-
Couldn't load subscription status.
- Fork 13.9k
CI: optimize ext4 mount options that can be toggled at runtime. #148085
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
Conversation
This comment has been minimized.
This comment has been minimized.
7c16001 to
026314e
Compare
|
|
||
| if isLinux && [[ $(findmnt -n /) =~ "ext4" ]] ; then | ||
| # noauto_da_alloc since auto_da_alloc causes sync IO for some common file creation patterns | ||
| # lazytime avoids sync IO when (rel)atime updates are applied |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is it possible to use noatime here?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Deleting stuff doesn't just update atime but also ctime and mtime, so I don't think noatime would make a difference here.
Anyway, the settings don't seem to make much of a difference, so this might not be due to sync-write-latency but due to read latency (enumerating the directories) or just being IOPS-starved (anemic cloud hardware).
I'll try running iostat alongside the deletion.
026314e to
837210f
Compare
Looks like this actually limited by read latency. |
fc302e1 to
549db0c
Compare
This comment has been minimized.
This comment has been minimized.
14c1db1 to
78306cb
Compare
78306cb to
964c658
Compare
|
Parallel deletion seems to help a bit, but not enough, trying something else in #148146 |
Another attempt at #82099. The "free up disk space" step is taking 10m, that's a bit much.