Hello, ceph-users.
Our mds servers keep segfaulting from a failed assertion, and for the first
time I can't find anyone else who's posted about this problem. None of them are
able to stay up, so our cephfs is down.
We recently had to truncate the journal log after an upgrade to nautilus, and
Currently I am getting the pool statistics (especially USED/MAX AVAIL) via the
command line:
ceph df -f json-pretty| jq '.pools[] | select(.name == "poolname") |
.stats.max_avail'
ceph df -f json-pretty| jq '.pools[] | select(.name == "poolname") |
.stats.bytes_used'
Command "ceph df" does not
On Fri, Oct 11, 2019 at 12:27 AM Kenneth Waegeman
wrote:
>
> Hi Brad, all,
>
> Pool 6 has min_size 2:
>
> pool 6 'metadata' replicated size 3 min_size 2 crush_rule 1 object_hash
> rjenkins pg_num 1024 pgp_num 1024 autoscale_mode warn last_change 172476
> flags hashpspool stripe_width 0
Hi, folks,
I have a middle-size Ceph cluster as cinder backup for openstack (queens).
Duing testing, one Ceph node went down unexpected and powered up again ca 10
minutes later, Ceph cluster starts PG recovery. To my surprise, VM IOPS drops
dramatically during Ceph recovery, from ca. 13K IOPS
Hi Brad, all,
Pool 6 has min_size 2:
pool 6 'metadata' replicated size 3 min_size 2 crush_rule 1 object_hash
rjenkins pg_num 1024 pgp_num 1024 autoscale_mode warn last_change 172476
flags hashpspool stripe_width 0 application cephfs
The output for all the inconsistent pgs is this:
{
I've also encountered this issue on a cluster yesterday; one CPU got
stuck in an infinite loop in get_obj_data::flush and it stopped
serving requests. I've updated the tracker issue accordingly.
Paul
--
Paul Emmerich
Looking for help with your Ceph cluster? Contact us at https://croit.io
Hi,
Currently I am getting the pool statistics (especially USED/MAX AVAIL) via the
command line:
ceph df -f json-pretty| jq '.pools[] | select(.name == "poolname") |
.stats.max_avail'
ceph df -f json-pretty| jq '.pools[] | select(.name == "poolname") |
.stats.bytes_used'
Command "ceph df" does
After trying to disable the paxos service trim temporarily (since that seemed
to trigger it initially), we now see this:
"assert_condition": "from != to",
"assert_func": "void PaxosService::trim(MonitorDBStore::TransactionRef,
version_t, version_t)",
"assert_file":
Thanks, Patrick. Looks like the fix is awaiting review, I guess my options
are to hold tight for 14.2.5 or patch myself if I get desperate. I've seen
this crash about 4 times over the past 96 hours, is there anything I can do
to mitigate the issue in the meantime?
On Wed, Oct 9, 2019 at 9:23 PM
Does pool 6 have min_size = 1 set?
https://tracker.ceph.com/issues/24994#note-5 would possibly be helpful
here, depending on what the output of the following command looks
like.
# rados list-inconsistent-obj [pgid] --format=json-pretty
On Thu, Oct 10, 2019 at 8:16 PM Kenneth Waegeman
wrote:
>
Hi all,
After some node failure and rebalancing, we have a lot of pg's in
inconsistent state. I tried to repair, but it din't work. This is also
in the logs:
2019-10-10 11:23:27.221 7ff54c9b0700 0 log_channel(cluster) log [DBG]
: 6.327 repair starts
2019-10-10 11:23:27.431 7ff5509b8700 -1
How do I Import an osdmap in Nautilus? I saw documentation for older version
but it seems one now can only export but not import anymore?
Am Donnerstag, 10. Oktober 2019, 08:52:03 OESZ hat Philippe D'Anjou
Folgendes geschrieben:
I dont think this has anything to do with CephFS, the
12 matches
Mail list logo
