Jon Leighton
Jon Leighton
#14170 probably introduced this problem. CC-ing those involved: @p0deje @ioquatix @aguspe
This is still an issue
@ninadbstack thanks, yes, I want to wait for asset discovery. the `sync` flag (from https://github.com/percy/cli/pull/1490) appears to block on Percy actually capturing the uploaded screenshots, which is not what we...
I agree that this should just be about `NameError`, as the other two can be retried via Active Job. > though maybe still Active Job should still have config of...
I find the `rubyLsp.runTestInTerminal` command useful because: * I can run all tests in the current file by triggering it on my top level `describe`/`context` block. This is much faster...
We upgraded to Postgres 16, and these queries are now running a bit faster (more like ~35ms avg). However, this may also have been due to us reducing job retention...
Thanks for the replies @bensheldon. > Cause I'm not seeing a good index specifically for the queue_name-limited situation. I tried to investigate this a bit. I generated a query without...
> fyi about: > > > Without being super familiar with Good Job's implementation details, my understanding is that this query obtains a lock for the job it is about...
> > Another maintenance thing you might try is running REINDEX TABLE CONCURRENTLY good_jobs; which I've heard could speed things up (let me know cause I could turn that into...
> The other thing we do plan to do is change our job retention to be very low, now that #1550 is merged - we'll retain discarded jobs a bit...