Better overview of what Rockstor/btrfs is doing

My NAS has high iowait sometimes. After a bit of investigation, I found a qgroup clean process causing it.

Would it be possible to get a better overview of scheduled tasks, even if we can’t change them, so we can troubleshoot what is going on?

1 Like

I have noticed that every time I log into the Web interface, a qgroup scan seems to run (Waking up all the disks).
I think it’s done on purpose, but it would be nice if some notice would show up, explaining what is happening.

2 Likes

this might be related to the dashboard metrics as some process must collect all the data, worth a try to disable

1 Like

Damn good detectives you are @roweryan @KarstenV @felixbrucker. I had put in a rescan while fetching usage information for 4.1 kernels. So @felixbrucker is pretty much right. Let me see if I can revise it as I am working on some fs related issues today anyway.

@suman, are these related logs expected?

BTRFS info (device sda): qgroup_rescan_init failed with -115
BTRFS info (device sda): qgroup_rescan_init failed with -115
BTRFS info (device sda): qgroup_rescan_init failed with -115
BTRFS info (device sda): qgroup_rescan_init failed with -115
BTRFS info (device sda): qgroup_rescan_init failed with -115
BTRFS info (device sda): qgroup scan completed (inconsistency flag cleared)

I see them all the time on my system when running DMESG.

I think its the system trying to initiate the qgroup scan, but failing, because the disks are spinned down. When the disks are eventually spinning, the scan completes.

At least thats what I guess is the case on my system.

The times between them are almost identical. I think this might be a bug…