|
2026-01-08 17:10:53
|
<ragesoss>
|
since yesterday, Wiki Education has been seeing a wave of intermittent errors related to OAuth login and token-fetching. any pointers to what might be going on?
|
|
2026-01-08 17:13:36
|
<ragesoss>
|
oh! it's 429 error
|
|
2026-01-08 17:13:49
|
<ragesoss>
|
that's annoying, having seen that before with login and OAuth tokens
|
|
2026-01-08 17:13:58
|
<taavi>
|
ragesoss: what does the HTTP response body say?
|
|
2026-01-08 17:14:47
|
<ragesoss>
|
Request served via cp4044 cp4044, Varnish XID 735497657<br>Upstream caches: cp4044 int<br>Error: 429, Too many requests (f061ab2) at Thu, 08 Jan 2026 17:11:08 GMT<br><details><summary>Sensitive client information</summary>IP address: 2600:[ip]</details>
|
|
2026-01-08 17:25:16
|
<ragesoss>
|
i've created an issue: https://phabricator.wikimedia.org/T414114
|
|
2026-01-08 23:31:53
|
<Dragonfly6-7>
|
I accidentally deleted an article with +1200 edits (I meant to only delete a single one), and when I try to restore it, I keep getting an error message.
|
|
2026-01-08 23:31:59
|
<Dragonfly6-7>
|
To avoid creating high replication lag, this transaction was aborted because the write duration (3.2673449516296) exceeded the 3 second limit. If you are changing many items at once, try doing multiple smaller operations instead.
|
|
2026-01-08 23:31:59
|
<Dragonfly6-7>
|
[7cff12f6-3c48-4794-a63a-76d06ef36647] 2026-01-08 23:30:20: Fatal exception of type "Wikimedia\Rdbms\DBTransactionSizeError"
|
|
2026-01-08 23:32:01
|
<Dragonfly6-7>
|
help help?
|
|
2026-01-08 23:36:11
|
<andre>
|
what's the question? "try doing multiple smaller operations instead"
|
|
2026-01-08 23:38:45
|
<Dragonfly6-7>
|
so is that "restore 100 diffs at a time" ?
|
|
2026-01-08 23:39:04
|
<Dragonfly6-7>
|
or...
|
|
2026-01-08 23:39:16
|
<Dragonfly6-7>
|
I'm not trying to change them
|
|
2026-01-08 23:39:54
|
<A_smart_kitten>
|
seems like that might be T239095
|
|
2026-01-08 23:39:55
|
<stashbot>
|
T239095: Undeleting pages with many revision results in DB exceptions - https://phabricator.wikimedia.org/T239095
|
|
2026-01-08 23:40:09
|
<zabe>
|
yeah undeletion still is not happening in batches iirc
|
|
2026-01-08 23:40:48
|
<zabe>
|
so you have to do the batching manually
|
|
2026-01-08 23:41:11
|
<Dragonfly6-7>
|
.... so, yes, select 20 revisions at a time and restore them?
|
|
2026-01-08 23:41:37
|
<A_smart_kitten>
|
disclaimer that i'm not an admin on a wiki... but i think you might be able to try with a higher number than that (hopefully)
|
|
2026-01-08 23:42:12
|
<A_smart_kitten>
|
maybe 500 revisions at a time might work? bit of a guess though...
|
|
2026-01-08 23:42:12
|
<zabe>
|
yeah, you could probably do like 200 or 300 or something in that area I would say
|
|
2026-01-08 23:42:18
|
<A_smart_kitten>
|
(i defer to zabe)
|
|
2026-01-08 23:43:25
|
<Dragonfly6-7>
|
any solution more efficient than clicking +1200 boxes by hand?
|
|
2026-01-08 23:43:44
|
<zabe>
|
You can try higher, you cannot really do anything wrong here, it would just end in another timeout if you try too many
|
|
2026-01-08 23:44:15
|
<zabe>
|
I would bet someone wrote a userscript for this issue
|
|
2026-01-08 23:45:05
|
<Dragonfly6-7>
|
you wouldn't have any idea what that script is or where to find it, would you
|
|
2026-01-08 23:47:03
|
<Reedy>
|
Can't you click, press shift, click and do a range?
|
|
2026-01-08 23:48:25
|
<Dragonfly6-7>
|
Thank you.
|
|
2026-01-08 23:49:21
|
<zabe>
|
TIL
|
|
2026-01-08 23:59:01
|
<Dragonfly6-7>
|
if the error message had said "If you are restoring many diffs at once, try etc", I would have known what to do.
|