https://bugzilla.wikimedia.org/show_bug.cgi?id=69613
--- Comment #15 from Niklas Laxström <[email protected]> --- Given the info provided by Nik above, it looks like gradual slowdown. What worries me is that the slow queries seem to block the MessageUpdateJobs. Increasing the timeout even more might not be a good idea in that light, or it might not make a difference. We could also try to limit the set of potential matches even further in the code. We should be very cautious about deploying the same algorithm to ElasticSearch (as per existing plans to stop using Solr). Nik knows better whether it might cause the same problem there, or whether that setup can make the problem smaller. I do have initial plans on working on translation memory (on my research time), but the ETA for that is more like end of this year. I would like to avoid is to not be able to run TM service at all until that. -- You are receiving this mail because: You are on the CC list for the bug. _______________________________________________ Wikibugs-l mailing list [email protected] https://lists.wikimedia.org/mailman/listinfo/wikibugs-l
