OSD showing up when all are down
garfield659
10 Posts
August 9, 2024, 7:58 pmQuote from garfield659 on August 9, 2024, 7:58 pmRunning PetaSAN 3.3.0 and when I have all the storage nodes turned off, the Dashboard is showing 16 total, 4 up, and 12 down. If I go do nodes, it shows all 4 of the Storage nodes (each has 4 drives) are all down. I also confirmed the servers are off as well. I don't know an easy way to post pictures of the issue.
root@gs-ss1-ps1:~# ceph -s
cluster:
id: b1042b3d-e000-4a7c-9b6e-c03a8a8ea373
health: HEALTH_WARN
1 datacenter (8 osds) down
8 osds down
2 hosts (8 osds) down
Reduced data availability: 513 pgs inactive
I am 100% certain all 16 drives are down.
services:
mon: 3 daemons, quorum gs-ss1-ps3,gs-ss1-ps1,gs-ss1-ps2 (age 3m)
mgr: gs-ss1-ps2(active, since 4m), standbys: gs-ss1-ps1, gs-ss1-ps3
osd: 16 osds: 4 up (since 57m), 12 in (since 39m)
data:
pools: 2 pools, 513 pgs
objects: 0 objects, 0 B
usage: 0 B used, 0 B / 0 B avail
pgs: 100.000% pgs unknown
513 unknown
root@gs-ss1-ps1:~#
Running PetaSAN 3.3.0 and when I have all the storage nodes turned off, the Dashboard is showing 16 total, 4 up, and 12 down. If I go do nodes, it shows all 4 of the Storage nodes (each has 4 drives) are all down. I also confirmed the servers are off as well. I don't know an easy way to post pictures of the issue.
root@gs-ss1-ps1:~# ceph -s
cluster:
id: b1042b3d-e000-4a7c-9b6e-c03a8a8ea373
health: HEALTH_WARN
1 datacenter (8 osds) down
8 osds down
2 hosts (8 osds) down
Reduced data availability: 513 pgs inactive
I am 100% certain all 16 drives are down.
services:
mon: 3 daemons, quorum gs-ss1-ps3,gs-ss1-ps1,gs-ss1-ps2 (age 3m)
mgr: gs-ss1-ps2(active, since 4m), standbys: gs-ss1-ps1, gs-ss1-ps3
osd: 16 osds: 4 up (since 57m), 12 in (since 39m)
data:
pools: 2 pools, 513 pgs
objects: 0 objects, 0 B
usage: 0 B used, 0 B / 0 B avail
pgs: 100.000% pgs unknown
513 unknown
root@gs-ss1-ps1:~#
admin
2,929 Posts
August 11, 2024, 4:07 pmQuote from admin on August 11, 2024, 4:07 pmThis is expected. Ceph is a distributed system, the detection of downed OSDs is done among running OSDs.
This is expected. Ceph is a distributed system, the detection of downed OSDs is done among running OSDs.
OSD showing up when all are down
garfield659
10 Posts
Quote from garfield659 on August 9, 2024, 7:58 pmRunning PetaSAN 3.3.0 and when I have all the storage nodes turned off, the Dashboard is showing 16 total, 4 up, and 12 down. If I go do nodes, it shows all 4 of the Storage nodes (each has 4 drives) are all down. I also confirmed the servers are off as well. I don't know an easy way to post pictures of the issue.
root@gs-ss1-ps1:~# ceph -s
cluster:
id: b1042b3d-e000-4a7c-9b6e-c03a8a8ea373
health: HEALTH_WARN
1 datacenter (8 osds) down
8 osds down
2 hosts (8 osds) down
Reduced data availability: 513 pgs inactive
I am 100% certain all 16 drives are down.
services:
mon: 3 daemons, quorum gs-ss1-ps3,gs-ss1-ps1,gs-ss1-ps2 (age 3m)
mgr: gs-ss1-ps2(active, since 4m), standbys: gs-ss1-ps1, gs-ss1-ps3
osd: 16 osds: 4 up (since 57m), 12 in (since 39m)data:
pools: 2 pools, 513 pgs
objects: 0 objects, 0 B
usage: 0 B used, 0 B / 0 B avail
pgs: 100.000% pgs unknown
513 unknownroot@gs-ss1-ps1:~#
Running PetaSAN 3.3.0 and when I have all the storage nodes turned off, the Dashboard is showing 16 total, 4 up, and 12 down. If I go do nodes, it shows all 4 of the Storage nodes (each has 4 drives) are all down. I also confirmed the servers are off as well. I don't know an easy way to post pictures of the issue.
root@gs-ss1-ps1:~# ceph -s
cluster:
id: b1042b3d-e000-4a7c-9b6e-c03a8a8ea373
health: HEALTH_WARN
1 datacenter (8 osds) down
8 osds down
2 hosts (8 osds) down
Reduced data availability: 513 pgs inactive
I am 100% certain all 16 drives are down.
services:
mon: 3 daemons, quorum gs-ss1-ps3,gs-ss1-ps1,gs-ss1-ps2 (age 3m)
mgr: gs-ss1-ps2(active, since 4m), standbys: gs-ss1-ps1, gs-ss1-ps3
osd: 16 osds: 4 up (since 57m), 12 in (since 39m)
data:
pools: 2 pools, 513 pgs
objects: 0 objects, 0 B
usage: 0 B used, 0 B / 0 B avail
pgs: 100.000% pgs unknown
513 unknown
root@gs-ss1-ps1:~#
admin
2,929 Posts
Quote from admin on August 11, 2024, 4:07 pmThis is expected. Ceph is a distributed system, the detection of downed OSDs is done among running OSDs.
This is expected. Ceph is a distributed system, the detection of downed OSDs is done among running OSDs.