20 May
2021
20 May
'21
6:56 p.m.
On 20 May 2021, at 19:47, Igor Fedotov
<ifedotov(a)suse.de> wrote:
which PR/ticket are you referring to?
This thread
https://lists.ceph.io/hyperkitty/list/ceph-users@ceph.io/thread/QMHLZOAY7LW…
<https://lists.ceph.io/hyperkitty/list/ceph-users@ceph.io/thread/QMHLZOAY7LWGBYA5UK53SZSHLRTPTAQY/>
And this ticket
https://tracker.ceph.com/issues/48385 <https://tracker.ceph.com/issues/48385>
All pools show me stored == used, even if all osd in/up/weighted on 14.2.21
RAW STORAGE:
CLASS SIZE AVAIL USED RAW USED %RAW USED
hdd 595 TiB 334 TiB 261 TiB 261 TiB 43.87
nvme 52 TiB 34 TiB 18 TiB 18 TiB 34.68
TOTAL 647 TiB 368 TiB 279 TiB 279 TiB 43.13
POOLS:
POOL ID PGS STORED OBJECTS USED
%USED MAX AVAIL
replicated_rbd 1 1024 29 TiB 7.74M 29 TiB
10.93 80 TiB
erasure_rbd_meta 6 16 2.9 MiB 154 2.9 MiB
0 7.3 TiB
erasure_rbd_data 7 512 44 TiB 11.64M 44 TiB
15.61 144 TiB
.rgw.root 12 16 11 KiB 22 11 KiB
0 80 TiB
default.rgw.control 13 16 0 B 8 0 B
0 80 TiB
default.rgw.meta 14 16 45 KiB 211 45 KiB
0 7.3 TiB
default.rgw.log 15 16 88 MiB 345 88 MiB
0 80 TiB
default.rgw.buckets.index 16 16 6.3 GiB 717 6.3 GiB
0.03 7.3 TiB
default.rgw.buckets.non-ec 17 16 810 KiB 5.74k 810 KiB
0 80 TiB
default.rgw.buckets.data 18 512 15 TiB 33.99M 15 TiB
5.70 80 TiB
replicated_rbd_nvme 19 256 3.3 TiB 880.95k 3.3 TiB
13.25 7.3 TiB
unredundant_rbd_nvme 20 128 3.9 TiB 1.03M 3.9 TiB
15.27 11 TiB
fs_data 21 256 11 TiB 12.32M 11 TiB
4.38 80 TiB
fs_meta 22 16 4.8 GiB 636.79k 4.8 GiB
0.02 7.3 TiB
fs_data_nvme 23 64 20 GiB 5.97k 20 GiB
0.09 7.3 TiB
kubernetes_rbd 24 32 142 GiB 36.71k 142 GiB
0.06 80 TiB
Thanks,
k
20 May
20 May
8:22 p.m.
New subject: ceph df: pool stored vs bytes_used -- raw or not?
I can confirm that we still occasionally see stored==used even with
14.2.21, but I didn't have time yet to debug the pattern behind the
observations. I'll let you know if we find anything useful.
.. Dan
On Thu, May 20, 2021, 6:56 PM Konstantin Shalygin <k0ste(a)k0ste.ru> wrote:
And this ticket
https://tracker.ceph.com/issues/48385 <
https://tracker.ceph.com/issues/48385>
All pools show me stored == used, even if all osd in/up/weighted on 14.2.21
RAW STORAGE:
CLASS SIZE AVAIL USED RAW USED %RAW USED
hdd 595 TiB 334 TiB 261 TiB 261 TiB 43.87
nvme 52 TiB 34 TiB 18 TiB 18 TiB 34.68
TOTAL 647 TiB 368 TiB 279 TiB 279 TiB 43.13
POOLS:
POOL ID PGS STORED OBJECTS
USED %USED MAX AVAIL
replicated_rbd 1 1024 29 TiB 7.74M
29 TiB 10.93 80 TiB
erasure_rbd_meta 6 16 2.9 MiB 154
2.9 MiB 0 7.3 TiB
erasure_rbd_data 7 512 44 TiB 11.64M
44 TiB 15.61 144 TiB
.rgw.root 12 16 11 KiB 22
11 KiB 0 80 TiB
default.rgw.control 13 16 0 B 8
0 B 0 80 TiB
default.rgw.meta 14 16 45 KiB 211
45 KiB 0 7.3 TiB
default.rgw.log 15 16 88 MiB 345
88 MiB 0 80 TiB
default.rgw.buckets.index 16 16 6.3 GiB 717
6.3 GiB 0.03 7.3 TiB
default.rgw.buckets.non-ec 17 16 810 KiB 5.74k
810 KiB 0 80 TiB
default.rgw.buckets.data 18 512 15 TiB 33.99M
15 TiB 5.70 80 TiB
replicated_rbd_nvme 19 256 3.3 TiB 880.95k
3.3 TiB 13.25 7.3 TiB
unredundant_rbd_nvme 20 128 3.9 TiB 1.03M
3.9 TiB 15.27 11 TiB
fs_data 21 256 11 TiB 12.32M
11 TiB 4.38 80 TiB
fs_meta 22 16 4.8 GiB 636.79k
4.8 GiB 0.02 7.3 TiB
fs_data_nvme 23 64 20 GiB 5.97k
20 GiB 0.09 7.3 TiB
kubernetes_rbd 24 32 142 GiB 36.71k
142 GiB 0.06 80 TiB
Thanks,
k
_______________________________________________
ceph-users mailing list -- ceph-users(a)ceph.io
To unsubscribe send an email to ceph-users-leave(a)ceph.io
On 20 May 2021, at 19:47, Igor Fedotov
<ifedotov(a)suse.de> wrote:
which PR/ticket are you referring to?
This thread
https://lists.ceph.io/hyperkitty/list/ceph-users@ceph.io/thread/QMHLZOAY7LW…
<
https://lists.ceph.io/hyperkitty/list/ceph-users@ceph.io/thread/QMHLZOAY7LW…
9 Jun
9 Jun
7:39 a.m.
New subject: ceph df: pool stored vs bytes_used -- raw or not?
Stored==used was resolved for this cluster. Actually problem is what you was discover in
previous year: zero's. Filestore lack of META counter - always zero. When I purged
last drained OSD from cluster - statistics becomes to normal immediately
Thanks,
k
On 20 May 2021, at 21:22, Dan van der Ster
<dan(a)vanderster.com> wrote:
I can confirm that we still occasionally see stored==used even with 14.2.21, but I
didn't have time yet to debug the pattern behind the observations. I'll let you
know if we find anything useful.
12:20 p.m.
New subject: ceph df: pool stored vs bytes_used -- raw or not?
Should we fire another ticket for that?
On 6/9/2021 8:39 AM, Konstantin Shalygin wrote:
Stored==used was resolved for this cluster. Actually
problem is what
you was discover in previous year: zero's. Filestore lack of META
counter - always zero. When I purged last drained OSD from cluster -
statistics becomes to normal immediately
Thanks,
k
On 20 May 2021, at 21:22, Dan van der Ster
<dan(a)vanderster.com
<mailto:dan@vanderster.com>> wrote:
I can confirm that we still occasionally see stored==used even with
14.2.21, but I didn't have time yet to debug the pattern behind the
observations. I'll let you know if we find anything useful.
15 Jun
15 Jun
2:21 p.m.
New subject: ceph df: pool stored vs bytes_used -- raw or not?
Fired https://tracker.ceph.com/issues/51223
k
On 9 Jun 2021, at 13:20, Igor Fedotov
<ifedotov(a)suse.de> wrote:
Should we fire another ticket for that?
1040
days inactive
1066
days old
4 comments
3 participants
participants (3)
-
Dan van der Ster -
Igor Fedotov -
Konstantin Shalygin