[ceph-users] Re: Full OSD with almost empty pools

[Adam Prycki]

Hi,

I've just tried to do it.
Fresh OSD is down gets stuck in preboot
It seems to slowly download maps from the cluster.
It's probably the issue Igor mentioned.


{
    "cluster_fsid": "bec60cda-a306-11ed-abd9-75488d4e8f4a",
    "osd_fsid": "fbfeba00-3daa-461e-9e02-5c257e677e8a",
    "whoami": 2349,
    "state": "preboot",
    "maps": "[1502335~121041]",
    "oldest_map": "1502335",
    "newest_map": "1623375",
    "cluster_osdmap_trim_lower_bound": 1502335,
    "num_pgs": 0
}

What's the healthy amount of maps OSD should store?

Best regards
Adam Prycki

W dniu 5.05.2025 o 17:11, Robert Sander pisze:

Am 5/5/25 um 16:58 schrieb Adam Prycki:

Evacuating all PGs from OSD doesn't lower it's usage.
For example

1814    ssd        0   1.00000  447 GiB  382 GiB  381 GiB  351 MiB 1.0 GiB  65 GiB  85.44  1.00    0      up I suspect its OSD/bluestore issue since data usage seems not to come from PGs.

What happens when you recreate this OSD?

Regards

smime.p7s
Description: Kryptograficzna sygnatura S/MIME

_______________________________________________
ceph-users mailing list -- ceph-users@ceph.io
To unsubscribe send an email to ceph-users-le...@ceph.io