I feel like I am either misunderstanding something here or have a bug. if a scrub Schedule is set to run weekly and the threshold is 35 days… is it still going to actually run every week, or is it only going to run the next week if it fails to finish the previous weeks attempt?
It will attempt to run every week. However, if the previous scrub task completed successfully, it will not run until 35 days.
So, if you set it to run on e.g. Sunday and complete the task, the next task will occur next Sunday after 35 days.
Okay, then I am having an issue here. I’ve been feeling like my disks have been much more active than usual on maintenance tasks and that seems to be what’s happening. Both my storage pools scrubbed last week.
But here we are a week later and they are running again:
I’m using the default schedules, so it should be another 4 weeks until it runs:
I’ve been doing some rebooting over the last couple months, as part of troubleshooting an app memory issue (also my UPS battery died and had to be replaced), so I figured I was resetting the timer with the shutdown, but I’ve been up and running continuously long enough to confirm this now.
Edit: Opened a ticket on Jira.
With the scrubbing options set, ZFS performs a daily check to see if the pool needs scrubbing, but only scrubs at configured intervals.
What were the results of the previuos scrub? Were there some errors / repairs (zpool status) ?
I have already opened ticket some days ago, look this thread.
Seems they fast fix the issue, but not seems available for now, so in meanwhile i have disabled Scrubs and i will perform them manually
Boot and Jails pool have already finished. No errors. Both data pools are still running, but no errors so far. Not sure how to view the results of past scrubs.
@oxyde Thanks for the heads up.