I'm reffering to this one. I've tried to use it from two different computers but always get this "Not Found" page. Is it just me or is this an ongoing issues that hasn't been resolved for a while? (I'm contacting you about this because the error message mentions contacting the project administrators and mentions you).
Page pile tool hasn't been working for a while
If you are ending up on the page you linked that sounds like a bug in the URLs generated by pagepile. That is one of Magnus' tools according to https://tools.wmflabs.org/admin/tool/pagepile. His enwiki user page recommends reporting issues at https://bitbucket.org/magnusmanske/
become toolname failing.
Hello, I created a new tool (https://toolsadmin.wikimedia.org/tools/id/holidays-viewer) about 5 hours ago and I'm getting an error when I run
become holidays-viewer. I've tried logging in and logging out several times but I'm still getting an error.
jeropbrenda@tools-sgebastion-07:~$ become holidays-viewer become: no such tool 'holidays-viewer' jeropbrenda@tools-sgebastion-07:~$ id uid=21205(jeropbrenda) gid=500(wikidev) groups=500(wikidev),50380(project-tools),54127(tools.holidays-viewer)
What could be the problem?
Probably related to the issues we have been having with the background process that creates the new tool directories and credentials. This service was restarted 2019-08-05T16:10:23Z which I would hope has made it provision your tool correctly.
Hi, we spoke about cross-wiki gadgets at Wikimania. Krinkle showed me a way to load the whole gadget+dependencies minified. It even supports automatic RTL conversion, using the user language.
// [[File:Krinkle_RTRC.js]] (mw.loader.getState('ext.gadget.rtrc') ? mw.loader.load('ext.gadget.rtrc') : mw.loader.load('https://www.mediawiki.org/w/load.php?modules=ext.gadget.rtrc&lang=' + mw.config.get('wgUserLanguage', 'en')));
The initial check is for loading a local version of the gadget, if such a version exists. Otherwise, it'll load the gadget from mediawiki.org.
Example taken from meta:User:Krinkle/Tools/Real-Time Recent Changes.
Following up on replication drift
Hi Bryan Following on from our discussion a few months ago I was just wondering where we were on the new servers. I see there's been progress on Phabricator but was wondering what the timelines were looking like?
I've just had an instance where one of the zombie categories has been successfully cleared, apparently by someone creating it on enwiki and it then getting deleted. I've got my own test running on another cat but I was wondering where we go from now. Is there going to be a complete ground-zero wipe of the Labs database, or is the idea that we keep the one we have but delete operations will work? I'm in no hurry if it's a question of waiting for ground zero, but if it's the latter then I might as well get on and delete-cycle the remaining zombies on the reports I work on. Cheers. Le Deluge (talk) 12:10, 25 August 2017 (UTC)
The new wiki replica cluster is fully populated and ready for use by tools. Just before Wikimania we created phab:T172704 to track efforts to attract beta testers of the new servers. I'm hoping to post something to labs-announce and our phame blog soon about that, but the current phab task has all the details that should be needed for people to start using the new servers.
Thanks, although I was hoping for a bit more of a phabricator-English translation, from the perspective of an occasional Quarry user. But no matter, I've been able to remove most of the zombies from my two main reports by editing or delete-cycling them, with one exception - but at least my OCD has been largely sated! [[en:Category:Law about religion in the United Kingdom]] is still stuck - it's possible that it's in a bit of the Labs db that's particularly out of sync, as it was one of a number of cats created at the same time that all had problems, but I was able to fix most of them. Any ideas about that one?
There is a blocker for converting Quarry to use the new database servers. The datasets_p database is not currently available on the new servers. There is yet another phab task (actually a whole pile of them) where we are trying to figure out what to do about this.
No worries. As I say, the worst of my itch has now been scratched. I take it that we'll see some speed improvements once Quarry is on the new servers? The query that started working again a few months ago has been timing out for the last six weeks or so, I assumed that was a combination of the current work and kids being on holiday.
We do hope that the new servers will be faster. We also hope to be able to allow longer running queries to the
wikireplica-analytics.eqiad.wmnet service name. That's the one that Quarry will be pointed at. The other service name is intended for tools which need quick responses.
Any idea yet when Quarry moves to the new server?
phab:T173511 is the main blocker for converting Quarry to the new servers. The
datasets_p data is currently only available on the old database servers.
Sidenote: There is a user-preference (default-on) for us to get web-Notifications about any "Email from other user", so it is no longer necessary to send talkpage notes about this. :-)
@AcagastyaI think I described what you need to do fairly completely in phab:T164910#3271720. The advice you have been given in has been mixed in part because using NodeJS to write bots is not completely common and in part because you have not referenced the previous communications. What you want to do is possible. See Amir's offer to work with you in phab:T169338#3396104 for one way you might get help beyond the instructions I have already written and the documentation at wikitech:Help:Tool_Labs/Kubernetes#Kubernetes_continuous_jobs
Hello and sorry
Hello, i would like to thanks you for the change Special:diff/3249645, unfortunaly my mouse didn't click on thanks but on [br]. I'm sorry for my revert : the good version is back now. Happy day.
No worries! I was about to start a chat with you about how to improve my edits to that page when you reverted the revert. :)
There are no older topics