Re: [ceph-users] Ceph RBD 0.78 Bug or feature?
On Fri, Apr 4, 2014 at 10:47 PM, Gruher, Joseph R wrote: > Meant to include this - what do these messages indicate? All systems have > 0.78. > > > > [1301268.557820] Key type ceph registered > > [1301268.558524] libceph: loaded (mon/osd proto 15/24) > > [1301268.579486] rbd: loaded rbd (rados block device) > > [1301268.582364] libceph: mon1 10.0.0.102:6789 feature set mismatch, my > 4a042a42 < server's 104a042a42, missing 10 That's CRUSH_V2 feature bit, it's supported starting with 3.13. Thanks, Ilya ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Ceph RBD 0.78 Bug or feature?
Aha – upgrade of kernel from 3.13 to 3.14 appears to have resolved the problem. Thanks, Joe From: Gruher, Joseph R Sent: Friday, April 04, 2014 11:48 AM To: Ирек Фасихов; Ilya Dryomov Cc: ceph-users@lists.ceph.com; Gruher, Joseph R Subject: RE: [ceph-users] Ceph RBD 0.78 Bug or feature? Meant to include this – what do these messages indicate? All systems have 0.78. [1301268.557820] Key type ceph registered [1301268.558524] libceph: loaded (mon/osd proto 15/24) [1301268.579486] rbd: loaded rbd (rados block device) [1301268.582364] libceph: mon1 10.0.0.102:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301268.582462] libceph: mon1 10.0.0.102:6789 socket error on read [1301278.589461] libceph: mon1 10.0.0.102:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301278.589558] libceph: mon1 10.0.0.102:6789 socket error on read [1301288.607615] libceph: mon1 10.0.0.102:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301288.607713] libceph: mon1 10.0.0.102:6789 socket error on read [1301298.625873] libceph: mon1 10.0.0.102:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301298.625970] libceph: mon1 10.0.0.102:6789 socket error on read [1301308.643936] libceph: mon0 10.0.0.101:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301308.644033] libceph: mon0 10.0.0.101:6789 socket error on read [1301318.662082] libceph: mon0 10.0.0.101:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301318.662179] libceph: mon0 10.0.0.101:6789 socket error on read [1301449.695232] libceph: mon0 10.0.0.101:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301449.695329] libceph: mon0 10.0.0.101:6789 socket error on read [1301459.716235] libceph: mon1 10.0.0.102:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301459.716332] libceph: mon1 10.0.0.102:6789 socket error on read [1301469.734425] libceph: mon1 10.0.0.102:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301469.734523] libceph: mon1 10.0.0.102:6789 socket error on read [1301479.752603] libceph: mon1 10.0.0.102:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301479.752700] libceph: mon1 10.0.0.102:6789 socket error on read [1301489.770773] libceph: mon1 10.0.0.102:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301489.770870] libceph: mon1 10.0.0.102:6789 socket error on read [1301499.788904] libceph: mon1 10.0.0.102:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301499.789001] libceph: mon1 10.0.0.102:6789 socket error on read $ ceph --version ceph version 0.78 (f6c746c314d7b87b8419b6e584c94bfe4511dbd4) $ ssh mohonpeak01 'ceph --version' ceph version 0.78 (f6c746c314d7b87b8419b6e584c94bfe4511dbd4) $ ssh mohonpeak02 'ceph --version' ceph version 0.78 (f6c746c314d7b87b8419b6e584c94bfe4511dbd4) $ ceph health detail HEALTH_WARN noscrub,nodeep-scrub flag(s) set noscrub,nodeep-scrub flag(s) set $ ceph status cluster b12ebb71-e4a6-41fa-8246-71cbfa09fb6e health HEALTH_WARN noscrub,nodeep-scrub flag(s) set monmap e1: 2 mons at {mohonpeak01=10.0.0.101:6789/0,mohonpeak02=10.0.0.102:6789/0}, election epoch 10, quorum 0,1 mohonpeak01,mohonpeak02 osdmap e216: 18 osds: 18 up, 18 in flags noscrub,nodeep-scrub pgmap v202112: 2784 pgs, 10 pools, 1637 GB data, 427 kobjects 2439 GB used, 12643 GB / 15083 GB avail 2784 active+clean From: Gruher, Joseph R Sent: Friday, April 04, 2014 11:44 AM To: 'Ирек Фасихов'; Ilya Dryomov Cc: ceph-users@lists.ceph.com<mailto:ceph-users@lists.ceph.com>; Gruher, Joseph R Subject: RE: [ceph-users] Ceph RBD 0.78 Bug or feature? Hi folks- Was this ever resolved? I’m not finding a resolution in the email chain, apologies if I am missing it. I am experiencing this same problem. Cluster works fine for object traffic, can’t seem to get rbd to work in 0.78. Worked fine in 0.72.2 for me. Running Ubuntu 13.04 with 3.12 kernel. $ rbd create rbd/myimage --size 102400 $ sudo rbd map rbd/myimage rbd: add failed: (5) Input/output error $ rbd ls rbd myimage $ Thanks, Joe From: ceph-users-boun...@lists.ceph.com<mailto:ceph-users-boun...@lists.ceph.com> [mailto:ceph-users-boun...@lists.ceph.com] On Behalf Of ??? Sent: Tuesday, March 25, 2014 1:59 AM To: Ilya Dryomov Cc: ceph-users@lists.ceph.com<mailto:ceph-users@lists.ceph.com> Subject: Re: [ceph-users] Ceph RBD 0.78 Bug or feature? Ilya, set "chooseleaf_vary_r 0", but no map rbd images. [root@ceph01 cluster]# rbd map rbd/tst 2014-03-25 12:48:14.318167 7f4
Re: [ceph-users] Ceph RBD 0.78 Bug or feature?
Meant to include this – what do these messages indicate? All systems have 0.78. [1301268.557820] Key type ceph registered [1301268.558524] libceph: loaded (mon/osd proto 15/24) [1301268.579486] rbd: loaded rbd (rados block device) [1301268.582364] libceph: mon1 10.0.0.102:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301268.582462] libceph: mon1 10.0.0.102:6789 socket error on read [1301278.589461] libceph: mon1 10.0.0.102:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301278.589558] libceph: mon1 10.0.0.102:6789 socket error on read [1301288.607615] libceph: mon1 10.0.0.102:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301288.607713] libceph: mon1 10.0.0.102:6789 socket error on read [1301298.625873] libceph: mon1 10.0.0.102:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301298.625970] libceph: mon1 10.0.0.102:6789 socket error on read [1301308.643936] libceph: mon0 10.0.0.101:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301308.644033] libceph: mon0 10.0.0.101:6789 socket error on read [1301318.662082] libceph: mon0 10.0.0.101:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301318.662179] libceph: mon0 10.0.0.101:6789 socket error on read [1301449.695232] libceph: mon0 10.0.0.101:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301449.695329] libceph: mon0 10.0.0.101:6789 socket error on read [1301459.716235] libceph: mon1 10.0.0.102:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301459.716332] libceph: mon1 10.0.0.102:6789 socket error on read [1301469.734425] libceph: mon1 10.0.0.102:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301469.734523] libceph: mon1 10.0.0.102:6789 socket error on read [1301479.752603] libceph: mon1 10.0.0.102:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301479.752700] libceph: mon1 10.0.0.102:6789 socket error on read [1301489.770773] libceph: mon1 10.0.0.102:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301489.770870] libceph: mon1 10.0.0.102:6789 socket error on read [1301499.788904] libceph: mon1 10.0.0.102:6789 feature set mismatch, my 4a042a42 < server's 104a042a42, missing 10 [1301499.789001] libceph: mon1 10.0.0.102:6789 socket error on read $ ceph --version ceph version 0.78 (f6c746c314d7b87b8419b6e584c94bfe4511dbd4) $ ssh mohonpeak01 'ceph --version' ceph version 0.78 (f6c746c314d7b87b8419b6e584c94bfe4511dbd4) $ ssh mohonpeak02 'ceph --version' ceph version 0.78 (f6c746c314d7b87b8419b6e584c94bfe4511dbd4) $ ceph health detail HEALTH_WARN noscrub,nodeep-scrub flag(s) set noscrub,nodeep-scrub flag(s) set $ ceph status cluster b12ebb71-e4a6-41fa-8246-71cbfa09fb6e health HEALTH_WARN noscrub,nodeep-scrub flag(s) set monmap e1: 2 mons at {mohonpeak01=10.0.0.101:6789/0,mohonpeak02=10.0.0.102:6789/0}, election epoch 10, quorum 0,1 mohonpeak01,mohonpeak02 osdmap e216: 18 osds: 18 up, 18 in flags noscrub,nodeep-scrub pgmap v202112: 2784 pgs, 10 pools, 1637 GB data, 427 kobjects 2439 GB used, 12643 GB / 15083 GB avail 2784 active+clean From: Gruher, Joseph R Sent: Friday, April 04, 2014 11:44 AM To: 'Ирек Фасихов'; Ilya Dryomov Cc: ceph-users@lists.ceph.com; Gruher, Joseph R Subject: RE: [ceph-users] Ceph RBD 0.78 Bug or feature? Hi folks- Was this ever resolved? I’m not finding a resolution in the email chain, apologies if I am missing it. I am experiencing this same problem. Cluster works fine for object traffic, can’t seem to get rbd to work in 0.78. Worked fine in 0.72.2 for me. Running Ubuntu 13.04 with 3.12 kernel. $ rbd create rbd/myimage --size 102400 $ sudo rbd map rbd/myimage rbd: add failed: (5) Input/output error $ rbd ls rbd myimage $ Thanks, Joe From: ceph-users-boun...@lists.ceph.com [mailto:ceph-users-boun...@lists.ceph.com] On Behalf Of ???? ??? Sent: Tuesday, March 25, 2014 1:59 AM To: Ilya Dryomov Cc: ceph-users@lists.ceph.com Subject: Re: [ceph-users] Ceph RBD 0.78 Bug or feature? Ilya, set "chooseleaf_vary_r 0", but no map rbd images. [root@ceph01 cluster]# rbd map rbd/tst 2014-03-25 12:48:14.318167 7f44717f7760 2 auth: KeyRing::load: loaded key file /etc/ceph/ceph.client.admin.keyring rbd: add failed: (5) Input/output error [root@ceph01 cluster]# cat /var/log/messages | tail Mar 25 12:45:06 ceph01 kernel: libceph: osdc handle_map corrupt msg Mar 25 12:45:06 ceph01 kernel: libceph: mon2 192.168.100.203:6789<http://192.168.100.203:6789> session established Mar 25 12:46:33 ceph01 kernel: libceph: client1124
Re: [ceph-users] Ceph RBD 0.78 Bug or feature?
Hi folks- Was this ever resolved? I’m not finding a resolution in the email chain, apologies if I am missing it. I am experiencing this same problem. Cluster works fine for object traffic, can’t seem to get rbd to work in 0.78. Worked fine in 0.72.2 for me. Running Ubuntu 13.04 with 3.12 kernel. $ rbd create rbd/myimage --size 102400 $ sudo rbd map rbd/myimage rbd: add failed: (5) Input/output error $ rbd ls rbd myimage $ Thanks, Joe From: ceph-users-boun...@lists.ceph.com [mailto:ceph-users-boun...@lists.ceph.com] On Behalf Of ??? Sent: Tuesday, March 25, 2014 1:59 AM To: Ilya Dryomov Cc: ceph-users@lists.ceph.com Subject: Re: [ceph-users] Ceph RBD 0.78 Bug or feature? Ilya, set "chooseleaf_vary_r 0", but no map rbd images. [root@ceph01 cluster]# rbd map rbd/tst 2014-03-25 12:48:14.318167 7f44717f7760 2 auth: KeyRing::load: loaded key file /etc/ceph/ceph.client.admin.keyring rbd: add failed: (5) Input/output error [root@ceph01 cluster]# cat /var/log/messages | tail Mar 25 12:45:06 ceph01 kernel: libceph: osdc handle_map corrupt msg Mar 25 12:45:06 ceph01 kernel: libceph: mon2 192.168.100.203:6789<http://192.168.100.203:6789> session established Mar 25 12:46:33 ceph01 kernel: libceph: client11240 fsid 10b46114-ac17-404e-99e3-69b34b85c901 Mar 25 12:46:33 ceph01 kernel: libceph: got v 13 cv 11 > 9 of ceph_pg_pool Mar 25 12:46:33 ceph01 kernel: libceph: osdc handle_map corrupt msg Mar 25 12:46:33 ceph01 kernel: libceph: mon2 192.168.100.203:6789<http://192.168.100.203:6789> session established Mar 25 12:48:14 ceph01 kernel: libceph: client11313 fsid 10b46114-ac17-404e-99e3-69b34b85c901 Mar 25 12:48:14 ceph01 kernel: libceph: got v 13 cv 11 > 9 of ceph_pg_pool Mar 25 12:48:14 ceph01 kernel: libceph: osdc handle_map corrupt msg Mar 25 12:48:14 ceph01 kernel: libceph: mon0 192.168.100.201:6789<http://192.168.100.201:6789> session established I do not really understand this error. CRUSH correct. Thanks. 2014-03-25 12:26 GMT+04:00 Ilya Dryomov mailto:ilya.dryo...@inktank.com>>: On Tue, Mar 25, 2014 at 8:38 AM, Ирек Фасихов mailto:malm...@gmail.com>> wrote: > Hi, Ilya. > > I added the files(crushd and osddump) to a folder in GoogleDrive. > > https://drive.google.com/folderview?id=0BxoNLVWxzOJWX0NLV1kzQ1l3Ymc&usp=sharing OK, so this has nothing to do with caching. You have chooseleaf_vary_r set to 1 in your crushmap. This is a new crush tunable, which was introduced long after 3.14 merge window closed. It will be supported starting with 3.15, until then you should be able to do ceph osd getcrushmap -o /tmp/crush crushtool -i /tmp/crush --set-chooseleaf_vary_r 0 -o /tmp/crush.new ceph osd setcrushmap -i /tmp/crush.new to disable it. Thanks, Ilya -- С уважением, Фасихов Ирек Нургаязович Моб.: +79229045757 ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Ceph RBD 0.78 Bug or feature?
Thanks, Ilya. 2014-03-29 22:06 GMT+04:00 Ilya Dryomov : > On Sat, Mar 29, 2014 at 5:15 PM, Ирек Фасихов wrote: > > Ilya, hi. Maybe you have the required patches for the kernel? > > Hi, > > It turned out there was a problem with userspace. If you grab the > latest ceph.git master, you should be able to use cache pools with > 3.14-rc7, which is what I assume you are running. > > If you want the specifics, 7a1990b66ec29f4f3e1659df813ef09831a17cfe is > the commit that merged the fix. > > Thanks, > > Ilya > -- С уважением, Фасихов Ирек Нургаязович Моб.: +79229045757 ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Ceph RBD 0.78 Bug or feature?
On Sat, Mar 29, 2014 at 5:15 PM, Ирек Фасихов wrote: > Ilya, hi. Maybe you have the required patches for the kernel? Hi, It turned out there was a problem with userspace. If you grab the latest ceph.git master, you should be able to use cache pools with 3.14-rc7, which is what I assume you are running. If you want the specifics, 7a1990b66ec29f4f3e1659df813ef09831a17cfe is the commit that merged the fix. Thanks, Ilya ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Ceph RBD 0.78 Bug or feature?
Ilya, hi. Maybe you have the required patches for the kernel? 2014-03-25 14:51 GMT+04:00 Ирек Фасихов : > Yep, so works. > > > 2014-03-25 14:45 GMT+04:00 Ilya Dryomov : > > On Tue, Mar 25, 2014 at 12:00 PM, Ирек Фасихов wrote: >> > Hmmm, create another image in another pool. Pool without cache tier. >> > >> > [root@ceph01 cluster]# rbd create test/image --size 102400 >> > [root@ceph01 cluster]# rbd -p test ls -l >> > NAME SIZE PARENT FMT PROT LOCK >> > image 102400M 1 >> > [root@ceph01 cluster]# ceph osd dump | grep test >> > pool 4 'test' replicated size 3 min_size 2 crush_ruleset 0 object_hash >> > rjenkins pg_num 100 pgp_num 100 last_change 2049 owner 0 flags >> hashpspool >> > stripe_width 0 >> > >> > Get the same error... >> > >> > [root@ceph01 cluster]# rbd map -p test image >> > rbd: add failed: (5) Input/output error >> > >> > Mar 25 13:53:56 ceph01 kernel: libceph: client11343 fsid >> > 10b46114-ac17-404e-99e3-69b34b85c901 >> > Mar 25 13:53:56 ceph01 kernel: libceph: got v 13 cv 11 > 9 of >> ceph_pg_pool >> > Mar 25 13:53:56 ceph01 kernel: libceph: osdc handle_map corrupt msg >> > Mar 25 13:53:56 ceph01 kernel: libceph: mon0 192.168.100.201:6789session >> > established >> > >> > Maybe I'm doing wrong? >> >> No, the problem here is that the pool with hit_set stuff set still >> exists and therefore is present in the osdmap. You'll have to remove >> that pool with something like >> >> # I assume "cache" is the name of the cache pool >> ceph osd tier remove-overlay rbd >> ceph osd tier remove rbd cache >> ceph osd pool delete cache cache --yes-i-really-really-mean-it >> >> in order to be able to map test/image. >> >> Thanks, >> >> Ilya >> > > > > -- > С уважением, Фасихов Ирек Нургаязович > Моб.: +79229045757 > -- С уважением, Фасихов Ирек Нургаязович Моб.: +79229045757 ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Ceph RBD 0.78 Bug or feature?
Yep, so works. 2014-03-25 14:45 GMT+04:00 Ilya Dryomov : > On Tue, Mar 25, 2014 at 12:00 PM, Ирек Фасихов wrote: > > Hmmm, create another image in another pool. Pool without cache tier. > > > > [root@ceph01 cluster]# rbd create test/image --size 102400 > > [root@ceph01 cluster]# rbd -p test ls -l > > NAME SIZE PARENT FMT PROT LOCK > > image 102400M 1 > > [root@ceph01 cluster]# ceph osd dump | grep test > > pool 4 'test' replicated size 3 min_size 2 crush_ruleset 0 object_hash > > rjenkins pg_num 100 pgp_num 100 last_change 2049 owner 0 flags hashpspool > > stripe_width 0 > > > > Get the same error... > > > > [root@ceph01 cluster]# rbd map -p test image > > rbd: add failed: (5) Input/output error > > > > Mar 25 13:53:56 ceph01 kernel: libceph: client11343 fsid > > 10b46114-ac17-404e-99e3-69b34b85c901 > > Mar 25 13:53:56 ceph01 kernel: libceph: got v 13 cv 11 > 9 of > ceph_pg_pool > > Mar 25 13:53:56 ceph01 kernel: libceph: osdc handle_map corrupt msg > > Mar 25 13:53:56 ceph01 kernel: libceph: mon0 192.168.100.201:6789session > > established > > > > Maybe I'm doing wrong? > > No, the problem here is that the pool with hit_set stuff set still > exists and therefore is present in the osdmap. You'll have to remove > that pool with something like > > # I assume "cache" is the name of the cache pool > ceph osd tier remove-overlay rbd > ceph osd tier remove rbd cache > ceph osd pool delete cache cache --yes-i-really-really-mean-it > > in order to be able to map test/image. > > Thanks, > > Ilya > -- С уважением, Фасихов Ирек Нургаязович Моб.: +79229045757 ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Ceph RBD 0.78 Bug or feature?
On Tue, Mar 25, 2014 at 12:00 PM, Ирек Фасихов wrote: > Hmmm, create another image in another pool. Pool without cache tier. > > [root@ceph01 cluster]# rbd create test/image --size 102400 > [root@ceph01 cluster]# rbd -p test ls -l > NAME SIZE PARENT FMT PROT LOCK > image 102400M 1 > [root@ceph01 cluster]# ceph osd dump | grep test > pool 4 'test' replicated size 3 min_size 2 crush_ruleset 0 object_hash > rjenkins pg_num 100 pgp_num 100 last_change 2049 owner 0 flags hashpspool > stripe_width 0 > > Get the same error... > > [root@ceph01 cluster]# rbd map -p test image > rbd: add failed: (5) Input/output error > > Mar 25 13:53:56 ceph01 kernel: libceph: client11343 fsid > 10b46114-ac17-404e-99e3-69b34b85c901 > Mar 25 13:53:56 ceph01 kernel: libceph: got v 13 cv 11 > 9 of ceph_pg_pool > Mar 25 13:53:56 ceph01 kernel: libceph: osdc handle_map corrupt msg > Mar 25 13:53:56 ceph01 kernel: libceph: mon0 192.168.100.201:6789 session > established > > Maybe I'm doing wrong? No, the problem here is that the pool with hit_set stuff set still exists and therefore is present in the osdmap. You'll have to remove that pool with something like # I assume "cache" is the name of the cache pool ceph osd tier remove-overlay rbd ceph osd tier remove rbd cache ceph osd pool delete cache cache --yes-i-really-really-mean-it in order to be able to map test/image. Thanks, Ilya ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Ceph RBD 0.78 Bug or feature?
Hmmm, create another image in another pool. Pool without cache tier. [root@ceph01 cluster]# rbd create test/image --size 102400 [root@ceph01 cluster]# rbd -p test ls -l NAME SIZE PARENT FMT PROT LOCK image 102400M 1 [root@ceph01 cluster]# ceph osd dump | grep test pool 4 'test' replicated size 3 min_size 2 crush_ruleset 0 object_hash rjenkins pg_num 100 pgp_num 100 last_change 2049 owner 0 flags hashpspool stripe_width 0 Get the same error... [root@ceph01 cluster]# rbd map -p test image rbd: add failed: (5) Input/output error Mar 25 13:53:56 ceph01 kernel: libceph: client11343 fsid 10b46114-ac17-404e-99e3-69b34b85c901 Mar 25 13:53:56 ceph01 kernel: libceph: got v 13 cv 11 > 9 of ceph_pg_pool Mar 25 13:53:56 ceph01 kernel: libceph: osdc handle_map corrupt msg Mar 25 13:53:56 ceph01 kernel: libceph: mon0 192.168.100.201:6789 session established Maybe I'm doing wrong? Thanks. 2014-03-25 13:34 GMT+04:00 Ilya Dryomov : > On Tue, Mar 25, 2014 at 10:59 AM, Ирек Фасихов wrote: > > Ilya, set "chooseleaf_vary_r 0", but no map rbd images. > > > > [root@ceph01 cluster]# rbd map rbd/tst > > 2014-03-25 12:48:14.318167 7f44717f7760 2 auth: KeyRing::load: loaded > key > > file /etc/ceph/ceph.client.admin.keyring > > rbd: add failed: (5) Input/output error > > > > [root@ceph01 cluster]# cat /var/log/messages | tail > > Mar 25 12:45:06 ceph01 kernel: libceph: osdc handle_map corrupt msg > > Mar 25 12:45:06 ceph01 kernel: libceph: mon2 192.168.100.203:6789session > > established > > Mar 25 12:46:33 ceph01 kernel: libceph: client11240 fsid > > 10b46114-ac17-404e-99e3-69b34b85c901 > > Mar 25 12:46:33 ceph01 kernel: libceph: got v 13 cv 11 > 9 of > ceph_pg_pool > > Mar 25 12:46:33 ceph01 kernel: libceph: osdc handle_map corrupt msg > > Mar 25 12:46:33 ceph01 kernel: libceph: mon2 192.168.100.203:6789session > > established > > Mar 25 12:48:14 ceph01 kernel: libceph: client11313 fsid > > 10b46114-ac17-404e-99e3-69b34b85c901 > > Mar 25 12:48:14 ceph01 kernel: libceph: got v 13 cv 11 > 9 of > ceph_pg_pool > > Mar 25 12:48:14 ceph01 kernel: libceph: osdc handle_map corrupt msg > > Mar 25 12:48:14 ceph01 kernel: libceph: mon0 192.168.100.201:6789session > > established > > > > I do not really understand this error. CRUSH correct. > > Ah, this error means that the kernel received an unsupported version of > osdmap. Strictly speaking, kernel client doesn't fully support caching > yet. The reason, once again, is that tiering agent and some of the > hit_set stuff are post 3.14. It will probably make 3.15, but I'll have > to get back to you on that. > > Sorry for not letting you know earlier, I got confused. > > Thanks, > > Ilya > -- С уважением, Фасихов Ирек Нургаязович Моб.: +79229045757 ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Ceph RBD 0.78 Bug or feature?
On Tue, Mar 25, 2014 at 10:59 AM, Ирек Фасихов wrote: > Ilya, set "chooseleaf_vary_r 0", but no map rbd images. > > [root@ceph01 cluster]# rbd map rbd/tst > 2014-03-25 12:48:14.318167 7f44717f7760 2 auth: KeyRing::load: loaded key > file /etc/ceph/ceph.client.admin.keyring > rbd: add failed: (5) Input/output error > > [root@ceph01 cluster]# cat /var/log/messages | tail > Mar 25 12:45:06 ceph01 kernel: libceph: osdc handle_map corrupt msg > Mar 25 12:45:06 ceph01 kernel: libceph: mon2 192.168.100.203:6789 session > established > Mar 25 12:46:33 ceph01 kernel: libceph: client11240 fsid > 10b46114-ac17-404e-99e3-69b34b85c901 > Mar 25 12:46:33 ceph01 kernel: libceph: got v 13 cv 11 > 9 of ceph_pg_pool > Mar 25 12:46:33 ceph01 kernel: libceph: osdc handle_map corrupt msg > Mar 25 12:46:33 ceph01 kernel: libceph: mon2 192.168.100.203:6789 session > established > Mar 25 12:48:14 ceph01 kernel: libceph: client11313 fsid > 10b46114-ac17-404e-99e3-69b34b85c901 > Mar 25 12:48:14 ceph01 kernel: libceph: got v 13 cv 11 > 9 of ceph_pg_pool > Mar 25 12:48:14 ceph01 kernel: libceph: osdc handle_map corrupt msg > Mar 25 12:48:14 ceph01 kernel: libceph: mon0 192.168.100.201:6789 session > established > > I do not really understand this error. CRUSH correct. Ah, this error means that the kernel received an unsupported version of osdmap. Strictly speaking, kernel client doesn't fully support caching yet. The reason, once again, is that tiering agent and some of the hit_set stuff are post 3.14. It will probably make 3.15, but I'll have to get back to you on that. Sorry for not letting you know earlier, I got confused. Thanks, Ilya ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Ceph RBD 0.78 Bug or feature?
I also added a new log in Google Drive. https://drive.google.com/folderview?id=0BxoNLVWxzOJWX0NLV1kzQ1l3Ymc&usp=sharing 2014-03-25 12:59 GMT+04:00 Ирек Фасихов : > Ilya, set "chooseleaf_vary_r 0", but no map rbd images. > > [root@ceph01 cluster]# *rbd map rbd/tst* > 2014-03-25 12:48:14.318167 7f44717f7760 2 auth: KeyRing::load: loaded key > file /etc/ceph/ceph.client.admin.keyring > rbd: add failed: (5) Input/output error > > [root@ceph01 cluster]# *cat /var/log/messages | tail* > Mar 25 12:45:06 ceph01 kernel: libceph: *osdc handle_map corrupt msg* > Mar 25 12:45:06 ceph01 kernel: libceph: mon2 192.168.100.203:6789 session > established > Mar 25 12:46:33 ceph01 kernel: libceph: client11240 fsid > 10b46114-ac17-404e-99e3-69b34b85c901 > Mar 25 12:46:33 ceph01 kernel: libceph: got v 13 cv 11 > 9 of ceph_pg_pool > Mar 25 12:46:33 ceph01 kernel: libceph: osdc handle_map corrupt msg > Mar 25 12:46:33 ceph01 kernel: libceph: mon2 192.168.100.203:6789 session > established > Mar 25 12:48:14 ceph01 kernel: libceph: client11313 fsid > 10b46114-ac17-404e-99e3-69b34b85c901 > Mar 25 12:48:14 ceph01 kernel: libceph: got v 13 cv 11 > 9 of ceph_pg_pool > Mar 25 12:48:14 ceph01 kernel: libceph: osdc handle_map corrupt msg > Mar 25 12:48:14 ceph01 kernel: libceph: mon0 192.168.100.201:6789 session > established > > I do not really understand this error. CRUSH correct. > > Thanks. > > > > 2014-03-25 12:26 GMT+04:00 Ilya Dryomov : > > On Tue, Mar 25, 2014 at 8:38 AM, Ирек Фасихов wrote: >> > Hi, Ilya. >> > >> > I added the files(crushd and osddump) to a folder in GoogleDrive. >> > >> > >> https://drive.google.com/folderview?id=0BxoNLVWxzOJWX0NLV1kzQ1l3Ymc&usp=sharing >> >> OK, so this has nothing to do with caching. You have chooseleaf_vary_r >> set to 1 in your crushmap. This is a new crush tunable, which was >> introduced long after 3.14 merge window closed. It will be supported >> starting with 3.15, until then you should be able to do >> >> ceph osd getcrushmap -o /tmp/crush >> crushtool -i /tmp/crush --set-chooseleaf_vary_r 0 -o /tmp/crush.new >> ceph osd setcrushmap -i /tmp/crush.new >> >> to disable it. >> >> Thanks, >> >> Ilya >> > > > > -- > С уважением, Фасихов Ирек Нургаязович > Моб.: +79229045757 > -- С уважением, Фасихов Ирек Нургаязович Моб.: +79229045757 ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Ceph RBD 0.78 Bug or feature?
Ilya, set "chooseleaf_vary_r 0", but no map rbd images. [root@ceph01 cluster]# *rbd map rbd/tst* 2014-03-25 12:48:14.318167 7f44717f7760 2 auth: KeyRing::load: loaded key file /etc/ceph/ceph.client.admin.keyring rbd: add failed: (5) Input/output error [root@ceph01 cluster]# *cat /var/log/messages | tail* Mar 25 12:45:06 ceph01 kernel: libceph: *osdc handle_map corrupt msg* Mar 25 12:45:06 ceph01 kernel: libceph: mon2 192.168.100.203:6789 session established Mar 25 12:46:33 ceph01 kernel: libceph: client11240 fsid 10b46114-ac17-404e-99e3-69b34b85c901 Mar 25 12:46:33 ceph01 kernel: libceph: got v 13 cv 11 > 9 of ceph_pg_pool Mar 25 12:46:33 ceph01 kernel: libceph: osdc handle_map corrupt msg Mar 25 12:46:33 ceph01 kernel: libceph: mon2 192.168.100.203:6789 session established Mar 25 12:48:14 ceph01 kernel: libceph: client11313 fsid 10b46114-ac17-404e-99e3-69b34b85c901 Mar 25 12:48:14 ceph01 kernel: libceph: got v 13 cv 11 > 9 of ceph_pg_pool Mar 25 12:48:14 ceph01 kernel: libceph: osdc handle_map corrupt msg Mar 25 12:48:14 ceph01 kernel: libceph: mon0 192.168.100.201:6789 session established I do not really understand this error. CRUSH correct. Thanks. 2014-03-25 12:26 GMT+04:00 Ilya Dryomov : > On Tue, Mar 25, 2014 at 8:38 AM, Ирек Фасихов wrote: > > Hi, Ilya. > > > > I added the files(crushd and osddump) to a folder in GoogleDrive. > > > > > https://drive.google.com/folderview?id=0BxoNLVWxzOJWX0NLV1kzQ1l3Ymc&usp=sharing > > OK, so this has nothing to do with caching. You have chooseleaf_vary_r > set to 1 in your crushmap. This is a new crush tunable, which was > introduced long after 3.14 merge window closed. It will be supported > starting with 3.15, until then you should be able to do > > ceph osd getcrushmap -o /tmp/crush > crushtool -i /tmp/crush --set-chooseleaf_vary_r 0 -o /tmp/crush.new > ceph osd setcrushmap -i /tmp/crush.new > > to disable it. > > Thanks, > > Ilya > -- С уважением, Фасихов Ирек Нургаязович Моб.: +79229045757 ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Ceph RBD 0.78 Bug or feature?
On Tue, Mar 25, 2014 at 8:38 AM, Ирек Фасихов wrote: > Hi, Ilya. > > I added the files(crushd and osddump) to a folder in GoogleDrive. > > https://drive.google.com/folderview?id=0BxoNLVWxzOJWX0NLV1kzQ1l3Ymc&usp=sharing OK, so this has nothing to do with caching. You have chooseleaf_vary_r set to 1 in your crushmap. This is a new crush tunable, which was introduced long after 3.14 merge window closed. It will be supported starting with 3.15, until then you should be able to do ceph osd getcrushmap -o /tmp/crush crushtool -i /tmp/crush --set-chooseleaf_vary_r 0 -o /tmp/crush.new ceph osd setcrushmap -i /tmp/crush.new to disable it. Thanks, Ilya ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Ceph RBD 0.78 Bug or feature?
Hi, Ilya. I added the files(crushd and osddump) to a folder in GoogleDrive. https://drive.google.com/folderview?id=0BxoNLVWxzOJWX0NLV1kzQ1l3Ymc&usp=sharing 2014-03-25 0:19 GMT+04:00 Ilya Dryomov : > On Mon, Mar 24, 2014 at 9:46 PM, Ирек Фасихов wrote: > > Kernel module support RBD, Ilya Dryomov? > > > > Thanks? > > > > > > 2014-03-24 23:32 GMT+04:00 Michael J. Kidd : > > > >> The message 'feature set mismatch' is exactly what Gregory was talking > >> about. > >> > >> This indicates that you are using a CRUSH feature in your Ceph > environment > >> that the Kernel RBD client doesn't understand. So, it's unable to > >> communicate. In this instance, it's likely to do with the cache pool > >> itself, but I'm not terribly familiar with the features support in 3.14 > >> kernel... > >> > >> Thanks, > >> > >> Michael J. Kidd > >> Sr. Storage Consultant > >> Inktank Professional Services > >> > >> > >> On Mon, Mar 24, 2014 at 2:58 PM, Ирек Фасихов > wrote: > >>> > >>> Hi, Gregory! > >>> I think that there is no interesting :). > >>> > >>> dmesg: > >>> Key type dns_resolver registered > >>> Key type ceph registered > >>> libceph: loaded (mon/osd proto 15/24) > >>> rbd: loaded (major 252) > >>> libceph: mon1 192.168.100.202:6789 feature set mismatch, my > 384a042a42 < > >>> server's 2384a042a42, missing 200 > >>> libceph: mon1 192.168.100.202:6789 socket error on read > >>> libceph: mon1 192.168.100.202:6789 feature set mismatch, my > 384a042a42 < > >>> server's 2384a042a42, missing 200 > >>> libceph: mon1 192.168.100.202:6789 socket error on read > >>> libceph: mon2 192.168.100.203:6789 feature set mismatch, my > 384a042a42 < > >>> server's 2384a042a42, missing 200 > >>> libceph: mon2 192.168.100.203:6789 socket error on read > >>> libceph: mon2 192.168.100.203:6789 feature set mismatch, my > 384a042a42 < > >>> server's 2384a042a42, missing 200 > >>> libceph: mon2 192.168.100.203:6789 socket error on read > >>> libceph: mon1 192.168.100.202:6789 feature set mismatch, my > 384a042a42 < > >>> server's 2384a042a42, missing 200 > >>> libceph: mon1 192.168.100.202:6789 socket error on read > >>> libceph: mon1 192.168.100.202:6789 feature set mismatch, my > 384a042a42 < > >>> server's 2384a042a42, missing 200 > >>> libceph: mon1 192.168.100.202:6789 socket error on read > >>> libceph: mon1 192.168.100.202:6789 feature set mismatch, my > 384a042a42 < > >>> server's 2384a042a42, missing 200 > >>> libceph: mon1 192.168.100.202:6789 socket error on read > >>> libceph: mon2 192.168.100.203:6789 feature set mismatch, my > 384a042a42 < > >>> server's 2384a042a42, missing 200 > >>> libceph: mon2 192.168.100.203:6789 socket error on read > >>> libceph: mon2 192.168.100.203:6789 feature set mismatch, my > 384a042a42 < > >>> server's 2384a042a42, missing 200 > >>> libceph: mon2 192.168.100.203:6789 socket error on read > >>> libceph: mon0 192.168.100.201:6789 feature set mismatch, my > 384a042a42 < > >>> server's 2384a042a42, missing 200 > >>> libceph: mon0 192.168.100.201:6789 socket error on read > >>> libceph: mon1 192.168.100.202:6789 feature set mismatch, my > 384a042a42 < > >>> server's 2384a042a42, missing 200 > >>> libceph: mon1 192.168.100.202:6789 socket error on read > >>> libceph: mon2 192.168.100.203:6789 feature set mismatch, my > 384a042a42 < > >>> server's 2384a042a42, missing 200 > >>> libceph: mon2 192.168.100.203:6789 socket error on read > > Hi, > > 200 is bit 41, so it's either crush tunables v3 or primary > affinity. Do you have crush chooseleaf_vary_r set to a non-default > value or primary-affinity adjusted with the 'ceph osd primary-affinity' > command? > > Can you add the output of 'ceph osd dump --format=json-pretty' and the > output of 'ceph osd getcrushmap -o /tmp/crush; crushtool -d /tmp/crush' > to your Drive folder so I can check myself? > > Thanks, > > Ilya > -- С уважением, Фасихов Ирек Нургаязович Моб.: +79229045757 ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Ceph RBD 0.78 Bug or feature?
On Mon, Mar 24, 2014 at 9:46 PM, Ирек Фасихов wrote: > Kernel module support RBD, Ilya Dryomov? > > Thanks? > > > 2014-03-24 23:32 GMT+04:00 Michael J. Kidd : > >> The message 'feature set mismatch' is exactly what Gregory was talking >> about. >> >> This indicates that you are using a CRUSH feature in your Ceph environment >> that the Kernel RBD client doesn't understand. So, it's unable to >> communicate. In this instance, it's likely to do with the cache pool >> itself, but I'm not terribly familiar with the features support in 3.14 >> kernel... >> >> Thanks, >> >> Michael J. Kidd >> Sr. Storage Consultant >> Inktank Professional Services >> >> >> On Mon, Mar 24, 2014 at 2:58 PM, Ирек Фасихов wrote: >>> >>> Hi, Gregory! >>> I think that there is no interesting :). >>> >>> dmesg: >>> Key type dns_resolver registered >>> Key type ceph registered >>> libceph: loaded (mon/osd proto 15/24) >>> rbd: loaded (major 252) >>> libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < >>> server's 2384a042a42, missing 200 >>> libceph: mon1 192.168.100.202:6789 socket error on read >>> libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < >>> server's 2384a042a42, missing 200 >>> libceph: mon1 192.168.100.202:6789 socket error on read >>> libceph: mon2 192.168.100.203:6789 feature set mismatch, my 384a042a42 < >>> server's 2384a042a42, missing 200 >>> libceph: mon2 192.168.100.203:6789 socket error on read >>> libceph: mon2 192.168.100.203:6789 feature set mismatch, my 384a042a42 < >>> server's 2384a042a42, missing 200 >>> libceph: mon2 192.168.100.203:6789 socket error on read >>> libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < >>> server's 2384a042a42, missing 200 >>> libceph: mon1 192.168.100.202:6789 socket error on read >>> libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < >>> server's 2384a042a42, missing 200 >>> libceph: mon1 192.168.100.202:6789 socket error on read >>> libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < >>> server's 2384a042a42, missing 200 >>> libceph: mon1 192.168.100.202:6789 socket error on read >>> libceph: mon2 192.168.100.203:6789 feature set mismatch, my 384a042a42 < >>> server's 2384a042a42, missing 200 >>> libceph: mon2 192.168.100.203:6789 socket error on read >>> libceph: mon2 192.168.100.203:6789 feature set mismatch, my 384a042a42 < >>> server's 2384a042a42, missing 200 >>> libceph: mon2 192.168.100.203:6789 socket error on read >>> libceph: mon0 192.168.100.201:6789 feature set mismatch, my 384a042a42 < >>> server's 2384a042a42, missing 200 >>> libceph: mon0 192.168.100.201:6789 socket error on read >>> libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < >>> server's 2384a042a42, missing 200 >>> libceph: mon1 192.168.100.202:6789 socket error on read >>> libceph: mon2 192.168.100.203:6789 feature set mismatch, my 384a042a42 < >>> server's 2384a042a42, missing 200 >>> libceph: mon2 192.168.100.203:6789 socket error on read Hi, 200 is bit 41, so it's either crush tunables v3 or primary affinity. Do you have crush chooseleaf_vary_r set to a non-default value or primary-affinity adjusted with the 'ceph osd primary-affinity' command? Can you add the output of 'ceph osd dump --format=json-pretty' and the output of 'ceph osd getcrushmap -o /tmp/crush; crushtool -d /tmp/crush' to your Drive folder so I can check myself? Thanks, Ilya ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Ceph RBD 0.78 Bug or feature?
Kernel module support RBD, Ilya Dryomov? Thanks? 2014-03-24 23:32 GMT+04:00 Michael J. Kidd : > The message 'feature set mismatch' is exactly what Gregory was talking > about. > > This indicates that you are using a CRUSH feature in your Ceph environment > that the Kernel RBD client doesn't understand. So, it's unable to > communicate. In this instance, it's likely to do with the cache pool > itself, but I'm not terribly familiar with the features support in 3.14 > kernel... > > Thanks, > > Michael J. Kidd > Sr. Storage Consultant > Inktank Professional Services > > > On Mon, Mar 24, 2014 at 2:58 PM, Ирек Фасихов wrote: > >> Hi, Gregory! >> I think that there is no interesting :). >> >> *dmesg:* >> Key type dns_resolver registered >> Key type ceph registered >> libceph: loaded (mon/osd proto 15/24) >> rbd: loaded (major 252) >> libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < >> server's 2384a042a42, missing 200 >> libceph: mon1 192.168.100.202:6789 socket error on read >> libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < >> server's 2384a042a42, missing 200 >> libceph: mon1 192.168.100.202:6789 socket error on read >> libceph: mon2 192.168.100.203:6789 feature set mismatch, my 384a042a42 < >> server's 2384a042a42, missing 200 >> libceph: mon2 192.168.100.203:6789 socket error on read >> libceph: mon2 192.168.100.203:6789 feature set mismatch, my 384a042a42 < >> server's 2384a042a42, missing 200 >> libceph: mon2 192.168.100.203:6789 socket error on read >> libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < >> server's 2384a042a42, missing 200 >> libceph: mon1 192.168.100.202:6789 socket error on read >> libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < >> server's 2384a042a42, missing 200 >> libceph: mon1 192.168.100.202:6789 socket error on read >> libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < >> server's 2384a042a42, missing 200 >> libceph: mon1 192.168.100.202:6789 socket error on read >> libceph: mon2 192.168.100.203:6789 feature set mismatch, my 384a042a42 < >> server's 2384a042a42, missing 200 >> libceph: mon2 192.168.100.203:6789 socket error on read >> libceph: mon2 192.168.100.203:6789 feature set mismatch, my 384a042a42 < >> server's 2384a042a42, missing 200 >> libceph: mon2 192.168.100.203:6789 socket error on read >> libceph: mon0 192.168.100.201:6789 feature set mismatch, my 384a042a42 < >> server's 2384a042a42, missing 200 >> libceph: mon0 192.168.100.201:6789 socket error on read >> libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < >> server's 2384a042a42, missing 200 >> libceph: mon1 192.168.100.202:6789 socket error on read >> libceph: mon2 192.168.100.203:6789 feature set mismatch, my 384a042a42 < >> server's 2384a042a42, missing 200 >> libceph: mon2 192.168.100.203:6789 socket error on read >> >> >> >> 2014-03-24 21:16 GMT+04:00 Gregory Farnum : >> >> I don't remember what features should exist where, but I expect that >>> the cluster is making use of features that the kernel client doesn't >>> support yet (despite the very new kernel). Have you checked to see if >>> there's anything interesting in dmesg? >>> -Greg >>> Software Engineer #42 @ http://inktank.com | http://ceph.com >>> >>> >>> On Mon, Mar 24, 2014 at 1:30 AM, Ирек Фасихов wrote: >>> > Created cache pool for documentation: >>> > http://ceph.com/docs/master/dev/cache-pool/ >>> > >>> > ceph osd pool create cache 100 >>> > ceph osd tier add rbd cache >>> > ceph osd tier cache-mode cache writeback >>> > ceph osd tier set-overlay rbd cache >>> > ceph osd pool set cache hit_set_type bloom >>> > ceph osd pool set cache hit_set_count 1 >>> > ceph osd pool set cache hit_set_period 600 >>> > ceph osd pool set cache target_max_bytes 100 #10Gb >>> > >>> > ceph osd tree: >>> > # idweight type name up/down reweight >>> > -1 6 root default >>> > -2 2 host ceph01 >>> > 0 1 osd.0 up 1 >>> > 3 1 osd.3 up 1 >>> > -3 2 host ceph02 >>> > 1 1 osd.1 up 1 >>> > 4 1 osd.4 up 1 >>> > -4 2 host ceph03 >>> > 2 1 osd.2 up 1 >>> > 5 1 osd.5 up 1 >>> > >>> > rbd -p rbd ls -l >>> > NAME SIZE PARENT FMT PROT LOCK >>> > test 10240M 1 >>> > >>> > rbd map rbd/test >>> > rbd: add failed: (5) Input/output error >>> > >>> > uname -a >>> > Linux ceph01.bank-hlynov.ru 3.14.0-rc7-bank-hlynov.ru #1 SMP Mon Mar >>> 17 >>> > 11:49:22 MSK 2014 x86_64 x86_64 x86_64 GNU/Linux. Linux CentOS 6.5 >>> > >>> > Debug mode and strace in logs. >>> > >>> > >>> https://drive.google.com/folderview?id=0BxoNLVWxzOJWX0NLV1kzQ1l3Ymc&usp=sharin
Re: [ceph-users] Ceph RBD 0.78 Bug or feature?
The message 'feature set mismatch' is exactly what Gregory was talking about. This indicates that you are using a CRUSH feature in your Ceph environment that the Kernel RBD client doesn't understand. So, it's unable to communicate. In this instance, it's likely to do with the cache pool itself, but I'm not terribly familiar with the features support in 3.14 kernel... Thanks, Michael J. Kidd Sr. Storage Consultant Inktank Professional Services On Mon, Mar 24, 2014 at 2:58 PM, Ирек Фасихов wrote: > Hi, Gregory! > I think that there is no interesting :). > > *dmesg:* > Key type dns_resolver registered > Key type ceph registered > libceph: loaded (mon/osd proto 15/24) > rbd: loaded (major 252) > libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < > server's 2384a042a42, missing 200 > libceph: mon1 192.168.100.202:6789 socket error on read > libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < > server's 2384a042a42, missing 200 > libceph: mon1 192.168.100.202:6789 socket error on read > libceph: mon2 192.168.100.203:6789 feature set mismatch, my 384a042a42 < > server's 2384a042a42, missing 200 > libceph: mon2 192.168.100.203:6789 socket error on read > libceph: mon2 192.168.100.203:6789 feature set mismatch, my 384a042a42 < > server's 2384a042a42, missing 200 > libceph: mon2 192.168.100.203:6789 socket error on read > libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < > server's 2384a042a42, missing 200 > libceph: mon1 192.168.100.202:6789 socket error on read > libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < > server's 2384a042a42, missing 200 > libceph: mon1 192.168.100.202:6789 socket error on read > libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < > server's 2384a042a42, missing 200 > libceph: mon1 192.168.100.202:6789 socket error on read > libceph: mon2 192.168.100.203:6789 feature set mismatch, my 384a042a42 < > server's 2384a042a42, missing 200 > libceph: mon2 192.168.100.203:6789 socket error on read > libceph: mon2 192.168.100.203:6789 feature set mismatch, my 384a042a42 < > server's 2384a042a42, missing 200 > libceph: mon2 192.168.100.203:6789 socket error on read > libceph: mon0 192.168.100.201:6789 feature set mismatch, my 384a042a42 < > server's 2384a042a42, missing 200 > libceph: mon0 192.168.100.201:6789 socket error on read > libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < > server's 2384a042a42, missing 200 > libceph: mon1 192.168.100.202:6789 socket error on read > libceph: mon2 192.168.100.203:6789 feature set mismatch, my 384a042a42 < > server's 2384a042a42, missing 200 > libceph: mon2 192.168.100.203:6789 socket error on read > > > > 2014-03-24 21:16 GMT+04:00 Gregory Farnum : > > I don't remember what features should exist where, but I expect that >> the cluster is making use of features that the kernel client doesn't >> support yet (despite the very new kernel). Have you checked to see if >> there's anything interesting in dmesg? >> -Greg >> Software Engineer #42 @ http://inktank.com | http://ceph.com >> >> >> On Mon, Mar 24, 2014 at 1:30 AM, Ирек Фасихов wrote: >> > Created cache pool for documentation: >> > http://ceph.com/docs/master/dev/cache-pool/ >> > >> > ceph osd pool create cache 100 >> > ceph osd tier add rbd cache >> > ceph osd tier cache-mode cache writeback >> > ceph osd tier set-overlay rbd cache >> > ceph osd pool set cache hit_set_type bloom >> > ceph osd pool set cache hit_set_count 1 >> > ceph osd pool set cache hit_set_period 600 >> > ceph osd pool set cache target_max_bytes 100 #10Gb >> > >> > ceph osd tree: >> > # idweight type name up/down reweight >> > -1 6 root default >> > -2 2 host ceph01 >> > 0 1 osd.0 up 1 >> > 3 1 osd.3 up 1 >> > -3 2 host ceph02 >> > 1 1 osd.1 up 1 >> > 4 1 osd.4 up 1 >> > -4 2 host ceph03 >> > 2 1 osd.2 up 1 >> > 5 1 osd.5 up 1 >> > >> > rbd -p rbd ls -l >> > NAME SIZE PARENT FMT PROT LOCK >> > test 10240M 1 >> > >> > rbd map rbd/test >> > rbd: add failed: (5) Input/output error >> > >> > uname -a >> > Linux ceph01.bank-hlynov.ru 3.14.0-rc7-bank-hlynov.ru #1 SMP Mon Mar 17 >> > 11:49:22 MSK 2014 x86_64 x86_64 x86_64 GNU/Linux. Linux CentOS 6.5 >> > >> > Debug mode and strace in logs. >> > >> > >> https://drive.google.com/folderview?id=0BxoNLVWxzOJWX0NLV1kzQ1l3Ymc&usp=sharing >> > >> > >> > -- >> > С уважением, Фасихов Ирек Нургаязович >> > Моб.: +79229045757 >> > >> > ___ >> > ceph-users mailing list >> > ceph-users@lists.ceph.com >> > http://lists.ceph.co
Re: [ceph-users] Ceph RBD 0.78 Bug or feature?
Hi, Gregory! I think that there is no interesting :). *dmesg:* Key type dns_resolver registered Key type ceph registered libceph: loaded (mon/osd proto 15/24) rbd: loaded (major 252) libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < server's 2384a042a42, missing 200 libceph: mon1 192.168.100.202:6789 socket error on read libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < server's 2384a042a42, missing 200 libceph: mon1 192.168.100.202:6789 socket error on read libceph: mon2 192.168.100.203:6789 feature set mismatch, my 384a042a42 < server's 2384a042a42, missing 200 libceph: mon2 192.168.100.203:6789 socket error on read libceph: mon2 192.168.100.203:6789 feature set mismatch, my 384a042a42 < server's 2384a042a42, missing 200 libceph: mon2 192.168.100.203:6789 socket error on read libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < server's 2384a042a42, missing 200 libceph: mon1 192.168.100.202:6789 socket error on read libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < server's 2384a042a42, missing 200 libceph: mon1 192.168.100.202:6789 socket error on read libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < server's 2384a042a42, missing 200 libceph: mon1 192.168.100.202:6789 socket error on read libceph: mon2 192.168.100.203:6789 feature set mismatch, my 384a042a42 < server's 2384a042a42, missing 200 libceph: mon2 192.168.100.203:6789 socket error on read libceph: mon2 192.168.100.203:6789 feature set mismatch, my 384a042a42 < server's 2384a042a42, missing 200 libceph: mon2 192.168.100.203:6789 socket error on read libceph: mon0 192.168.100.201:6789 feature set mismatch, my 384a042a42 < server's 2384a042a42, missing 200 libceph: mon0 192.168.100.201:6789 socket error on read libceph: mon1 192.168.100.202:6789 feature set mismatch, my 384a042a42 < server's 2384a042a42, missing 200 libceph: mon1 192.168.100.202:6789 socket error on read libceph: mon2 192.168.100.203:6789 feature set mismatch, my 384a042a42 < server's 2384a042a42, missing 200 libceph: mon2 192.168.100.203:6789 socket error on read 2014-03-24 21:16 GMT+04:00 Gregory Farnum : > I don't remember what features should exist where, but I expect that > the cluster is making use of features that the kernel client doesn't > support yet (despite the very new kernel). Have you checked to see if > there's anything interesting in dmesg? > -Greg > Software Engineer #42 @ http://inktank.com | http://ceph.com > > > On Mon, Mar 24, 2014 at 1:30 AM, Ирек Фасихов wrote: > > Created cache pool for documentation: > > http://ceph.com/docs/master/dev/cache-pool/ > > > > ceph osd pool create cache 100 > > ceph osd tier add rbd cache > > ceph osd tier cache-mode cache writeback > > ceph osd tier set-overlay rbd cache > > ceph osd pool set cache hit_set_type bloom > > ceph osd pool set cache hit_set_count 1 > > ceph osd pool set cache hit_set_period 600 > > ceph osd pool set cache target_max_bytes 100 #10Gb > > > > ceph osd tree: > > # idweight type name up/down reweight > > -1 6 root default > > -2 2 host ceph01 > > 0 1 osd.0 up 1 > > 3 1 osd.3 up 1 > > -3 2 host ceph02 > > 1 1 osd.1 up 1 > > 4 1 osd.4 up 1 > > -4 2 host ceph03 > > 2 1 osd.2 up 1 > > 5 1 osd.5 up 1 > > > > rbd -p rbd ls -l > > NAME SIZE PARENT FMT PROT LOCK > > test 10240M 1 > > > > rbd map rbd/test > > rbd: add failed: (5) Input/output error > > > > uname -a > > Linux ceph01.bank-hlynov.ru 3.14.0-rc7-bank-hlynov.ru #1 SMP Mon Mar 17 > > 11:49:22 MSK 2014 x86_64 x86_64 x86_64 GNU/Linux. Linux CentOS 6.5 > > > > Debug mode and strace in logs. > > > > > https://drive.google.com/folderview?id=0BxoNLVWxzOJWX0NLV1kzQ1l3Ymc&usp=sharing > > > > > > -- > > С уважением, Фасихов Ирек Нургаязович > > Моб.: +79229045757 > > > > ___ > > ceph-users mailing list > > ceph-users@lists.ceph.com > > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > > > -- С уважением, Фасихов Ирек Нургаязович Моб.: +79229045757 ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Ceph RBD 0.78 Bug or feature?
I don't remember what features should exist where, but I expect that the cluster is making use of features that the kernel client doesn't support yet (despite the very new kernel). Have you checked to see if there's anything interesting in dmesg? -Greg Software Engineer #42 @ http://inktank.com | http://ceph.com On Mon, Mar 24, 2014 at 1:30 AM, Ирек Фасихов wrote: > Created cache pool for documentation: > http://ceph.com/docs/master/dev/cache-pool/ > > ceph osd pool create cache 100 > ceph osd tier add rbd cache > ceph osd tier cache-mode cache writeback > ceph osd tier set-overlay rbd cache > ceph osd pool set cache hit_set_type bloom > ceph osd pool set cache hit_set_count 1 > ceph osd pool set cache hit_set_period 600 > ceph osd pool set cache target_max_bytes 100 #10Gb > > ceph osd tree: > # idweight type name up/down reweight > -1 6 root default > -2 2 host ceph01 > 0 1 osd.0 up 1 > 3 1 osd.3 up 1 > -3 2 host ceph02 > 1 1 osd.1 up 1 > 4 1 osd.4 up 1 > -4 2 host ceph03 > 2 1 osd.2 up 1 > 5 1 osd.5 up 1 > > rbd -p rbd ls -l > NAME SIZE PARENT FMT PROT LOCK > test 10240M 1 > > rbd map rbd/test > rbd: add failed: (5) Input/output error > > uname -a > Linux ceph01.bank-hlynov.ru 3.14.0-rc7-bank-hlynov.ru #1 SMP Mon Mar 17 > 11:49:22 MSK 2014 x86_64 x86_64 x86_64 GNU/Linux. Linux CentOS 6.5 > > Debug mode and strace in logs. > > https://drive.google.com/folderview?id=0BxoNLVWxzOJWX0NLV1kzQ1l3Ymc&usp=sharing > > > -- > С уважением, Фасихов Ирек Нургаязович > Моб.: +79229045757 > > ___ > ceph-users mailing list > 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] Ceph RBD 0.78 Bug or feature?
Created cache pool for documentation: http://ceph.com/docs/master/dev/cache-pool/ *ceph osd pool create cache 100* *ceph osd tier add rbd cache* *ceph osd tier cache-mode cache writeback* *ceph osd tier set-overlay rbd cache* *ceph osd pool set cache hit_set_type bloom* *ceph osd pool set cache hit_set_count 1* *ceph osd pool set cache hit_set_period 600* *ceph osd pool set cache target_max_bytes 100 #10Gb* *ceph osd tree:* # idweight type name up/down reweight -1 6 root default -2 2 host ceph01 0 1 osd.0 up 1 3 1 osd.3 up 1 -3 2 host ceph02 1 1 osd.1 up 1 4 1 osd.4 up 1 -4 2 host ceph03 2 1 osd.2 up 1 5 1 osd.5 up 1 *rbd -p rbd ls -l * NAME SIZE PARENT FMT PROT LOCK test 10240M 1 *rbd map rbd/test* rbd: add failed: (5) Input/output error *uname -a* Linux ceph01.bank-hlynov.ru 3.14.0-rc7-bank-hlynov.ru #1 SMP Mon Mar 17 11:49:22 MSK 2014 x86_64 x86_64 x86_64 GNU/Linux. Linux CentOS 6.5 Debug mode and strace in logs. https://drive.google.com/folderview?id=0BxoNLVWxzOJWX0NLV1kzQ1l3Ymc&usp=sharing -- С уважением, Фасихов Ирек Нургаязович Моб.: +79229045757 ceph.conf Description: Binary data ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com