Re: S3website range requests - possible issue
I was checking the tracker again and I found already fixed issue that seems to be connected with this issue. https://tracker.ceph.com/issues/44508 Here is the PR that fixes it https://github.com/ceph/ceph/pull/33807 What I’m still not understanding is why this is only happening when using s3website api. Is there someone who could shed some light on this? Regards, Ondrej
hey Ondrej, thanks for creating the tracker issue https://tracker.ceph.com/issues/62938. i added a comment there, and opened a fix in https://github.com/ceph/ceph/pull/53602 for the only issue i was able to identify On Wed, Sep 20, 2023 at 9:20 PM Ondřej Kukla <ondrej.kukla@cdn77.com> wrote:
I was checking the tracker again and I found already fixed issue that seems to be connected with this issue.
https://tracker.ceph.com/issues/44508
Here is the PR that fixes it https://github.com/ceph/ceph/pull/33807
What I’m still not understanding is why this is only happening when using s3website api.
Is there someone who could shed some light on this?
Regards,
Ondrej _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
Hello Casey, Thanks a lot for that. I’ve forgot to mention that in my previous message that I was able to trigger the prefetch by header bytes=1-10 You can see the the read 1~10 in the osd logs I’ve sent here - https://pastebin.com/nGQw4ugd Which is wierd as it seems that it is not the same you were able to replicate. Ondrej
On 22. 9. 2023, at 21:52, Casey Bodley <cbodley@redhat.com> wrote:
hey Ondrej,
thanks for creating the tracker issue https://tracker.ceph.com/issues/62938. i added a comment there, and opened a fix in https://github.com/ceph/ceph/pull/53602 for the only issue i was able to identify
On Wed, Sep 20, 2023 at 9:20 PM Ondřej Kukla <ondrej.kukla@cdn77.com> wrote:
I was checking the tracker again and I found already fixed issue that seems to be connected with this issue.
https://tracker.ceph.com/issues/44508
Here is the PR that fixes it https://github.com/ceph/ceph/pull/33807
What I’m still not understanding is why this is only happening when using s3website api.
Is there someone who could shed some light on this?
Regards,
Ondrej _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
that first "read 0~4194304" is probably what i fixed in https://github.com/ceph/ceph/pull/53602, but it's hard to tell from osd log where these osd ops are coming from. why are there several [read 1~10] requests after that? the rgw log would be more useful for debugging, with --debug-rgw=20 and --debug-ms=1 to show the osd ops/replies On Fri, Sep 22, 2023 at 4:00 PM Ondřej Kukla <ondrej.kukla@cdn77.com> wrote:
Hello Casey,
Thanks a lot for that.
I’ve forgot to mention that in my previous message that I was able to trigger the prefetch by header bytes=1-10
You can see the the read 1~10 in the osd logs I’ve sent here - https://pastebin.com/nGQw4ugd
Which is wierd as it seems that it is not the same you were able to replicate.
Ondrej
On 22. 9. 2023, at 21:52, Casey Bodley <cbodley@redhat.com> wrote:
hey Ondrej,
thanks for creating the tracker issue https://tracker.ceph.com/issues/62938. i added a comment there, and opened a fix in https://github.com/ceph/ceph/pull/53602 for the only issue i was able to identify
On Wed, Sep 20, 2023 at 9:20 PM Ondřej Kukla <ondrej.kukla@cdn77.com> wrote:
I was checking the tracker again and I found already fixed issue that seems to be connected with this issue.
https://tracker.ceph.com/issues/44508
Here is the PR that fixes it https://github.com/ceph/ceph/pull/33807
What I’m still not understanding is why this is only happening when using s3website api.
Is there someone who could shed some light on this?
Regards,
Ondrej _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
Hello Casey, Thanks a lot for that. I’ve forgot to mention that in my previous message that I was able to trigger the prefetch by header bytes=1-10 You can see the the read 1~10 in the osd logs I’ve sent here - https://pastebin.com/nGQw4ugd Which is wierd as it seems that it is not the same you were able to replicate. Ondrej
On 22. 9. 2023, at 21:52, Casey Bodley <cbodley@redhat.com> wrote:
hey Ondrej,
thanks for creating the tracker issue https://tracker.ceph.com/issues/62938. i added a comment there, and opened a fix in https://github.com/ceph/ceph/pull/53602 for the only issue i was able to identify
On Wed, Sep 20, 2023 at 9:20 PM Ondřej Kukla <ondrej.kukla@cdn77.com> wrote:
I was checking the tracker again and I found already fixed issue that seems to be connected with this issue.
https://tracker.ceph.com/issues/44508
Here is the PR that fixes it https://github.com/ceph/ceph/pull/33807
What I’m still not understanding is why this is only happening when using s3website api.
Is there someone who could shed some light on this?
Regards,
Ondrej _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
participants (3)
-
Casey Bodley
-
Ondřej Kukla
-
Ondřej Kukla