DB Timeouts
I've been getting timeouts leading to notifications/timelines becoming unresponsive since 2.0.4 stable OTP. It seems others have been mentioning issues lately too? Anyway in syslog I get things like this:
May 16 20:53:44 localhost pleroma[25540]: 20:53:44.268 [error] GenServer Oban.Queue.ActivityExpiration.Producer terminating May 16 20:53:44 localhost pleroma[25540]: ** (DBConnection.ConnectionError) connection not available and request was dropped from queue after 1997ms. This means requests are coming in and your connection pool cannot serve them fast enough. You can address this by: May 16 20:53:44 localhost pleroma[25540]: 1. By tracking down slow queries and making sure they are running fast enough May 16 20:53:44 localhost pleroma[25540]: 2. Increasing the pool_size (albeit it increases resource consumption) May 16 20:53:44 localhost pleroma[25540]: 3. Allow requests to wait longer by increasing :queue_target and :queue_interval May 16 20:53:44 localhost pleroma[25540]: See DBConnection.start_link/2 for more information May 16 20:53:44 localhost pleroma[25540]: (db_connection) lib/db_connection.ex:745: DBConnection.run/3 May 16 20:53:44 localhost pleroma[25540]: (stdlib) gen_server.erl:637: :gen_server.try_dispatch/4 May 16 20:53:44 localhost pleroma[25540]: (stdlib) gen_server.erl:711: :gen_server.handle_msg/6 May 16 20:53:44 localhost pleroma[25540]: (stdlib) proc_lib.erl:249: :proc_lib.init_p_do_apply/3 May 16 20:53:44 localhost pleroma[25540]: Last message: :poll May 16 20:53:44 localhost pleroma[25540]: State: %Oban.Queue.Producer.State{circuit: :disabled, conf: %Oban.Config{circuit_backoff: 30000, crontab: [{%Oban.Crontab.Cron{days: [:*], hours: [0], minutes: [0], months: [:*], weekdays: [:*]}, Pleroma.Workers.Cron.ClearOauthTokenWorker, []}, {%Oban.Crontab.Cron{days: [:*], hours: [:*], minutes: [0], months: [:*], weekdays: [:*]}, Pleroma.Workers.Cron.StatsWorker, []}, {%Oban.Crontab.Cron{days: [:*], hours: [:*], minutes: [:*], months: [:*], weekdays: [:*]}, Pleroma.Workers.Cron.PurgeExpiredActivitiesWorker, []}, {%Oban.Crontab.Cron{days: [:*], hours: [0], minutes: [0], months: [:*], weekdays: [0]}, Pleroma.Workers.Cron.DigestEmailsWorker, []}, {%Oban.Crontab.Cron{days: [:*], hours: [0], minutes: [0], months: [:*], weekdays: [:*]}, Pleroma.Workers.Cron.NewUsersDigestWorker, []}], name: Oban, node: "pleroma@127.0.0.1", poll_interval: 1000, prefix: "public", prune: {:maxlen, 1500}, prune_interval: 60000, prune_limit: 5000, queues: [federator_incoming: 50, federator_outgoing: 50, web_push: 50, mailer: 10, transmogrifier: 20, scheduled_activities: 10, background: 5, remote_fetcher: 2, attachments_cleanup: 5, new_users_digest: 1, activity_expiration: 20], repo: Pleroma.Repo, rescue_after: 60, rescue_interval: 60000, shutdown_grace_period: 15000, verbose: false}, foreman: Oban.Queue.ActivityExpiration.Foreman, limit: 20, name: Oban.Queue.ActivityExpiration.Producer, nonce: "1h2jjuco", paused: false, poll_ref: #Reference<0.360831197.415760385.100970>, queue: "activity_expiration", rescue_ref: #Reference<0.360831197.415760385.100971>, running: %{}, started_at: ~U[2020-05-17 00:53:39.283787Z]} May 16 20:53:44 localhost pleroma[25540]: 20:53:44.268 [error] GenServer Oban.Queue.FederatorOutgoing.Producer terminating
I've tried a vacuum analyze and it seemed to help a bit but it's been intermittent. Any ideas on other things I can try?