very interesting LFX insights!
2 takeaways on PRs:
* gigantic PRs (+500 LoC) review process is probably less effective. less
time is spent on reviewing them, and they have less review comments than
large (100-499 LoC) PRs
this is sad, but also makes sense as people are reluctant to commit to such
l
There is no pool number in the crushmap. I tried to find out where and
how this pg_upmap_primary is stored exactly, but I couldn’t really
find anything useful. Since it is part of the ‚ceph osd dump‘ output,
I thought it might be in the osdmap, but none of the tools I tried
were giving me a
Hello Tobias,
Thank you so much for the helpful answer. In the meantime I found the
corresponding ticket (https://tracker.ceph.com/issues/64308).
So the fix will also be included in the upcoming Reef 18.2.5 release, which is
good news as our computing centre is planning to update to this version
Hello Markus,
Try using v17.2.8 that includes this change [1], please note that if you want
to upgrade
to Reef (v18.2.x) that same fix [2] is not yet released there, the fixed is
released in Squid v19.x
/Tobias
[1] https://github.com/ceph/ceph/pull/60458
[2] https://github.com/ceph/ceph/commit
Hello
Using boto3 trying to upload files to s3 bucket, when I am using boto version
1.35 it works fine, when upgrading it to 1.36 doesn’t work, facing below error,
python3 s3_upload.py
Error testing the new bucket: An error occurred (XAmzContentSHA256Mismatch)
when calling the PutObject opera
Hi,
skipping one release is supported. I upgraded a customer cluster a
month ago from 17.2.7 to 19.2.0. the only difference is that this is a
cephadm managed cluster.
Regards,
Eugen
Zitat von Simon Oosthoek :
Hi list,
we have a ceph cluster running ubuntu 22.04 and quincy packages from
Hi list,
We encountered a problem with presigned URLs for putting objects in connection
with CORS and S3 Object tagging in Ceph v17.2.7
It works fine with v 16.2.15 and it works fine if the tagging is disabled.
Here are the steps to reproduce:
1. Create a CORS rule for a bucket called "my-bucke
Hi Dev,
AWS recently updated their SDKs to enable CRC32 checksums on multiple
object operations by default. (boto >= 1.36.0)
This is a known issue and the reef backport has merged for v18.2.5 (not
released yet).
https://tracker.ceph.com/issues/64465
https://github.com/ceph/ceph/pull/58435
Becau
hi Dev,
On Tue, Feb 25, 2025 at 1:25 PM Devender Singh wrote:
>
> Hello
>
>
> Using boto3 trying to upload files to s3 bucket, when I am using boto version
> 1.35 it works fine, when upgrading it to 1.36 doesn’t work, facing below
> error,
>
> python3 s3_upload.py
> Error testing the new bucket
Hi list,
we have a ceph cluster running ubuntu 22.04 and quincy packages from ubuntu.
To keep up with the releases of Ceph, we want to upgrade to a more recent
stable release...
We could upgrade to ubuntu 24.04 LTS (re-install actually) but the ceph
version then skips 18.2.x and goes directly to
Ack. Another fine mess.
I was trying to clean things up and the process of tossing around OSD's
kept getting me reports of slow responses and hanging PG operations.
This is Ceph Pacific, by the way.
I found a deprecated server that claimed to have an OSD even though it
didn't show in either "cep
Hi,
as always, I looked briefly into the tracker but couldn't find
anything related (yet). Not sure if this has been reported or not, but
after I upgraded a customer cluster to Squid (19.2.0), Grafana was
also upgraded to version 9.4.12. And now the host details show all 126
OSDs for each
12 matches
Mail list logo
