Hello! We ran out of memory right after we upgrade...
# community-helpg
Guillermo Dorfman
11/28/2024, 4:54 PMHello! We ran out of memory right after we upgraded the cluster from v0.25.1 ti v27.1 is it possible that the automatic re-indexing of all the documents after the upgrade depleted the memory?
Guillermo Dorfman
11/28/2024, 5:08 PMA follow up question: If I edit the schemas and index less fields, re-indexing those collections would free-up memory?
j
Jason Bosco
11/28/2024, 5:14 PMis it possible that the automatic re-indexing of all the documents after the upgrade depleted the memory?Yeah that's possible. Also, v27.1 has a bug fix with facet index construction - the fix now indexes all facet values properly, when previously some values were missed from the facet index in previous versions. So that could cause an increase in memory as well.
If I edit the schemas and index less fields, re-indexing those collections would free-up memory?Yup
🙏 1
g
Guillermo Dorfman
11/28/2024, 6:01 PMGreat, thank you!
Guillermo Dorfman
11/28/2024, 6:48 PMError: t: Request failed with HTTP code 422 | Server said: Rejecting write: running out of resource type: OUT_OF_MEMORY
I cant drop the fields on the schema due to lack of memory, should I delete the collection and create it again?
j
Jason Bosco
11/28/2024, 6:49 PMI would recommend temporarily upgrading RAM, then doing the schema change and then scale back down