Bug #62704
closedCephfs different monitor has a different LAST_DEEP_SCRUB state
0%
Description
I ran 'ceph pg dump pgs' on each monitor node and find that the values of REPORTED are inconsistent. For example:
The result in Monitor 1:
1.72 67 0 0 0 0 247403520 0 0 4402 0 4402 active+clean 2023-09-02T07:50:31.065231+0000 1585'4702 1586:7591 [4,0,7] 4 [4,0,7] 4 1579'3650 2023-09-02T07:50:31.064729+0000 0'0 2023-08-31T02:43:38.055980+0000 0 1 periodic scrub scheduled @ 2023-09-03T14:09:43.741830+0000 67 0
While the one in Monitor 2:
1.72 67 0 0 0 0 247403520 0 0 4402 0 4402 active+clean 2023-09-02T07:50:31.065231+0000 1585'4702 1586:7592 [4,0,7] 4 [4,0,7] 4 1579'3650 2023-09-02T07:50:31.064729+0000 0'0 2023-08-31T02:43:38.055980+0000 0 1 periodic scrub scheduled @ 2023-09-03T14:09:43.741830+0000 67 0
And this is constantly inconsistent, even if I make no further actions from my client.
Updated by Venky Shankar 9 months ago
- Project changed from CephFS to RADOS
Unrelated to cephfs - moving to RADOS component.
Updated by Radoslaw Zarzynski 8 months ago
Is this consistent? Was there a third attempt, on mon node 1, showing 1586:7591
again?
And BTW, this is mgr-handled command so not seeing why mon should be involved.
Updated by fuchen ma 8 months ago
Radoslaw Zarzynski wrote:
Is this consistent? Was there a third attempt, on mon node 1, showing
1586:7591
again?
And BTW, this is mgr-handled command so not seeing why mon should be involved.
Tried for several times, still showing the same thing. Maybe not related to mon, so the title of this issue may be misunderstanding. Apologize for that.
Updated by Radoslaw Zarzynski 8 months ago
- Status changed from New to Need More Info
Hi! Could you please elaborate on the symptom? Was the integer values absolutely the same in the third round?
Updated by fuchen ma 8 months ago
Radoslaw Zarzynski wrote:
Hi! Could you please elaborate on the symptom? Was the integer values absolutely the same in the third round?
Yes, the '1586:7591' still exists on mon node 1. But, I'm new to ceph and rados, so I don't quite understand what the 'third round' means. When I tried 'ceph pg dump pgs' for more times, there are more inconsistent values besides '1586:7591' and '1586:7592'. But yes, the '1586:7591' remains there.
Updated by Radoslaw Zarzynski 8 months ago
By the "round" I meant an instance of ceph pg dump pgs
execution.
Anyway, how about moving to ceph-users as this might simply a misunderstanding of the interface.
Updated by Radoslaw Zarzynski 7 months ago
- Status changed from Need More Info to Closed