Zilch
IImmich
•Created by Zilch on 8/22/2024 in #help-desk-support
Unable to join new cnpg instance
Good Morning,
I know you guys aren't cloudnative-pg or pg_vectors support, but I'm having an issue that appears to be the result of something with pg_vectors and you all are waaaay more knowledgeable in this area than I- I also may not be the first or last of your users to encounter this.
My setup:
K8S
Cloudnative-pg version 16.4 with using the tensorchord cloudnative-pgvector.rs image with pgvector.rs 0.3.0
I recently replaced a node, in my cluster, and cloudnative began the join process to replace the postgres node that was removed during the swap, however, it now produces an error indicating that the backup fails as a result of a filename being too long for tar format
I've tried backing down to 16.3, but no change in behavior- it seems like this might be tied to the new pgvectors
I can paste the log as an additional comment due to length limits
7 replies