#metabrainz

/

      • iliekcomputers says hi from Bangkok 😊
      • 2018-01-04 00406, 2018

      • ruaok waves at iliekcomputers
      • 2018-01-04 00426, 2018

      • ruaok
        LB is down, but likely not our fault. the DB is freaking out. but we have a poor error message up there.
      • 2018-01-04 00421, 2018

      • iliekcomputers
        Yikes.
      • 2018-01-04 00446, 2018

      • iliekcomputers
        ruaok: open a ticket and I will take a look today?
      • 2018-01-04 00404, 2018

      • ruaok
        ok.
      • 2018-01-04 00421, 2018

      • ruaok
        first I gotta try and keep the sites up. everything is tipping over.
      • 2018-01-04 00439, 2018

      • iliekcomputers can't wait for Sentry to start working with LB tbh
      • 2018-01-04 00405, 2018

      • vishalchoudhary[ joined the channel
      • 2018-01-04 00404, 2018

      • djwhitey joined the channel
      • 2018-01-04 00457, 2018

      • djwhitey_ has quit
      • 2018-01-04 00417, 2018

      • xps2 has quit
      • 2018-01-04 00450, 2018

      • xps2 joined the channel
      • 2018-01-04 00426, 2018

      • samj1912
        ruaok: sorry, just woke up
      • 2018-01-04 00437, 2018

      • ruaok
        k. postgres is freaking out again.
      • 2018-01-04 00450, 2018

      • ruaok
        I stopped sir & solr, no change (not surprising)
      • 2018-01-04 00405, 2018

      • ruaok
        running a vacuum analyze now, no improvement so far.
      • 2018-01-04 00414, 2018

      • ruaok
        do you remember what else helped in the past?
      • 2018-01-04 00430, 2018

      • samj1912
        ruaok, vaccum analyze is what you want
      • 2018-01-04 00437, 2018

      • samj1912
        It should work once it's done
      • 2018-01-04 00406, 2018

      • ruaok
        do we not have the auto-vacuum thingy running?
      • 2018-01-04 00416, 2018

      • samj1912
        PG does. I think the problem is once sir does a re index with recordings, it somehow messes up PG
      • 2018-01-04 00440, 2018

      • samj1912
        In the past we have had such problems only when sir did a complete reindex
      • 2018-01-04 00457, 2018

      • ruaok
        and we just did, no?
      • 2018-01-04 00402, 2018

      • samj1912
        Yup
      • 2018-01-04 00414, 2018

      • ruaok
        from now on, when you re-index. run a vacuum anayze!
      • 2018-01-04 00423, 2018

      • ruaok gives samj1912 a light spanking
      • 2018-01-04 00423, 2018

      • samj1912
        But this time I did it against Master (bowie)
      • 2018-01-04 00444, 2018

      • samj1912
        Last time it was against queen
      • 2018-01-04 00445, 2018

      • ruaok
        clearly it matters not.
      • 2018-01-04 00404, 2018

      • samj1912
        We thought it was hot standby feedback that was causing the problem
      • 2018-01-04 00416, 2018

      • samj1912
        From today, it seems not
      • 2018-01-04 00429, 2018

      • samj1912
        Because PG should technically be doing this on its own.
      • 2018-01-04 00435, 2018

      • samj1912
        Weird that it's not
      • 2018-01-04 00454, 2018

      • samj1912
        I'll look into it with bitmap
      • 2018-01-04 00439, 2018

      • yvanzo
        Is it possible to read docker logs from another place than bowie?
      • 2018-01-04 00450, 2018

      • yvanzo
        docker logs --since takes forever
      • 2018-01-04 00436, 2018

      • samj1912
        yvanzo: redirect to stdout and then do a head?
      • 2018-01-04 00453, 2018

      • samj1912
      • 2018-01-04 00430, 2018

      • samj1912
        bitmap should be better informed about the value though
      • 2018-01-04 00443, 2018

      • yvanzo
        samj1912: I used --tail
      • 2018-01-04 00427, 2018

      • yvanzo
        (from the beginning)
      • 2018-01-04 00400, 2018

      • samj1912
        Worked?
      • 2018-01-04 00414, 2018

      • yvanzo
        Yes and no, it took several minutes and returned logs with unmatched timestamps.
      • 2018-01-04 00410, 2018

      • samj1912
        I generally do docker logs --since xyz 2&>1 | head -n 50
      • 2018-01-04 00430, 2018

      • Anirudh joined the channel
      • 2018-01-04 00434, 2018

      • xps2 has quit
      • 2018-01-04 00459, 2018

      • Anirudh
        Okay, so hey!
      • 2018-01-04 00400, 2018

      • yvanzo
        Ok, sorry, it is tail, not head :/
      • 2018-01-04 00413, 2018

      • Anirudh
        I'm Anirudh, and I'm participating in GCI
      • 2018-01-04 00442, 2018

      • Anirudh
        And I don't know what's going on, would someone just update me??
      • 2018-01-04 00456, 2018

      • yvanzo
        Welcome! Do you have any question?
      • 2018-01-04 00417, 2018

      • Anirudh
        No, I just was messing around when I landed here.... So....
      • 2018-01-04 00459, 2018

      • Anirudh
        I was actually thinking which task I should take up next
      • 2018-01-04 00400, 2018

      • samj1912
        ruaok: `1 S lxd 20934 20929 0 80 0 - 8642073 - 2017 ? 00:01:28 postgres: autovacuum launcher process ` autovaccum is on and running since 2017
      • 2018-01-04 00423, 2018

      • samj1912
        we will have to investigate why its not doing its job :\
      • 2018-01-04 00430, 2018

      • Anirudh has left the channel
      • 2018-01-04 00453, 2018

      • samj1912
        ugh dammnit, `hot_standby_feedback` is on on bowie :\
      • 2018-01-04 00457, 2018

      • Vini joined the channel
      • 2018-01-04 00408, 2018

      • samj1912
        I thought bitmap turned it off
      • 2018-01-04 00423, 2018

      • Major_Lurker
        Hi guys I take it the 502 errors are related to the above?
      • 2018-01-04 00441, 2018

      • samj1912
        Major_Lurker: yes
      • 2018-01-04 00453, 2018

      • Major_Lurker
        thought so thanks.... good luck :)
      • 2018-01-04 00405, 2018

      • rahul_india joined the channel
      • 2018-01-04 00426, 2018

      • samj1912
        its on bowie so it shouldn't matter but still
      • 2018-01-04 00433, 2018

      • samj1912
        ill ask bitmap to remove it
      • 2018-01-04 00454, 2018

      • rahul_india
        hello, I'm unable to access https://musicbrainz.org/ its showing error 502 "serwer is being naughty". what should i do?
      • 2018-01-04 00407, 2018

      • rahul_india
        server*
      • 2018-01-04 00431, 2018

      • Major_Lurker
        rahul_india, just wait they are working on it
      • 2018-01-04 00452, 2018

      • rahul_india
        okay
      • 2018-01-04 00430, 2018

      • rahul_india has quit
      • 2018-01-04 00421, 2018

      • samj1912
        ruaok: is the vaccum still running?
      • 2018-01-04 00453, 2018

      • samj1912
        *vacuum
      • 2018-01-04 00430, 2018

      • ruaok
        it finished.
      • 2018-01-04 00442, 2018

      • ruaok
        sigh
      • 2018-01-04 00459, 2018

      • samj1912
        just now?
      • 2018-01-04 00418, 2018

      • ruaok
        dunno. I went back to napping, since I've got the flu
      • 2018-01-04 00442, 2018

      • samj1912
        I dont think it helped
      • 2018-01-04 00416, 2018

      • ruaok
        sure doesn't look that way.
      • 2018-01-04 00436, 2018

      • ruaok
        but, I have no ideas on what else it could be.
      • 2018-01-04 00440, 2018

      • ruaok
        I can't think straight.
      • 2018-01-04 00405, 2018

      • samj1912
        should we restart postgres?
      • 2018-01-04 00439, 2018

      • Vini has quit
      • 2018-01-04 00408, 2018

      • ruaok
      • 2018-01-04 00427, 2018

      • ruaok
        do you know how to do that?
      • 2018-01-04 00432, 2018

      • samj1912
        nope
      • 2018-01-04 00411, 2018

      • samj1912
        just docker restart wont do?
      • 2018-01-04 00457, 2018

      • ruaok
        I dont know. nor do we want to corrupt the DB
      • 2018-01-04 00437, 2018

      • samj1912
        ruaok: did you run vacuum as postgres or musicbrainz?
      • 2018-01-04 00448, 2018

      • ruaok
        musicbrainz
      • 2018-01-04 00459, 2018

      • samj1912
        running it again seems to timeout
      • 2018-01-04 00421, 2018

      • samj1912
      • 2018-01-04 00440, 2018

      • samj1912
        should I redo it with set_statement_timeout 0?
      • 2018-01-04 00443, 2018

      • samj1912
        yeah
      • 2018-01-04 00459, 2018

      • ruaok
        it won't help to do it again.
      • 2018-01-04 00430, 2018

      • ruaok
        if it didn't fix anything, running again won't improve it.
      • 2018-01-04 00425, 2018

      • samj1912
        postgres doesn't have access to musicbrainz_db right?
      • 2018-01-04 00444, 2018

      • ruaok
        it should, why?
      • 2018-01-04 00444, 2018

      • ruaok
        we could try re-starting all the -web and -ws containers
      • 2018-01-04 00448, 2018

      • ruaok
        see if that improves things.
      • 2018-01-04 00457, 2018

      • samj1912
        okay
      • 2018-01-04 00450, 2018

      • yvanzo
        Can we stop sentry btw?
      • 2018-01-04 00439, 2018

      • yvanzo
        It is ~10% of queries.
      • 2018-01-04 00424, 2018

      • ruaok
        sure, we've done that in the past.
      • 2018-01-04 00430, 2018

      • rvedotrc has quit
      • 2018-01-04 00415, 2018

      • Vini joined the channel
      • 2018-01-04 00404, 2018

      • xps2 joined the channel
      • 2018-01-04 00434, 2018

      • rvedotrc joined the channel
      • 2018-01-04 00451, 2018

      • yvanzo
        done
      • 2018-01-04 00403, 2018

      • samj1912
      • 2018-01-04 00414, 2018

      • samj1912
        still running
      • 2018-01-04 00421, 2018

      • samj1912
        last time I think it took 2 hrs when bitmap ran it
      • 2018-01-04 00414, 2018

      • Dragonzeron has quit
      • 2018-01-04 00424, 2018

      • Vini has quit
      • 2018-01-04 00448, 2018

      • samj1912
        seems to be getting better?
      • 2018-01-04 00400, 2018

      • ruaok
        no
      • 2018-01-04 00430, 2018

      • ruaok
        maybe.
      • 2018-01-04 00432, 2018

      • samj1912
        ouch, load back to 40 :\
      • 2018-01-04 00439, 2018

      • ruaok
        the CPU on bowie is the key to look for.
      • 2018-01-04 00450, 2018

      • ruaok
        it dropped for a brief moment, but back above 90%
      • 2018-01-04 00425, 2018

      • ruaok
        we need to think about other things. vacuum analyze didn't work, nor will it work the second time.
      • 2018-01-04 00455, 2018

      • samj1912
        its not from the second time? given the age its from the first time?
      • 2018-01-04 00423, 2018

      • ruaok
        I ran one on bowie.
      • 2018-01-04 00406, 2018

      • samj1912
        yeah, it seems to be that one?
      • 2018-01-04 00432, 2018

      • samj1912
        I wasn't up at the time it ran
      • 2018-01-04 00405, 2018

      • ruaok
        your last two statements make no sense to me
      • 2018-01-04 00411, 2018

      • ruaok
        oh
      • 2018-01-04 00419, 2018

      • ruaok
        nm.
      • 2018-01-04 00436, 2018

      • ruaok
        you weren't up... got it.
      • 2018-01-04 00400, 2018

      • samj1912
        meanwhile, what should we do, I restarted a couple of containers
      • 2018-01-04 00405, 2018

      • samj1912
        didn't seem to help
      • 2018-01-04 00452, 2018

      • ruaok
        restarting postgres is probably the right thing to do, but without knowing how...
      • 2018-01-04 00419, 2018

      • samj1912
        yvanzo: do you have any idea?
      • 2018-01-04 00459, 2018

      • yvanzo
        samj1912: turning hot_standby_feedback off again?
      • 2018-01-04 00413, 2018

      • samj1912
        yvanzo: we will have to restart pg for that
      • 2018-01-04 00425, 2018

      • samj1912
        and its just `on` on the master
      • 2018-01-04 00433, 2018

      • samj1912
        pg docs say it shouldn't matter there
      • 2018-01-04 00426, 2018

      • samj1912
        I am just waiting for the vacuum analyze to complete, its almost running since 2 hrs now, should be near completion
      • 2018-01-04 00437, 2018

      • yvanzo
        samj1912: hot_standby is 'on' on bowie/master, and hot_standby_feedback is 'on' on queen/slave, that would just require restarting pg slave. But I'm not sure it would matter really.
      • 2018-01-04 00409, 2018

      • samj1912
        hot_standby_feedback should be off on queen? O.o
      • 2018-01-04 00426, 2018

      • samj1912
        im pretty sure bitmap turned it off
      • 2018-01-04 00459, 2018

      • samj1912
        yup, its off
      • 2018-01-04 00426, 2018

      • samj1912
        hot_standby is `on` on queen
      • 2018-01-04 00449, 2018

      • yvanzo
        Oops indeed, it is the opposite
      • 2018-01-04 00458, 2018

      • yvanzo
        :(
      • 2018-01-04 00435, 2018

      • samj1912
        yvanzo: do you know how to restart pg master?
      • 2018-01-04 00451, 2018

      • Vini joined the channel
      • 2018-01-04 00412, 2018

      • yvanzo
        nope
      • 2018-01-04 00446, 2018

      • naiveai
        Leo_Verto: do I base my branch off of pipenv-docker?
      • 2018-01-04 00448, 2018

      • samj1912
        yvanzo: do you have any idea what musicbrainz_ro does?