but hopefully we can figure something out, anyway :/
2013-04-10 10002, 2013
nikki
could it be something to do with cache stuff suddenly expiring?
2013-04-10 10010, 2013
ianmcorvidae
hm?
2013-04-10 10021, 2013
abelcheung
DB cache?
2013-04-10 10018, 2013
nikki
well, we cache data, right? perhaps it all suddenly gets deleted and then has to be refetched from the db and totoro is like "wtf guys not all at once D:"
2013-04-10 10038, 2013
nikki is just thinking of things that sound reasonably plausible :P
2013-04-10 10017, 2013
uptown1 joined the channel
2013-04-10 10032, 2013
ianmcorvidae
yeah, I dunno, that doesn't seem to explain it going for six hours or whatever :/
2013-04-10 10023, 2013
abelcheung
I would think about arrogant spiders or mirror first
and.... i suppose that's old version of db which is not in active use
2013-04-10 10030, 2013
ianmcorvidae
no, 20110516 is the current DB
2013-04-10 10032, 2013
ianmcorvidae
if that's what you mean
2013-04-10 10039, 2013
abelcheung
ouch
2013-04-10 10049, 2013
ianmcorvidae
that's when NGS was released, heh
2013-04-10 10007, 2013
abelcheung
some ulimit or pg limit might be at work too
2013-04-10 10021, 2013
ianmcorvidae
yeah, was looking for things in config
2013-04-10 10055, 2013
abelcheung
before Jan it's going up to 6k, but never above 4-5 ever since -- and it's like a steady rhythm, that's too unnatural
2013-04-10 10003, 2013
ianmcorvidae
we upgraded pg around 17UTC on the 5th
2013-04-10 10017, 2013
abelcheung
that is, less than a week ago?
2013-04-10 10041, 2013
ianmcorvidae
yes
2013-04-10 10059, 2013
ianmcorvidae
there was a major security vulnerability, thus
2013-04-10 10004, 2013
abelcheung
Probably can get some hint by enabling slow log on pgsql. But this is production DB....
2013-04-10 10034, 2013
ianmcorvidae
things time out and get cancelled anyway, I'm not sure it's slow queries -- it's apparently queries that are returning -- and returning a *lot*
2013-04-10 10025, 2013
ianmcorvidae
but apparently not returning it the whole way to the client, just internally
2013-04-10 10054, 2013
abelcheung
There could be multiple problems. For example, when you guys saying the lagging is cured, I am still constantly seeing 502
2013-04-10 10011, 2013
ianmcorvidae
well, there's multiple problems
2013-04-10 10045, 2013
ianmcorvidae
trying to figure out the one that's having the biggest effect first
2013-04-10 10058, 2013
ianmcorvidae
(e.g. some large releases have always 502'd repeatedly)
2013-04-10 10025, 2013
abelcheung
not just large releases :)
2013-04-10 10056, 2013
abelcheung
I'd say there's bigger problem when even single CD edits results in 502 repeatedly
2013-04-10 10054, 2013
ianmcorvidae
that is not a phenomenon that is well-reported; if it's a specific set of releases, rather than a cross-section, it's most likely related to those releases specifically unless it's during the blocks of time where we have known other problems
2013-04-10 10010, 2013
ianmcorvidae
you've seen the 5xx errors graph, so you can see what is anomaly and what isn't
2013-04-10 10029, 2013
abelcheung
it's easier when reading graphs, but for personal experience that's harded to tell
2013-04-10 10028, 2013
abelcheung
but slow query log can still help a bit too, at least help tracing what's the offending source
2013-04-10 10004, 2013
abelcheung
because the sudden surge causes exploding I/O and dragging everything down
2013-04-10 10059, 2013
abelcheung
especially if the surge can be irrelevant from increase of external requests
2013-04-10 10009, 2013
ianmcorvidae
hm
2013-04-10 10016, 2013
ianmcorvidae
it seems it is, at least, swapping during these periods
2013-04-10 10055, 2013
abelcheung
swapping is one of the worst things for DB
2013-04-10 10019, 2013
ianmcorvidae
yeah
2013-04-10 10025, 2013
ianmcorvidae
still unclear what's causing any of this though
2013-04-10 10036, 2013
ianmcorvidae
we're just getting an ever-better description of why it's bad :P
2013-04-10 10042, 2013
abelcheung
yup
2013-04-10 10010, 2013
abelcheung
back to square one: why the sudden increase in no. of processes
2013-04-10 10056, 2013
abelcheung
from less than 100 on average, jumping to max'ed 256
2013-04-10 10045, 2013
ianmcorvidae
hm
2013-04-10 10002, 2013
ianmcorvidae
what it *looks* like is as though it's not using pgbouncer, for some undefined reason
2013-04-10 10017, 2013
ianmcorvidae
looking at the yearly graph (we instituted pgbouncer last august IIRC
2013-04-10 10048, 2013
abelcheung
the graph for no. of db transactions?
2013-04-10 10017, 2013
abelcheung
Hmm, DB block hit is much lower since 7th
2013-04-10 10011, 2013
abelcheung
Urgh, tried with a few single CDs with >25 tracks, editing all tracks together also result in guaranteed 502 :-(
2013-04-10 10037, 2013
danoply joined the channel
2013-04-10 10040, 2013
DarkAudit joined the channel
2013-04-10 10043, 2013
v6lur joined the channel
2013-04-10 10048, 2013
gmk1 joined the channel
2013-04-10 10024, 2013
outsidecontext joined the channel
2013-04-10 10024, 2013
SultS joined the channel
2013-04-10 10002, 2013
nioncode joined the channel
2013-04-10 10048, 2013
jacobbrett joined the channel
2013-04-10 10034, 2013
jesus2099 joined the channel
2013-04-10 10036, 2013
jesus2099
it seems that editing an existing tracklist through TRACK PARSER doesn’t owrk any more… :/
2013-04-10 10009, 2013
nioncode joined the channel
2013-04-10 10004, 2013
Leftmost
I have a few CDs of field recordings of traditional music. They generally only have one or two tracks with any performer information, located in the liner notes. Should I just Various Artists it and [unknown] most of the artists?
