Actions
action #97412
openopenQA Project (public) - coordination #103944: [saga][epic] Scale up: More robust handling of diverse infrastructure with varying performance
openQA Project (public) - coordination #98463: [epic] Avoid too slow asset downloads leading to jobs exceeding the timeout with or run into auto_review:"(timeout: setup exceeded MAX_SETUP_TIME|Cache service queue already full)":retry
Reduce I/O load on OSD by using more cache size on workers with using free disk space when available instead of hardcoded space
Start date:
Due date:
% Done:
0%
Estimated time:
Description
Motivation¶
I am sure we could spread the load on OSD when many job start a bit if we manage to have less assets that need to be downloaded at the same time. We can increase the cache size, e.g. use all available space instead of the artifical limits of the cache directory.
The worker cache could just ensure a certain percentage of free disk space in the file system.
Acceptance criteria¶
- AC1: At least most workers on OSD use all available free disk space except for a configured ratio to keep free
Updated by okurz over 3 years ago
- Copied from action #96554: Mitigate on-going disk I/O alerts size:M added
Actions