TimescaleDB 2.4 / PostgreSQL 12 are not compressing or erasing old data.

[mbuyukkarakas]

Dear all

For the test purposes, I tried to install PGSQL 12 + Timescaledb 2.4. No problem during the setup. But after running zabbix services, I started to receive that kind of errors. And TimescaleDB is not compressing and erasing old data.

2021-10-24 21:56:33.975 EDT [16729] LOG: TimescaleDB background worker launcher connected to shared catalogs
2021-10-25 21:56:03.009 EDT [16731] WARNING: failed to launch job 1003 "Compression Policy [1003]": failed to start a background worker
[root@192 yum.repos.d]# cat /var/lib/pgsql/13/data/log/postgresql-Sat.log
2021-10-23 19:56:00.477 EDT [15787] WARNING: failed to launch job 1004 "Compression Policy [1004]": failed to start a background worker
2021-10-23 19:56:00.480 EDT [15787] WARNING: failed to launch job 1005 "Compression Policy [1005]": failed to start a background worker
2021-10-23 19:56:00.521 EDT [15787] WARNING: failed to launch job 1005 "Compression Policy [1005]": out of background workers

Which versions of TS and PG are compatible , anybody has an idea ? I was able to run smoothly with PG12 + TS 1.7 on Centos7. But as you know Centos7 is now unsupported and I need the correct / compatible versions.

thank you.
mehmet

[mbuyukkarakas]

Does somebody has any idea about this ? I dont want to believe nobody is using timescaledb with pgsql 12 or higher versions

best regards

[splitek]

I have PG12, TSDB 1.7 but CentOS 8 and Zabbix 5, no problems.

Maybe try to tune parameter max_background_workers:

https://docs.timescale.com/timescaledb/latest/how-to-guides/configuration/postgres-config/#workers

I just found this (this is recommendation for Zabbix 6.0 LTS): https://support.zabbix.com/browse/ZBXNEXT-6898