Page MenuHomeMiraheze
Create Task
Maniphest T5976

Add the EventStreams service
Closed, DeclinedPublic
Actions

Description

It has been discussed with various people that a tool to better monitor and combat cross-wiki vandalism is definitely needed on Miraheze. I haven't seen any alternatives that could work for us other than SWViewer, and that requires EventStreams.

Related Objects

Event Timeline

Reception123 triaged this task as Normal priority.Jul 27 2020, 11:13
Reception123 created this task.
Herald added a subscriber: RhinosF1. · View Herald TranscriptJul 27 2020, 11:13
Reception123 added a project: Site Reliability Engineering.Jul 27 2020, 11:14
Reception123 moved this task from Radar to New Features/Services on the Site Reliability Engineering board.
Reception123 added a subscriber: Paladox.
MacFan4000 subscribed.Sep 20 2020, 16:21

This would be useful for certain bots as well. For example: SignBot requires this. (https://phab.bots.miraheze.wiki/T86)

Reception123 added a comment.Sep 20 2020, 16:29

This does seem to require a lot of RAM, so not sure we can do this at this stage.

Dmehus added subscribers: Owen, Dmehus.Sep 20 2020, 18:59

Ah, so that might be the main issue here, the RAM usage? Perhaps a discussion with @Owen if we have room in the budget to increase the RAM allocated to one or more of our servers? If the budget is too tight, then I guess this is dependent on a rather significant donation from another benefactor?

Sario528 subscribed.Oct 27 2020, 11:34
Unknown Object (User) changed the task status from Open to Stalled.Dec 24 2020, 04:25
Unknown Object (User) lowered the priority of this task from Normal to Low.
Herald added a subscriber: Unknown Object (User). · View Herald TranscriptDec 24 2020, 04:25
Dmehus awarded a token.Dec 24 2020, 17:38
John edited projects, added Infrastructure (SRE); removed Site Reliability Engineering.Jan 30 2021, 11:45
John moved this task from Incoming to Long Term on the Infrastructure (SRE) board.Jan 30 2021, 12:17
John changed the task status from Stalled to Open.Mar 6 2021, 19:52
Redmin subscribed.Mar 7 2021, 05:39
John claimed this task.Mar 7 2021, 17:24
John subscribed.

Requires Kafka -> claiming as a investigatory exercise

John closed this task as Declined.Mar 7 2021, 17:56

Going to mark as declined for now - a quick look into this suggests we'd need around 16GB to be able to process just our RC traffic alone, ignoring jobs as well. The actual gain in deploying this service as well is very small, so I don't feel like I can justify allocating the remaining RAM on a single server to a Kafka deployment.

John mentioned this in T6006: Migrate to a Kafka Job Queue.Mar 7 2021, 17:57