Hi,
rbd list --pool erasure42 return a empty list.
that's because the rbd metadata is stored in a replicated pool. You
need to look into your replicated pool to delete the test image. If
you don't find it with:
rbd list --pool
you can inspect the rbd_data prefix of the EC chunks and
Hi,
is the balancer on? And which mode is enabled?
ceph balancer status
You definitely should split PGs, aim at 100 - 150 PGs per OSD at
first. I would inspect the PG sizes of the new OSDs:
ceph pg ls-by-osd 288 (column BYTES)
and compare them with older OSDs. If you have very large PG siz
Hey Brian,
The setting you're looking for is bluefs_buffered_io. This is very
much a YMMV setting, so it's best to test with both modes, but I
usually recommend turning it off for all but omap-intensive workloads
(e.g. RGW index) due to it causing writes to tend to be split up into
smaller pieces.
Before I replied, I wanted to renew my confidence and do a small test
in a lab environment. I also created a k4m2 pool with host as
failure-domain, started to write data chunks into it in a while loop
and then marked three of the OSDs "out" simultaneously. After a few
seconds of repeering b
Le 17/03/2025 à 13:46:58+, Eugen Block a écrit
Hi,
>
> that's because the rbd metadata is stored in a replicated pool. You need to
> look into your replicated pool to delete the test image. If you don't find
> it with:
Ok. I know that. But I didn't know I should remove the image metadata fr
600 MB/s is rather slow. With 10 GBit/s I regularly measure 1,28 GB/s bandwidth even with a single connection.The issue is latency not bandwidth!The latenc is bound by the CPU serving the osds when using decent NVMe storage is used.In an optimal world the network latency would be the limiting fact
Den mån 17 mars 2025 kl 14:48 skrev Joshua Baergen :
> Hey Brian,
>
> The setting you're looking for is bluefs_buffered_io. This is very
> much a YMMV setting, so it's best to test with both modes, but I
> usually recommend turning it off for all but omap-intensive workloads
> (e.g. RGW index) due
I have 6 ssd sata and 12 osd per server in a 24 server cluster. This
environment was created when it was in the natilus version.
I switched this environment to the Octopus version 6 months ago. The
cluster is working healthily.
I added 8 new servers, I created 6 ssd sata and 12 osd on these serve
Hello,
I am currently managing a Ceph cluster that consists of 3 racks, each with
4 OSD nodes. Each node contains 24 OSDs. I plan to add three new nodes, one
to each rack, to help alleviate the high OSD utilization.
The current highest OSD utilization is 85%. I am concerned about the
possibility
Hi Danish,
Have you tried restarting all RGWs or at least the one running the sync thread?
Unless the synchronization thread generates more errors than you have time to
clean up, you should be able to trim shard 31's error log entirely by looping
through all IDs (--marker).
Also, have you tr
Hi everyone,
In previous discussions, the Ceph Steering Committee tentatively agreed on
a Tentacle dev freeze around the end of March or mid-April. We would like
to revisit this and check in with all the tech leads to assess the
readiness level and ensure we're aligned on the timeline.
Please pro
Hi,
I got two pool one replica 3 and another with EC 4+2 for rbd.
During my test to access from a KVM hyperviseur I create a image with meta
data in the replica 3 pool and the data in the EC 4+2.
After I end the test (where everything works) I can't see anything in
erasure42 pool.
rbd lis
12 matches
Mail list logo
