hi! having this in logs (and some tests started to...
# pact-broker
d
hi! having this in logs (and some tests started to fail due to timeouts), any ideas ?
Copy code
2022-03-22T14:02:01.354538416Z stdout F /pact_broker/vendor/bundle/ruby/2.6.0/gems/sequel-5.49.0/lib/sequel/connection_pool/threaded.rb:264:in `raise_pool_timeout'
2022-03-22T14:02:01.354486796Z stdout F 2022-03-22 14:02:01.351645 E [8:puma server threadpool 005 logging.rb:48] PactBroker::Errors::ErrorLogger -- Error reference YEkrNbMqTK -- Exception: Sequel::PoolTimeout: timeout: 5.0, elapsed: 5.000220500980504Show context
b
looks like a pool timeout to me...
😛
how many connections does your database allow?
are you having a lot of concurrent activity? often a mono-repo build will kick off a bunch of activity at once and hammer your database.
you can control the number of connections in the thread pool and the time out in the config settings.
d
too many connections would kick different error, as i understand
but yes, feels like monorepo ddosing broker
how many connections does your database allow?
default one 🙂
Copy code
`PACT_BROKER_DATABASE_MAX_CONNECTIONS` - optional, defaults to 4. The maximum size of the connection pool.
`PACT_BROKER_DATABASE_POOL_TIMEOUT` - optional, 5 seconds by default. The number of seconds to wait if a connection cannot be acquired before raising an error.
b
Try upping the number of connections allowed by your db?