On Thu, Apr 30, 2020 at 2:49 PM Russell Blau <russblau@imapmail.org> wrote:
On Apr 28, 2020, at 8:31 AM, Manuel Arostegui <marostegui@wikimedia.org> wrote:
> This is because we are doing maintenance on the usual host that serves the slow queries, and we placed a different one and we (I) didn't change the query killer, which kills queries after 5 minutes.
> I am going to change it now, to give it some more room (but won't set the 4h query, as we need a balance there now that we are serving with less hosts).

Is there any ETA on when the query killer will get back to normal?


The query killer is now killing queries longer than 1h.
Unfortunately the maintenance will take a few more days, as we are facing some unexpected issues. 
If you want to follow this progress, this is the task we are using: https://phabricator.wikimedia.org/T249188

Sorry for the inconvenience. 
Wikimedia Cloud Services mailing list
Cloud@lists.wikimedia.org (formerly labs-l@lists.wikimedia.org)