Hi everyone, we've encountered an unusual thing in our setup(4 nodes, 48
OSDs, 3 monitors - ceph Jewel, Ubuntu 16.04 with kernel 4.4.0).
Yesterday, we were doing a HW upgrade of the nodes, so they went down
one by one - the cluster was in good shape during the upgrade, as we've
done this numerous times and we're quite sure that the redundancy wasn't
screwed up while doing this. However, during this upgrade one of the
clients that does backups to cephfs(mounted via the kernel driver)
failed to write the backup file correctly to the cluster with the
following trace after we turned off one of the nodes:
[2585732.529412] ffff8800baa279a8 ffffffff813fb2df ffff880236230e00
ffff8802339c0000
[2585732.529414] ffff8800baa28000 ffff88023fc96e00 7fffffffffffffff
ffff8800baa27b20
[2585732.529415] ffffffff81840ed0 ffff8800baa279c0 ffffffff818406d5
0000000000000000
[2585732.529417] Call Trace:
[2585732.529505] [<ffffffff813fb2df>] ? cpumask_next_and+0x2f/0x40
[2585732.529558] [<ffffffff81840ed0>] ? bit_wait+0x60/0x60
[2585732.529560] [<ffffffff818406d5>] schedule+0x35/0x80
[2585732.529562] [<ffffffff81843825>] schedule_timeout+0x1b5/0x270
[2585732.529607] [<ffffffff810642be>] ? kvm_clock_get_cycles+0x1e/0x20
[2585732.529609] [<ffffffff81840ed0>] ? bit_wait+0x60/0x60
[2585732.529611] [<ffffffff8183fc04>] io_schedule_timeout+0xa4/0x110
[2585732.529613] [<ffffffff81840eeb>] bit_wait_io+0x1b/0x70
[2585732.529614] [<ffffffff81840c6e>] __wait_on_bit_lock+0x4e/0xb0
[2585732.529652] [<ffffffff8118f3cb>] __lock_page+0xbb/0xe0
[2585732.529674] [<ffffffff810c4460>] ? autoremove_wake_function+0x40/0x40
[2585732.529676] [<ffffffff8119078d>] pagecache_get_page+0x17d/0x1c0
[2585732.529730] [<ffffffffc056b3a8>] ? ceph_pool_perm_check+0x48/0x700 [ceph]
[2585732.529732] [<ffffffff811907f6>] grab_cache_page_write_begin+0x26/0x40
[2585732.529738] [<ffffffffc056a6a8>] ceph_write_begin+0x48/0xe0 [ceph]
[2585732.529739] [<ffffffff8118fd6e>] generic_perform_write+0xce/0x1c0
[2585732.529763] [<ffffffff8122bdb9>] ? file_update_time+0xc9/0x110
[2585732.529769] [<ffffffffc05651c9>] ceph_write_iter+0xf89/0x1040 [ceph]
[2585732.529792] [<ffffffff81199c19>] ? __alloc_pages_nodemask+0x159/0x2a0
[2585732.529808] [<ffffffff8120fedb>] new_sync_write+0x9b/0xe0
[2585732.529811] [<ffffffff8120ff46>] __vfs_write+0x26/0x40
[2585732.529812] [<ffffffff812108c9>] vfs_write+0xa9/0x1a0
[2585732.529814] [<ffffffff81211585>] SyS_write+0x55/0xc0
[2585732.529817] [<ffffffff818447f2>] entry_SYSCALL_64_fastpath+0x16/0x71
I have encountered this behavior on Luminous, but not on Jewel. Anyone who has
a clue why the write fails? As far as i'm concerned, it should always work if
all the PGs are available. Thanks
Josef
_______________________________________________
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com