The indexing in connection with lexemes are now slow again. For instance, during creation of L270066, the form L270066-F3 is not available for use.

I am wondering whether @Bstorm could clarify if -m 1Gi is a typo and should have been -m 1G?

Thanks for the explanation and the restart

There might be a bug in webservice migrate: https://wikitech.wikimedia.org/w/index.php?title=Talk%3ANews%2F2020_Kubernetes_cluster_migration&type=revision&diff=1856814&oldid=1856342

I think was due to our tool (import error on lxml on new server) as well as the problem with virtual environment change during migration. It seems to work now. https://tools.wmflabs.org/scholia/

Also tracking this at https://github.com/fnielsen/scholia/issues/1013

Both the entity IDs and the label searches were slow. The search is now fast again, so I can make it "Resolved" (or something else?). I suppose it could have been a temporary lag in elasticsearch.

Is @Multichill able to add 172.16.0.0/12 to https://www.wikidata.org/wiki/MediaWiki:Autoblock_whitelist ?

This seems not to be confined to lexemes but also the case for Q-items.

A possible unrelated issue is Magnus Manske's sourcemd that is also in non-working condition. https://twitter.com/MagnusManske/status/1204000679116849152

My two tools came up nicely yesterday evening/night.

I think the design of Wikidata lexeme was inspired by the work around wordnets and Ontolex. In the wordnet world, a sense is the intermediate step between a lexeme and a concept/synset, see Ontolex specification where the sense is called LexicalSense https://www.w3.org/2016/05/ontolex/

I suppose this issue was inspired by the discussion after Lydia's talk at WikidataCon 2019: https://media.ccc.de/v/wikidatacon2019-2-wikidata_and_languages#t=2788 . I have now gone back to see how I actually handle there situations for Danish.

The query is a bit hard on WDQS. If one execute it twice then the second time can apparently use some caching from the first.

Here a variation on @Yurik's query with count on within-language forms: https://w.wiki/8y8 (current count is obfuscated by Tamil annotation). For instance, 'led' in Ordia shows 9 lexemes from 3 language: https://tools.wmflabs.org/ordia/representation/led

I count over 30 basic lexemes on https://en.wiktionary.org/wiki/for while there may be more when we start to count inflections and derived terms ...

Maybe the priority should just be "high" as few users might be viewing the 3D models?

There was some work on I18N branches in Ordia: https://github.com/fnielsen/ordia/branches It is not yet implemented in Scholia

I have changed the priority to unbreak now. I do not know if that is appropriate. The problem is that it is currently not possible to interact (basically) with 3D models on the Wikimedia sites, - only by downloading and viewing them in e.g. meshlab.

I can confirm this on Ubuntu Firefox 69.0:

It seems that this was the only one.

I have now edited the lexeme back and forth (https://www.wikidata.org/w/index.php?title=Lexeme:L10531&action=history) which switched the language to wd:Q9056.

Thanks, that was apparently not submitted but just shown as preview

There has been a recent namespace change announced at https://lists.wikimedia.org/pipermail/wikidata/2019-August/013350.html but this was with dcatap.

Yes, there is a Q/P mixup. DESCRIBE wd:P20980928 gives 512 results.

Works for me now too. Thanks!

Yes. It is back.

Yes, fixed at my end too.

My browser is Firefox.

The lexeme has these language fields for word stem.

Yes, I experience no error on the main sight now.

The error also appears when the first gloss is reedited

If the page is reloaded by the browser then it is possible to enter the second gloss.

This may have been a temporary error. It works now.

BTW: This affect the ShEx validator.

Quite a number of panels in Scholia has been affected by the change/bug, e.g., https://tools.wmflabs.org/scholia/author/Q20980928 "Number of publications per year", "Number of pages per year" and "Citations by year".

As a tool provider it would (also) be nice to have the data transposed - so to speak: Eg. https://tools.wmflabs.org/toolviews/api/v1/tool/scholia where the returned data is across multiple days.

Further development issues are noted here: https://github.com/fnielsen/scholia/issues/721

This was related to Wikimedia-Hackathon-2019.

Javascript implementation being experiment on here: https://codepen.io/fnielsen/pen/oReLKE

Maybe this is not a bug per se, but intentional? There is not yet a Fon Wikipedia, - it is on Incubator.

Yes, it has been working fine now.

This seems to work fine now.

Another way to generate status code 500 is by, e.g.,

There seems to be another issue with the Wikidata.org/w/api.php?action=wbsearchentities: T222347. In the case of T222333 the result is just empty, while T222347 returns status code 500.

Commands such as these seems to have fixed it:

webservice --backend=kubernetes python shell
cd www/python
python3 /usr/lib/python3/dist-packages/virtualenv.py --python=python3.4 venv3.4
ln -s venv3.4 venv
cd ordia ; pip install -rrequirements.txt

It is a python3.4/python3.5 issue? The Virtualenv I can construct is 3.5. Kubernetes start Python 3.4.2.

Ordia's uwsgi.log looks basically the same as QuickCategories. And with restart I get the same message.

Yes. This may have been a temporary issue.

currentDay gadget for P813 now in troubles. I wonder if the fix/changes related to this issue.

The fix works for me. Thanks!

The issue must related to a change happening on 13 March or a few days before.

There have been slow response from Toolforge services. I use sourcemd and scholia. sourcemd was just slow, and have been slow from time to time. In scholia I have seen error messages like: