[ceph-users] troubleshooting ceph rdma performance
Hello All, I have been collecting performance numbers on our ceph cluster, and I had noticed a very poor throughput on ceph async+rdma when compared with tcp. I was wondering what tunings/settings should I do to the cluster that would improve the ceph rdma (async+rdma) performance. Currently, from what we see: Ceph rdma throughput is less than half of the ceph tcp throughput (ran fio over iscsi mounted disks). Our ceph cluster has 8 nodes and configured with two networks, cluster and client networks. Can someone please shed some light. I'd be glad to provide any further information regarding the setup. Thanks in Advance, Raju ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] permission errors rolling back ceph cluster to v13
Thanks Greg. I think I have to re-install ceph v13 from scratch then. -Raju From: Gregory Farnum Sent: 09 August 2018 01:54 To: Raju Rangoju Cc: ceph-users@lists.ceph.com Subject: Re: [ceph-users] permission errors rolling back ceph cluster to v13 On Tue, Aug 7, 2018 at 6:27 PM Raju Rangoju mailto:ra...@chelsio.com>> wrote: Hi, I have been running into some connection issues with the latest ceph-14 version, so we thought the feasible solution would be to roll back the cluster to previous version (ceph-13.0.1) where things are known to work properly. I’m wondering if rollback/downgrade is supported at all ? After compiling/starting ceph-13 I’m running into a permission error. Basically it complains about the incompatibility of disk layout (ceph-13 mimic vs ceph-14 nautilus) 2018-08-07 10:41:00.580 2b391528e080 -1 ERROR: on disk data includes unsupported features: compat={},rocompat={},incompat={11=nautilus ondisk layout} 2018-08-07 10:41:00.580 2b391528e080 -1 error checking features: (1) Operation not permitted 2018-08-07 10:41:11.161 2b16a7d14080 0 set uid:gid to 167:167 (ceph:ceph) 2018-08-07 10:41:11.161 2b16a7d14080 0 ceph version 13.0.1-3266-g6b59fbf (6b59fbfcc6bbfd67193e1c1e142b478ddd68aab4) mimic (dev), process (unknown), pid 14013 2018-08-07 10:41:11.161 2b16a7d14080 0 pidfile_write: ignore empty --pid-file I thought allowing permissions to mon would fix it (see below), but apparently ceph command hangs. So it didn’t allow permissions. ceph auth add osd.44 osd 'allow *' mon 'allow profile osd' -i /var/lib/ceph/osd/ceph-44/keyring [root@hadoop1 my-ceph]# ceph -s 2018-08-07 10:59:59.325 2b5c26347700 0 monclient(hunting): authenticate timed out after 300 2018-08-07 10:59:59.325 2b5c26347700 0 librados: client.admin authentication error (110) Connection timed out [errno 110] error connecting to the cluster Has anyone tried ceph rollback before? Any help is greatly appreciated. Unfortunately this is generally not possible. Disk encodings change across major versions and the old code can't understand what's been written down once the new code runs. -Greg Thanks, Raj ___ ceph-users mailing list ceph-users@lists.ceph.com<mailto:ceph-users@lists.ceph.com> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
[ceph-users] permission errors rolling back ceph cluster to v13
Hi, I have been running into some connection issues with the latest ceph-14 version, so we thought the feasible solution would be to roll back the cluster to previous version (ceph-13.0.1) where things are known to work properly. I'm wondering if rollback/downgrade is supported at all ? After compiling/starting ceph-13 I'm running into a permission error. Basically it complains about the incompatibility of disk layout (ceph-13 mimic vs ceph-14 nautilus) 2018-08-07 10:41:00.580 2b391528e080 -1 ERROR: on disk data includes unsupported features: compat={},rocompat={},incompat={11=nautilus ondisk layout} 2018-08-07 10:41:00.580 2b391528e080 -1 error checking features: (1) Operation not permitted 2018-08-07 10:41:11.161 2b16a7d14080 0 set uid:gid to 167:167 (ceph:ceph) 2018-08-07 10:41:11.161 2b16a7d14080 0 ceph version 13.0.1-3266-g6b59fbf (6b59fbfcc6bbfd67193e1c1e142b478ddd68aab4) mimic (dev), process (unknown), pid 14013 2018-08-07 10:41:11.161 2b16a7d14080 0 pidfile_write: ignore empty --pid-file I thought allowing permissions to mon would fix it (see below), but apparently ceph command hangs. So it didn't allow permissions. ceph auth add osd.44 osd 'allow *' mon 'allow profile osd' -i /var/lib/ceph/osd/ceph-44/keyring [root@hadoop1 my-ceph]# ceph -s 2018-08-07 10:59:59.325 2b5c26347700 0 monclient(hunting): authenticate timed out after 300 2018-08-07 10:59:59.325 2b5c26347700 0 librados: client.admin authentication error (110) Connection timed out [errno 110] error connecting to the cluster Has anyone tried ceph rollback before? Any help is greatly appreciated. Thanks, Raj ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] issues with ceph nautilus version
Hey Igor, patch that you pointed worked for me. Thanks Again. From: ceph-users On Behalf Of Igor Fedotov Sent: 20 June 2018 21:55 To: ceph-users@lists.ceph.com Subject: Re: [ceph-users] issues with ceph nautilus version Hi Raju, This is a bug in new BlueStore's bitmap allocator. This PR will most probably fix that: https://github.com/ceph/ceph/pull/22610 Also you may try to switch bluestore and bluefs allocators (bluestore_allocator and bluefs_allocator parameters respectively) to stupid and restart OSDs. This should help. Thanks, Igor On 6/20/2018 6:41 PM, Raju Rangoju wrote: Hi, Recently I have upgraded my ceph cluster to version 14.0.0 - nautilus(dev) from ceph version 13.0.1, after this, I noticed some weird data usage numbers on the cluster. Here are the issues I'm seeing... 1. The data usage reported is much more than what is available usage: 16 EiB used, 164 TiB / 158 TiB avail before this upgradation, it used to report properly usage: 1.10T used, 157T / 158T avail 1. it reports that all the osds/pool are full Can someone please shed some light? Any helps is greatly appreciated. [root@hadoop1 my-ceph]# ceph --version ceph version 14.0.0-480-g6c1e8ee (6c1e8ee14f9b25dc96684dbc1f8c8255c47f0bb9) nautilus (dev) [root@hadoop1 my-ceph]# ceph -s cluster: id: ee4660fd-167b-42e6-b27b-126526dab04d health: HEALTH_ERR 87 full osd(s) 11 pool(s) full services: mon: 3 daemons, quorum hadoop1,hadoop4,hadoop6 mgr: hadoop6(active), standbys: hadoop1, hadoop4 mds: cephfs-1/1/1 up {0=hadoop3=up:creating}, 2 up:standby osd: 88 osds: 87 up, 87 in data: pools: 11 pools, 32588 pgs objects: 0 objects, 0 B usage: 16 EiB used, 164 TiB / 158 TiB avail pgs: 32588 active+clean Thanks in advance -Raj ___ ceph-users mailing list ceph-users@lists.ceph.com<mailto:ceph-users@lists.ceph.com> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] issues with ceph nautilus version
Hi Igor, Great! Thanks for the quick response. Will try the fix and let you know how it goes. -Raj From: ceph-users On Behalf Of Igor Fedotov Sent: 20 June 2018 21:55 To: ceph-users@lists.ceph.com Subject: Re: [ceph-users] issues with ceph nautilus version Hi Raju, This is a bug in new BlueStore's bitmap allocator. This PR will most probably fix that: https://github.com/ceph/ceph/pull/22610 Also you may try to switch bluestore and bluefs allocators (bluestore_allocator and bluefs_allocator parameters respectively) to stupid and restart OSDs. This should help. Thanks, Igor On 6/20/2018 6:41 PM, Raju Rangoju wrote: Hi, Recently I have upgraded my ceph cluster to version 14.0.0 - nautilus(dev) from ceph version 13.0.1, after this, I noticed some weird data usage numbers on the cluster. Here are the issues I'm seeing... 1. The data usage reported is much more than what is available usage: 16 EiB used, 164 TiB / 158 TiB avail before this upgradation, it used to report properly usage: 1.10T used, 157T / 158T avail 1. it reports that all the osds/pool are full Can someone please shed some light? Any helps is greatly appreciated. [root@hadoop1 my-ceph]# ceph --version ceph version 14.0.0-480-g6c1e8ee (6c1e8ee14f9b25dc96684dbc1f8c8255c47f0bb9) nautilus (dev) [root@hadoop1 my-ceph]# ceph -s cluster: id: ee4660fd-167b-42e6-b27b-126526dab04d health: HEALTH_ERR 87 full osd(s) 11 pool(s) full services: mon: 3 daemons, quorum hadoop1,hadoop4,hadoop6 mgr: hadoop6(active), standbys: hadoop1, hadoop4 mds: cephfs-1/1/1 up {0=hadoop3=up:creating}, 2 up:standby osd: 88 osds: 87 up, 87 in data: pools: 11 pools, 32588 pgs objects: 0 objects, 0 B usage: 16 EiB used, 164 TiB / 158 TiB avail pgs: 32588 active+clean Thanks in advance -Raj ___ ceph-users mailing list ceph-users@lists.ceph.com<mailto:ceph-users@lists.ceph.com> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
[ceph-users] issues with ceph nautilus version
Hi, Recently I have upgraded my ceph cluster to version 14.0.0 - nautilus(dev) from ceph version 13.0.1, after this, I noticed some weird data usage numbers on the cluster. Here are the issues I'm seeing... 1. The data usage reported is much more than what is available usage: 16 EiB used, 164 TiB / 158 TiB avail before this upgradation, it used to report properly usage: 1.10T used, 157T / 158T avail 1. it reports that all the osds/pool are full Can someone please shed some light? Any helps is greatly appreciated. [root@hadoop1 my-ceph]# ceph --version ceph version 14.0.0-480-g6c1e8ee (6c1e8ee14f9b25dc96684dbc1f8c8255c47f0bb9) nautilus (dev) [root@hadoop1 my-ceph]# ceph -s cluster: id: ee4660fd-167b-42e6-b27b-126526dab04d health: HEALTH_ERR 87 full osd(s) 11 pool(s) full services: mon: 3 daemons, quorum hadoop1,hadoop4,hadoop6 mgr: hadoop6(active), standbys: hadoop1, hadoop4 mds: cephfs-1/1/1 up {0=hadoop3=up:creating}, 2 up:standby osd: 88 osds: 87 up, 87 in data: pools: 11 pools, 32588 pgs objects: 0 objects, 0 B usage: 16 EiB used, 164 TiB / 158 TiB avail pgs: 32588 active+clean Thanks in advance -Raj ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
[ceph-users] issue with OSD class path in RDMA mode
Hello, I'm trying to run iscsi tgtd on ceph cluster. When do 'rbd list' I see below errors. [root@ceph1 ceph]# rbd list 2018-05-30 18:19:02.227 2ae7260a8140 -1 librbd::api::Image: list_images: error listing image in directory: (5) Input/output error 2018-05-30 18:19:02.227 2ae7260a8140 -1 librbd: error listing v2 images: (5) Input/output error rbd: list: (5) Input/output error I have followed '/ceph/doc/dev/osd-class-path.rst' [1] and found that this is an issue with OSD class path, and I have added the osd_class_dir to ceph.conf. After adding osd class path, 'rbd list' worked perfectly fine in TCP mode and no issues were seen. But in RDMA(async) mode, 26 out of 88 osds are down. No errors are seen in osd logs. Can someone please shed some light? [root@hadoop1 my-ceph]# ceph -s cluster: id: ee4660fd-167b-42e6-b27b-126526dab04d health: HEALTH_WARN 1 filesystem is degraded 39 osds down 4 hosts (44 osds) down Reduced data availability: 23944 pgs inactive services: mon: 3 daemons, quorum hadoop1,hadoop4,hadoop6 mgr: hadoop1(active), standbys: hadoop6, hadoop4 mds: cephfs-1/1/1 up {0=hadoop3=up:replay}, 2 up:standby osd: 88 osds: 26 up, 88 in data: pools: 13 pools, 23944 pgs objects: 0 objects, 0 usage: 0 used, 0 / 0 avail pgs: 100.000% pgs unknown 23944 unknown I am more than happy to provide any extra information necessary. [1] /root/ceph/doc/dev/osd-class-path.rst === OSD class path issues === :: 2011-12-05 17:41:00.994075 7ffe8b5c3760 librbd: failed to assign a block name for image create error: error 5: Input/output error This usually happens because your OSDs can't find ``cls_rbd.so``. They search for it in ``osd_class_dir``, which may not be set correctly by default (http://tracker.ceph.com/issues/1722). Most likely it's looking in ``/usr/lib/rados-classes`` instead of ``/usr/lib64/rados-classes`` - change ``osd_class_dir`` in your ``ceph.conf`` and restart the OSDs to fix it. Thanks, Raju ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
[ceph-users] exporting cephfs as nfs share on RDMA transport
Hi, I am testing CEPH over RDMA, for one of the tests I had to export ceph filesystem as NFS share on RDMA transport. For TCP transport, I used ganesha as NFS server that runs in user space and supports the cephFS FSAL using libcephfs, and it worked perfectly fine. However, my requirement was to export cephFS as NFS share on RDMA transport instead of TCP. So, I was wondering if there is something like ganesha to achieve the same on RDMA transport. Can someone please provide some pointers on this? Appreciate any help. Thanks, Raju ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com