great - graphs look even better today (apologies for the 2+ week blip) -- so i think this is resolved

In T431493#12189339, @Ottomata wrote:

Hi, just sharing something I learned in T423583: mediawiki.page_change.v1 event - Add revision revert details:

revisions are only tagged as reverted if they were reverted within $wgRevertedTagMaxDepth (15, I think?) revisions of the reverted-to revision.

Perhaps for a constructive edits metric it practically doesn't matter, but it does mean that in cases where a revert to a revision happens within 48 hours but where there are more than 15 revisions after that reverted-to revision, the intermediate revisions will not be tagged as reverted.

Also ran:

SELECT
  dt,
  meta.stream,
  mediawiki.database,
  page.namespace_id,
  page.revision_id
FROM event.product_metrics_web_base
WHERE year = 2026 AND month = 8 AND day = 5
  AND page.revision_id IS NOT NULL
  AND page.namespace_id IS NOT NULL
LIMIT 20;

which returned:

dt				stream			database	namespace_id	revision_id
2026-08-05T14:26:33Z	product_metrics.web_base	wikidatawiki	0		2526755317
2026-08-05T14:23:42Z	product_metrics.web_base	nlwiki		0		71701110
2026-08-05T14:24:19Z	product_metrics.web_base	ptwikisource	106		557779
2026-08-05T14:19:29Z	product_metrics.web_base	commonswiki	6		1256023382
2026-08-05T14:26:23Z	product_metrics.web_base	wikidatawiki	0		2526762295
2026-08-05T14:21:06Z	product_metrics.web_base	frwikisource	104		15959818
2026-08-05T14:20:07Z	product_metrics.web_base	dewiki		5		269341283
2026-08-05T14:23:42Z	product_metrics.web_base	nlwiki		0		71701156
2026-08-05T14:26:32Z	product_metrics.web_base	wikidatawiki	0		2526756951
2026-08-05T14:26:08Z	product_metrics.web_base	viwiki		0		75394259
2026-08-05T14:19:39Z	product_metrics.web_base	commonswiki	6		1256017524
2026-08-05T14:19:47Z	product_metrics.web_base	commonswiki	6		1256011713
2026-08-05T14:21:11Z	product_metrics.web_base	frwiktionary	0		39773453
2026-08-05T14:26:19Z	product_metrics.web_base	wikidatawiki	0		2526764209
2026-08-05T14:19:48Z	product_metrics.web_base	commonswiki	6		1256011127
2026-08-05T14:20:59Z	product_metrics.web_base	fonwiki		0		38811
2026-08-05T14:26:33Z	product_metrics.web_base	wikidatawiki	0		2526755305
2026-08-05T14:26:28Z	product_metrics.web_base	wikidatawiki	0		2526760062
2026-08-05T14:19:33Z	product_metrics.web_base	commonswiki	6		1256021335
2026-08-05T14:19:27Z	product_metrics.web_base	commonswiki	6		1256025090

So confirming that data collection is happening with the contextual attributes specified in the ticket description.

and it appears that evengate validation errors related to the mediawiki.product_metrics.wikifunctions_ui stream are dropping steadily tho slowly 😌

ok - i think we're finally getting somewhere - https://grafana-rw.wikimedia.org/goto/dfuar4s42ek8wc?orgId=default shows events finally showing up after the last backport fix to bump the wikilambda/ui_actions schema to reference the most recent web base schema 🎉

So I ran the follow queries on a stat machine:

@RLazarus graciously deployed the puppet patch at EOD today and when we did a test run of the maintenance script, realized there was an error (s/o to @Scott_French for the fix) so I quickly (possibly too boldly) pushed/self-merged https://gerrit.wikimedia.org/r/c/mediawiki/extensions/WikimediaEvents/+/1320295 and backported https://gerrit.wikimedia.org/r/c/mediawiki/extensions/WikimediaEvents/+/1320296

The edit_survived instrument is functionally complete in code and the maintenance script is scheduled for deployment on 8/4/26 (I will try to get it deployed earlier if possible)

thanks @mpopov - do we have an idea of how long this instrument should run? and traffic constraints?

Per discussion with the team, there was a consensus that the python and jobs metrics are not valuable for what we're looking in terms for GB health metrics. It'd be great if there were some metrics geared toward database queries, etc (we can inquire with GrowthBook). So for now we are treating this ticket as an investigation spike and closing for now

What GrowthBook's OTel instrumentation actually emits

In T431493#12120652, @DLynch wrote:

The main thing that needs to be done with said patch is tuning it to work out what the right interval to run it at is.

Thanks @BTullis for your corrections! Super helpful - definitely got some things wrong in the proposal - my bad. Let me sort out what assumptions you debunked first, then propose next steps.

These are all things I tried to learn about by reading the current code and config and full disclosure, by conferring with Claude (in the sandbox!):

In T415370#12100764, @SBisson wrote:

One of the CX app is using mw.eventLog.submit(...), does it need to be migrated as well?

for WebUIScroll instrument - mediawiki.web_ui_scroll_migrated should maybe be deleted - added to ACs

tysm! works great

hi @Sfaci - quick Q - does

I'm probably forgetting the context for this ticket - are we sure we don't want to support custom schema ids? it seems like if the base schemas for whatever reason don't capture enough info, that we would still want to enable product teams to use custom schemas for edge cases?

as confirmed in Slack, MP java lib isn't being used in production and we haven't cut a release in a long time - i went ahead and merged the MR anyway so this ticket can be closed out -- we'll probably remove the Java lib and related infrastructure as part of some future cleanup
cc @KReid-WMF

just an fyi - we removed Saved Groups functionality from the TK backend for now in https://gitlab.wikimedia.org/repos/data-engineering/test-kitchen/-/merge_requests/361 << some form of this will probably need to be reverted once we have a path forward to dealing with saved groups