How to get current dirty data size？ #9620
-
System information
Describe the problem you're observingI had set zfs_dirty_data_max but how to check the usage to adjust this parameter? Describe how to reproduce the problemInclude any warning/errors/backtraces from the system logs |
Beta Was this translation helpful? Give feedback.
Replies: 5 comments
-
|
This is tracked in |
Beta Was this translation helpful? Give feedback.
-
|
For telemetry services, that value is too volatile to be useful. We do track the contents of |
Beta Was this translation helpful? Give feedback.
-
|
@richardelling where could I found the two dashboard?thanks. |
Beta Was this translation helpful? Give feedback.
-
|
@richardelling, can you share the Dashboards? |
Beta Was this translation helpful? Give feedback.
-
|
The OpenZFS wiki seems to indicate that you can track dirty data by reviewing the ndirty column in /proc/spl/kstat/zfs/{poolname}/txgs. I've been watching mine, and it definitely spikes up and down when I restart a VM workload. Saw it exceed my zfs_dirty_data_max by ~1.5 MB once. Otherwise it always stays below. |
Beta Was this translation helpful? Give feedback.
For telemetry services, that value is too volatile to be useful. We do track the contents of
/proc/spl/kstat/zfs/dmu_txin bothnode_exporterandtelegrafThe data there can be useful for monitoring delays due to the write throttle and I've got a dashboard or two that show examples of how to look at that data.