2 months ago
I've recently migrated my Umami instance from v2 to v3, and noticed an unexpected increase of my postgres volume usage from 3gb to a 97% of maximum size.
I threat is as unexpected since my data has not mostly changed, i've tried running a query that returns me the size of each table and size matches (around ~3gb of data).
I can say that during the update I had some trouble running a migration which throws me that error:
Context: Timed out trying to acquire a postgres advisory lock (SELECT pg_advisory_lock(72707369)). Timeout: 10000ms. See https://pris.ly/d/migrate-advisory-locking for details.
Running the same migration via IntelliJ query or dbeaver throws me a Unsufficent disk space after some minutes (https://github.com/umami-software/umami/blob/master/prisma/migrations/09_update_hostname_region/migration.sql) so I continued skipping it.
The thing is, after this, it seems the volume has increased significantly in size, so I wanted to see if there's anything left hanging there that's taking up space.
Attachments
2 Replies
2 months ago
That's should be normal, the graph in the usage dashboard is cumulative, so it will always increase
fra
That's should be normal, the graph in the usage dashboard is cumulative, so it will always increase
2 months ago
What's strange to me is that It seems to me like all the available space is being hung up on that query that was giving me an error. I went from 60% to about 97% volume space overnight.
I currently have more than ~200k saved events on Umami and with that query I wrote the used space by that table (including indexes) were around 1GB. I expected to be able to store at least double or triple the size, but now I don't see any space left even though I think 'm not actually using it.
I created this service 4 years ago, usage cannot be increased by 2GB like this in 1 hour after that update. I had to update because of react CVE, for some reason the service went down by itself