Implement real time updates and gradually enable it on production to see if we can handle the extra load.
Description
Details
Subject | Repo | Branch | Lines +/- | |
---|---|---|---|---|
Implement RT updates for completion | mediawiki/extensions/CirrusSearch | master | +707 -83 | |
Implement RT updates for completion | mediawiki/extensions/CirrusSearch | es5 | +1 K -128 |
Status | Subtype | Assigned | Task | ||
---|---|---|---|---|---|
Resolved | debt | T151324 [epic] System level upgrade for cirrus / elasticsearch | |||
Resolved | • Deskana | T154501 [Epic, Q3 Goal] Upgrade search systems to Elasticsearch 5 | |||
Declined | None | T154503 Implement real-time updates to completion suggester index in Elasticsearch 5 |
Event Timeline
It is unclear whether this investigation can be done before upgrading to ES5 (tracked in T154501) right now.
Change 342819 had a related patch set uploaded (by DCausse):
[mediawiki/extensions/CirrusSearch] Implement RT updates for completion
Change 342820 had a related patch set uploaded (by DCausse):
[mediawiki/extensions/CirrusSearch] Implement RT updates for completion
moving back to in progress, the first patch proposed is flawed and most probably won't work due to race conditions and is likely to cause much more issues than it solves.
We should figure out if we want to dedicate the time needed to figure this out. There is an email thread going around about this to determine if/how/when.
We have decided not to continue work on this for now. There's a lot of open questions, and it's very unclear from where we stand right now how much work this would be. It's quite possible that we'd open a can of worms trying to do this due to the complexity and it'd turn into a time sink. The user value is still clear, but it seems it's not in line with the expected effort to implement it.
Change 342820 abandoned by DCausse:
Implement RT updates for completion
Reason:
too complex and costly to be worthwhile
This does not define as stalled ("If a report is waiting for further input (e.g. from its reporter or a third party) and can currently not be acted on"). Hence resetting task status and lowering priority. If this task is out of scope and nobody should ever work on this, then this task should have the "Declined" status instead.
Closing out low/est priority tasks over 6 months old with no activity within last 6 months in order to clean out the backlog of tickets we will not be addressing in the near term. Please feel free to reopen if you think a ticket is important, but bare in mind that given current priorities and resourcing, it is unlikely for the Search team to pick up these tasks for the indefinite future. We hope that the requested changes have either been addressed by or made irrelevant by work the team has done or is doing -- e.g. upgrading Elasticsearch to a newer version will solve various ES-related problems -- or will be subsumed by future work in a more generalized way.
Re-opening tasks and removing from team workboard per IRC feedback given yesterday and discussion with MPham.