| 23 Aug 2023 |
@linus:schreibt.jetzt | I think it has a 30s timeout | 11:07:52 |
Julien | I yes | 11:16:12 |
Julien | * Ah yes | 11:16:21 |
Julien | It looks that that’s the problem | 11:16:29 |
Julien | * It looks like that’s the problem | 11:16:39 |
@linus:schreibt.jetzt | I think the suggested chunking params are pretty heavy on the database, so you might need to adjust the chunking, or you might need a more powerful or maybe just better tuned postgres server | 11:19:24 |
@linus:schreibt.jetzt | that is, adjust the chunking so that you get fewer chunks, but also less effective deduplication | 11:19:41 |
raitobezarius | postgres defaults are extremely conservative | 11:23:32 |
raitobezarius | tuning postgres is probably the easiest | 11:23:38 |
@linus:schreibt.jetzt | I'm trying that out now | 11:30:32 |
@linus:schreibt.jetzt | using https://pgtune.leopard.in.ua/ and the web application profile | 11:30:44 |
@linus:schreibt.jetzt | though maybe desktop is more suitable given that it's not a dedicated machine | 11:31:39 |
@linus:schreibt.jetzt | well, I'll find out :) | 11:31:47 |
Julien | I’ll use my Raito script to tune postgres | 11:50:43 |
raitobezarius | krkrkrkrkr | 11:54:46 |
@linus:schreibt.jetzt | hm, it seems to result in more successes, but doesn't entirely eliminate the failures | 12:03:42 |
raitobezarius | what is the nature of pgsql connections that attic does? | 12:04:52 |
raitobezarius | read or write? | 12:04:55 |
@linus:schreibt.jetzt | I could also imagine that atticd doesn't reuse connections well enough | 12:05:01 |
raitobezarius | in some cases, a pooler could help | 12:05:04 |
raitobezarius | yeah | 12:05:07 |
@linus:schreibt.jetzt | it does pool them | 12:05:08 |
@linus:schreibt.jetzt | Aug 23 13:58:54 garage-muc atticd[286]: 2023-08-23T11:58:54.590061Z ERROR tower_http::trace::on_failure: response failed classification=Status code: 500 Internal Server Error latency=30003 ms
Aug 23 14:02:44 garage-muc atticd[286]: 2023-08-23T12:02:44.976132Z ERROR attic_server::error: Database error: Failed to acquire connection from pool
| 12:05:45 |
raitobezarius | how many connections do you allow now? | 12:06:02 |
@linus:schreibt.jetzt | I have no idea :D | 12:06:08 |
@linus:schreibt.jetzt | maybe I should just dig into it rather than putting every intermediate thought in here :p | 12:06:39 |
raitobezarius | max_connections is the right knob | 12:06:51 |
@linus:schreibt.jetzt | I don't think this is on the postgres side, it's on the attic side | 12:07:15 |
raitobezarius | Ah right I agree | 12:07:27 |
raitobezarius | https://github.com/NixOS/nixos-org-configurations/tree/master/metrics/fastly | 13:19:45 |