Hi Frank,
The only time I've seen something like this was when we accidently changed
a subtree pin to a different rank, causing a huge amount of mds export work
to be queued up between MDSs. In that incident, we just waited until it
completed... It took around 30 minutes, after which all the log s
Guten Tag.
Our file system is out of operation (mimic 13.2.10). Our MDSes are choking on
an operation:
2021-09-19 02:23:36.432664 mon.ceph-01 mon.0 192.168.32.65:6789/0 185676 :
cluster [WRN] Health check failed: 1 MDSs repor
t slow requests (MDS_SLOW_REQUEST)
[...]
2021-09-19 02:23:34.909269 m
It looks similar to these:
https://tracker.ceph.com/issues/39987 [The user reporting this was me as well]
https://tracker.ceph.com/issues/42338
Issue 39987 was fixed a long time ago by Zheng Yan. A search for "currently
failed to authpin, subtree is being exported" only returns hits regarding th
Hi,
after upgrading the cluster from 16.2.5 to 16.2.6, several OSDs crashed
and refuse to start due to rocksdb corruption, eg:
2021-09-19T15:47:10.611+0200 7f8bc1f0e700 4 rocksdb:
[compaction/compaction_job.cc:1680] [default] Compaction start summary:
Base version 6 Base level 0,
Hi Dan,
thanks for looking at this. I had to take action and restarted 2 of our 4
active MDSes, which flushed out the stuck operations. I'm pretty sure it was a
real deadlock, most clients were blocked already. There was almost no fs
activity and MDS CPU usage was below 2%.
We are running a ve
I just deleted the rados object from .rgw.data.root and this removed the
bucket.instance, but this did not solve the problem.
It looks like there is some access error when I try to radosgw-admin
metadata sync init.
The 403 http response code on the post to the /admin/realm/period endpoint.
I chec
Hi,
You can arbitrarily add or remove the cache tier, there's no problem with
that. The problem is that cache tier doesn't work well, I tried it in front
of replicated and EC-pools with very mixed results: when it worked there
wasn't as much of a speed/latency benefit as one would expect from
NVME
Hello,
I'll start by explaining what I have done. I was adding some new storage
in attempt to setup a cache pool according to
https://docs.ceph.com/en/latest/dev/cache-pool/ by doing the following.
1. I upgraded all servers in cluster to ceph 15.2.14 which put the
system into recovery for ou
And we are quite happy with our cache tier. When we got new HDD OSDs
we tested if things would improve without the tier but we had to stick
to it, otherwise working with our VMs was almost impossible. But this
is an RBD cache so I can't tell how the other protocols perform with a
cache tier
Hi,
Yes! I did play with another cluster before and forgot to completely
clear that node! And the fsid "46e2b13c-dab7-11eb-810b-a5ea707f1ea1"
from that cluster. But then there is an error in CEPH. Because the
mon the existing cluster complained about (with fsid
"1ef45b26-dbac-11eb-a357-61
10 matches
Mail list logo