hi! i'm working with one of the core
hi! i'm working with one of the core contributors over at zerosync. We're consistently seeing connection interruptions in the replication stream, and can't 100% confirm whether this is a neon issue or not. Its an issue that seemingly exists only with large replications (a million+ rows).
I'm wondering if we might be able to get some help from someone at neon in debugging the issue and getting better access to logs - particularly
wal inspections. I need superuser permissions to be able to install CREATE EXTENSION pg_walinspect; which would help us alot.
I also am having a hard time getting the datadog integration for postgres logs to show anything other then an 'error' status, with no understanding of what the error actually is or how to fix it.3 Replies
other-emerald•5mo ago
Ho @krystian We should set up a Slack channel for the Zero team. In the meantime, @raoufchebri @David Gomes Can we help @krystian ?
wise-white•5mo ago
Hey @krystian what's the instance size and how much in storage?
Can you DM or email your project-id? raouf@neon.tech
stormy-gold•5mo ago
sure thing!
just DM'd you
1 <-> 8 CPU, storage says 38.28GB
although the size of any replicas i create is around 1.6GB