[ceph-users] jewel 10.2.5 cephfs fsync write issue
hi guys: I am using kernel client (4.9) to mount cephfs (10.2.5) that just upgraded from Hammer(0.94.9). It became slow while doing fsync write VS same process in Hammer. (Yes, I am sure, fsync is the key.) 1. anyone knows what's goning on ? 2. any way to import that? === btw, actually, It's the same slow if I deploy a new cluster, with kernel client 4.2 or fuse client. ~ ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Issue with upgrade from 0.94.9 to 10.2.5
On 01/17/2017 12:52 PM, Piotr Dałek wrote:
During our testing we found out that during upgrade from 0.94.9 to 10.2.5
we're hitting issue http://tracker.ceph.com/issues/17386 ("Upgrading 0.94.6
-> 0.94.9 saturating mon node networking"). Apparently, there's a few
commits for both hammer and jewel which are supposed to fix this issue for
upgrades from 0.94.6 to 0.94.9 (and possibly for others), but we're still
seeing this upgrading to Jewel, and symptoms are exactly same - after
upgrading MONs, each not yet upgraded OSD takes full OSDMap from monitors
after failing the CRC check. Anyone else encountered this?
http://tracker.ceph.com/issues/18582
--
Piotr Dałek
piotr.da...@corp.ovh.com
https://www.ovh.com/us/
___
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] failing to respond to capability release, mds cache size?
On Tue, Jan 17, 2017 at 10:56 AM, Darrell Enns wrote: > Thanks for the info, I'll be sure to "dump_ops_in_flight" and "session ls" if > it crops up again. Is there any other info you can think of that might be > useful? I want to make sure I capture all the evidence needed if it happens > again. If you can dump the MDS cache that might turn out to be useful too; otherwise there's probably not much you can get unless your kernel includes dynamic debug and you can play with the /sys/kernel/debug/ceph that is described at http://docs.ceph.com/docs/master/cephfs/troubleshooting/#kernel-mount-debugging But I'll leave that up to Zheng or somebody who works on the kernel client. -Greg ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
[ceph-users] Ceph uses more raw space than expected
Hi all, I'm running a 6 node 24 OSD cluster, Jewel 10.2.5 with kernel 4.8. I put about 1TB of data in the cluster, with all pools having size 3. Yet about 5TB of raw disk is used as opposed to the expected 3TB. result of ceph -s: pgmap v1057361: 2400 pgs, 3 pools, 984 GB data, 125 Mobjects 5039 GB used, 12353 GB / 17393 GB avail 2398 active+clean 2 active+clean+scrubbing Result of ceph df: GLOBAL: SIZE AVAIL RAW USED %RAW USED 17393G 12353G5039G 28.97 POOLS: NAMEID USED %USED MAX AVAIL OBJECTS rbd 1801G 17.12 3880G206299 cephfs_data 2182G 4.49 3880G 130875410 cephfs_metadata 3 32540k 0 3880G201555 Result of ceph osd dump: pool 1 'rbd' replicated size 3 min_size 1 crush_ruleset 0 object_hash rjenkins pg_num 800 pgp_num 800 last_change 482 flags hashpspool stripe_width 0 pool 2 'cephfs_data' replicated size 3 min_size 1 crush_ruleset 0 object_hash rjenkins pg_num 800 pgp_num 800 last_change 410 flags hashpspool crash_replay_interval 45 stripe_width 0 pool 3 'cephfs_metadata' replicated size 3 min_size 1 crush_ruleset 0 object_hash rjenkins pg_num 800 pgp_num 800 last_change 408 flags hashpspool stripe_width 0 The cluster was set up with ceph-deploy. So each OSD drive is formatted with xfs. One more thing I should mention is that I'm using the experimental directory fragmentation feature in cephfs. After scouring the mailing list I've found this other post that seems to be related or the same issue: http://lists.ceph.com/pipermail/ceph-users-ceph.com/2013-October/004901.html Does anyone know if this is a bug or is it legitimate overhead that I failed to account for? Thanks, Pavel ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
[ceph-users] RadosGW Performance on Copy
Performance related data I found so far on RadosGW don't contain anything on Copy operation. Can anyone comment on the Copy operation? Is it basically just a GET + INSERT? Thanks, Eric ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
[ceph-users] GSOC 2017 Submissions Open Tomorrow
Hey cephers, Just wanted to send a friendly reminder that the Google Summer of Code program has their submissions window opening tomorrow, and will remain open for 3 weeks. If you are interested in mentoring or suggesting a project please contact me as soon as possible. Thanks. -- Best Regards, Patrick McGarry Director Ceph Community || Red Hat http://ceph.com || http://community.redhat.com @scuttlemonkey || @ceph ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
[ceph-users] rgw static website docs 404
Are there docs on the RGW static website feature? I found 'rgw enable static website' config setting only via the mailing list. A search for 'static' on ceph.com turns up release notes, but no other documentation. Anyone have pointers on how to set this up and what i can do with it? Does it require using dns based buckets, for example? I'd like to be able to hit a website with http:, ideally. (without the browser forcing it to download) thanks, -Ben ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] rgw static website docs 404
Aha! Found some docs here in the RHCS site: https://access.redhat.com/documentation/en/red-hat-ceph-storage/2/paged/object-gateway-guide-for-red-hat-enterprise-linux/chapter-2-configuration Really, ceph.com should have all this too... -Ben On Wed, Jan 18, 2017 at 5:15 PM, Ben Hines wrote: > Are there docs on the RGW static website feature? > > I found 'rgw enable static website' config setting only via the mailing > list. A search for 'static' on ceph.com turns up release notes, but no > other documentation. Anyone have pointers on how to set this up and what i > can do with it? Does it require using dns based buckets, for example? I'd > like to be able to hit a website with http:, > ideally. (without the browser forcing it to download) > > thanks, > > -Ben > ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
[ceph-users] Does this indicate a "CPU bottleneck"?
Hi, everyone. Recently, we did some stress test on ceph using three machines. We tested the IOPS of the whole small cluster when there are 1~8 OSDs per machines separately and the result is as follows: OSD num per machine fio iops 1 10k 2 16.5k 3 22k 4 23.5k 5 26k 6 27k 7 27k 8 28k As shown above, it seems that there is some kind of bottleneck when there are more than 4 OSDs per machine. Meanwhile, we observed that the CPU %idle during the test, shown below, has also some kind of correlation with the number of OSDs per machine. OSD num per machine CPU idle 1 74% 2 52% 3 30% 4 25% 5 24% 6 17% 7 14% 8 11% It seems that with the number of OSDs per machine increasing, the CPU idle time is reducing and the reduce rate Is also decreasing, can we come to the conclusion that CPU is the performance bottleneck in this test? Thank you:) ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
