Hi Luis,
I think this is caused by the issue covered by
https://tracker.ceph.com/issues/53266
Gabi is working on that AFAIK.
You might want to verify my hypothesis by setting fast_osd_shutdown to
false and checking startup times afterwards.
Not sure it makes sense to use that setting permanently though as I'm
unaware about all the consequences.
Thanks,
Igor
On 3/2/2022 4:02 PM, Luis Domingues wrote:
Hi all,
We have installed Qincy 17.1.0 on our lab cluster. The update from Pacific 16.2.6 went quite smoothly, but now we have a weird issue.
Everytime we want to restart some OSD running on HDD, they take about 6 minutes to be up, and we have this error on the logs:
```
2022-03-02T12:41:45.863+0000 7f5857f463c0 -1 bluestore::NCB::__restore_allocator::Failed open_for_read with error-code -2
2022-03-02T12:47:54.225+0000 7f5857f463c0 -1 osd.13 21874 log_to_monitors true2022-03-02T12:47:54.882+0000 7f58479d8700 -1 osd.13 21929 set_numa_affinity unable to identify public interface '' numa node: (2) No such file or directory
```
The same error can be found on OSD running on SSDs, but there is no delay, the OSD starts within 10 seconds.
Luis Domingues
Proton AG
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
--
Igor Fedotov
Ceph Lead Developer
Looking for help with your Ceph cluster? Contact us at https://croit.io
croit GmbH, Freseniusstr. 31h, 81247 Munich
CEO: Martin Verges - VAT-ID: DE310638492
Com. register: Amtsgericht Munich HRB 231263
Web: https://croit.io | YouTube: https://goo.gl/PGE1Bx
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
