-
-
Notifications
You must be signed in to change notification settings - Fork 720
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
Added support for ZFS pool io stats monitoring #374
Added support for ZFS pool io stats monitoring #374
Conversation
this would be great to have. it's rather surprising to not see ZFS-related disk I/O stats in btop. i would understand we don't have it per dataset, but surely it should be registered in the merged view! also, it would be nice to have pools instead of dataset in the disk usage listing, does this PR change that? |
No, this PR is focused only on ZFS pool io monitoring, I didn't alter the listings logic in any way. |
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.
Looks great, other than the requested change, awesome work :)
@simplepad |
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.
Awesome :)
Thanks for contributing!
I haven't tried the new btop yet, but in reviewing the comments above I don't see the
I only see it on my much older Ubuntu ZFS versions:
|
I was testing Debian Bullseye's last version of zfs-utils, and it had this file. Looking at the commit of openzfs you provided, it does look like my implementation is not compatible with the latest versions of zfs. That said, there shouldn't be any errors using the latest btop on your system, as it checks if this file exists before using it. Also, could you please open an issue to track this? |
This pull request adds support for ZFS pool io monitoring, using file
/proc/spl/kstat/zfs/*pool_name*/io
.Reads and writes are recorded as bytes, not sectors, so there is no need to multiply it by 512.
Since zfs stat collection is different from block device stat collection,
disk
has to store filesystem type to determine which collection method should be used.Also,
io_ticks
for zfs are recorded separetely for reads and writes, so to get a total I addedio_ticks_read
andio_ticks_write
variables.