your website ? Or use img_auth.php ? Because this problem often demonstrates when there are problems with CORS.

@Krinkle You've added the MW-1.31-release project, so does that mean this should also be back-ported to 1.31?

Change 427644 had a related patch set uploaded (by Santhosh; owner: Santhosh):
[mediawiki/services/cxserver@master] Do not gather categories under transclusions