This is somewhat high priority since it seems to be somewhat broken on the change to the new server; for now, I've simply switched to Cosmos, but a better solution is needed.

This will probably be helped a lot by T47.

Fixed with the move to the new server.

This is now done; pending moving all of the databases over.

This has now been done.

The server is now set up using Proxmox, configuration is ongoing, and it is expected by EOW that we can have this live.

This will most likely be done by virtualisation with Proxmox, running on a bare metal server.

We had another round of spam attacks recently, and as a result, there were several measures taken.

This has now been solved by normalising the cookie values that reach the backend.

This is slightly harder, due to the fact that S3 doesn't allow much scalability in terms of adding headers; notwithstanding this, we can probably do this on Fastly's end? Maybe.

This is blocked due to memory issues with PHP 8; tracked upstream in MediaWiki, here.

Slightly behind on this due to some unforeseen circumstances, but 1.40 is now live on testingoa.telepedia.net.

The provisional date for this is Tuesday, 24 Oct, at 5pm BST.

Update on this, we're not caching through Ezoic anymore, but we could still benefit from this; we still don't need a cache server, because we're caching through Fastly, but we can probably use HAProxy as a load balancer—it offers better configuration than Hetzner.

Still needed, but will need to be sent on S3's end.

This is now solved, hopefully, by adding a query parameter to the url in CF.

Ok. I'll confirm this.

Tracking new cache updates using CloudFlare inside of CloudFlare to see if the caching and speed gets better over the coming weeks.

@CJWorldGame32125 pinging this, if I do not recieve a response by the end of the weekend, I will assume the issue with notifications is fixed. I see the comment image issue now and will address.

Enabling $wgNativeImageLazyLoading to see if this makes a difference.

I am unfortunately unable to replicate this. I can clear all of your Echo notifications to see if this helps. Please confirm if you are okay with this.

And yes, the Notifications issue is non existent in other wikis.

The upvote/downvote issue occurs in every page.

Stalled waiting replies from reporter, @CJWorldGame32125.

These are two separate issues, and need two separate tickets.

Not entirely sure what the issue was, here, but it seemed to be fixed by restarting Apache. Some folks over on WMF Phabricator are saying that it could correlate to a CURL issue, which is tracked on Ubuntu forums, here.

Error seems to be:

[FileOperation] S3FileBackend: found backend with S3 buckets: static.telepedia.net/testingoawiki, static.telepedia.net/testingoawiki/thumb, static.telepedia.net/testingoawiki/deleted, static.telepedia.net/testingoawiki/temp.
[objectcache] RedisBagOStuff debug: get(WANCache:global:filerepo-file:testingoawiki:66667aed064b92f5a6af29684c1efc23aa395c66|#|v) on 127.0.0.1:6379: success
[objectcache] RedisBagOStuff debug: get(WANCache:global:filerepo-file-redirect:testingoawiki:fba2115d27b4a24592ec186c646ddc8e|#|v) on 127.0.0.1:6379: success
[objectcache] RedisBagOStuff debug: get(WANCache:global:filerepo-Metadata:28ff4c02c0553abe5a5c8a67d719fc31047a7116|#|v) on 127.0.0.1:6379: success
ForeignAPIRepo: HTTP GET: https://commons.wikimedia.org/w/api.php?titles=File%3ARefiner%C3%ADa_CEPSA_Gibraltar-San_Roque.jpg&iiprop=timestamp%7Cuser%7Ccomment%7Curl%7Csize%7Csha1%7Cmetadata%7Cmime%7Cmediatype%7Cextmetadata&prop=imageinfo&iimetadataversion=2&iiextmetadatamultilang=1&format=json&action=query&redirects=true&uselang=en
[http] * Error fetching URL: SSL: no alternative certificate subject name matches target host name 'commons.wikimedia.org'
* There was a problem during the HTTP request: 0 Error

Perfect. Thank you.

The list of interwiki prefixes on Meta is intentionally left short due to the fact that these are global interwiki prefixes. I've added Miraheze, Shoutwiki, Wikimedia Commons, Wikipedia, and Fandom to the list globally. You _should_ be able to edit locally ones you need as well if there are any missing.

The prototype does not include any caching, but here are the thoughts on the process:

• Cache the announcements in Redis (or whatever cache), for the maximum period of time; this means that everytime someone visits S:Announcements, it isn't calling the DB again.
• Use the job queue to send the Echo notification; also provide variable $wgAnnouncementUseJQ which will be useful if wikis have disabled the JQ and run it via cron; we don't want notifications being stuck and all ran at night.
• Add timestamps for the announcement and when it is expired.

Now fixed. It seems as though a wiki being deleted and being made invisible depends on the "missing wiki" file. This is not idea, but it works for now.