-
Notifications
You must be signed in to change notification settings - Fork 896
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Bug]: autovacuum worker took too long to start; canceled #7485
Comments
This is quite an old version of TimescaleDB. I suggest updating to a newer version to see if the problem persists. |
The version updates of TimescaleDB are relatively frequent. Version 2.11.1 was released in June of last year, and now the latest version is already 2.17.2. Rapid version updates indicate that the software is active, which is a good thing. However, for our application, upgrading the version may involve some costs, unless we can confirm this issue caused by TimescaleDB. Our data volume is relatively large, with approximately 5 million records per hour. The link #6730 mentions that a table with 1 million records may trigger a failure to start autovacuum. From the logs above, it appears that our database not only failed to start autovacuum at that time, but also experienced many slow SQL queries, resulting in the inability to establish new connections to the database. |
Having looked at this a bit, I found an interesting post on hackers mailing list: Seems very similar to what you are experiencing, also on Windows platform. Are you able to make new connections to the server once this starts happening? I assume that's what is going on so the autovacuum workers cannot start. Have you reached max_connections perhaps? At this point, it seems like it has nothing to do with Timescaledb but rather a PG/Windows issue. Hope this helps. |
Thank you for your reply and findings. The case you discovered is very similar to ours. At that time, new connections could not be established either, and we hadn't reached the max_connections limit (we calculated that the max_connections was set to 200, but there were less than 50 connections at the time of the failure). Another similarity is that both cases used table spaces set up on other disks. This case is from 2017, has Postgres not resolved this issue after all this time? |
What type of bug is this?
Locking issue, Performance issue, Unexpected error
What subsystems and features are affected?
Background worker
What happened?
TimescaleDB version affected
2.11.1
PostgreSQL version used
15.2
What operating system did you use?
windows 10
What installation method did you use?
Source
What platform did you run on?
On prem/Self-hosted
Relevant log output and stack trace
How can we reproduce the bug?
The text was updated successfully, but these errors were encountered: