Karl-Aksel Puulmann
Karl-Aksel Puulmann
I think this can be solved by setting "Restrict who can push to matching branches" and making sure that commits are made by posthog bot not ourselves. :)
Not necessarily a blocker, but a thing to consider here - can we give actionable feedback if the install fails in a simple way? Without `--debug` I believe it would...
This is caused by finalizers in the clickhouse-operator. It might be a 'feature' in that it doesn't drop the data. A one-command workaround is `kubectl patch chi posthog -n posthog...
The merged fix didn't quite work - the app is connecting to clickhouse on different ports: - App is using 9000 or 9443 by default depending on whether connection is...
cc @ellie @guidoiaquinti I think we'll need 2 (up to 4) env variables introduced into the app: - CLICKHOUSE_HTTP_PORT - CLICKHOUSE_TCP_PORT The defaults would depend on whether `CLICKHOUSE_SECURE` See also...
I think the issue might be that if a previous request is in-flight then `shutdownAsync` is not awaiting those.
On benchmarking: - The result is IMO a bit inconclusive - benchmarking results under
> As for "benchmarking", I'll take a look at the other shard explicitly, although I suspect querying on all replicas should have given me that? No because the whole query...
cc @yakkomajuri might be a good bug to pick up :)
Would you feel comfortable only appending these metrics only on self-hosted?