https://jira.whamcloud.com This file is an XML representation of an issue en-us 9.4.14 940014 05-12-2023 https://jira.whamcloud.com/browse/LU-2117 Lustre <p>Running sanityn I hit this assertion today.<br/> I have a kernel crashdump if more info is needed file</p> <div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"> <pre class="code-java">[ 4848.384105] Lustre: DEBUG MARKER: == sanityn test 18: mmap sanity check =================================== 00:56:56 (1349758616) [ 4848.455414] LustreError: 380:0:(lov_object.c:392:lov_attr_get_raid0()) ASSERTION( atomic_read(&amp;lsm-&gt;lsm_refc) &gt; 1 ) failed: [ 4848.456184] LustreError: 380:0:(lov_object.c:392:lov_attr_get_raid0()) LBUG [ 4848.456559] Pid: 380, comm: mmap_sanity [ 4848.456863] [ 4848.456864] Call Trace: [ 4848.457463] [&lt;ffffffffa0b34915&gt;] libcfs_debug_dumpstack+0x55/0x80 [libcfs] [ 4848.457854] [&lt;ffffffffa0b34f27&gt;] lbug_with_loc+0x47/0xb0 [libcfs] [ 4848.458219] [&lt;ffffffffa059008d&gt;] lov_attr_get_raid0+0x2fd/0x340 [lov] [ 4848.458606] [&lt;ffffffffa058debe&gt;] lov_attr_get+0x1e/0x60 [lov] [ 4848.459007] [&lt;ffffffffa0cabb48&gt;] cl_object_attr_get+0x78/0x180 [obdclass] [ 4848.459400] [&lt;ffffffffa0ab7636&gt;] ccc_lock_state+0x196/0x3b0 [lustre] [ 4848.459786] [&lt;ffffffffa0cb2da8&gt;] cl_lock_state_signal+0x68/0x160 [obdclass] [ 4848.460114] [&lt;ffffffffa0cb2f24&gt;] cl_lock_state_set+0x84/0x1d0 [obdclass] [ 4848.460514] [&lt;ffffffffa0cb3c82&gt;] cl_wait_try+0x122/0x310 [obdclass] [ 4848.460890] [&lt;ffffffffa0cb7d95&gt;] cl_wait+0x65/0x270 [obdclass] [ 4848.461301] [&lt;ffffffffa0cbe388&gt;] cl_io_lock+0x498/0x5b0 [obdclass] [ 4848.461694] [&lt;ffffffffa0cbe542&gt;] cl_io_loop+0xa2/0x1b0 [obdclass] [ 4848.461994] [&lt;ffffffffa0a6c22b&gt;] ll_file_io_generic+0x42b/0x550 [lustre] [ 4848.462377] [&lt;ffffffffa0a6d11c&gt;] ll_file_aio_write+0x13c/0x2c0 [lustre] [ 4848.462715] [&lt;ffffffffa0a6d409&gt;] ll_file_write+0x169/0x2a0 [lustre] [ 4848.463053] [&lt;ffffffff8117b2e8&gt;] vfs_write+0xb8/0x1a0 [ 4848.463057] [&lt;ffffffff810d5192&gt;] ? audit_syscall_entry+0x272/0x2a0 [ 4848.463060] [&lt;ffffffff8117bbb1&gt;] sys_write+0x51/0x90 [ 4848.463064] [&lt;ffffffff8100b0f2&gt;] system_call_fastpath+0x16/0x1b [ 4848.463068] [ 4848.463552] Kernel panic - not syncing: LBUG [ 4848.463554] Pid: 380, comm: mmap_sanity Not tainted 2.6.32-debug #6 [ 4848.463555] Call Trace: [ 4848.463559] [&lt;ffffffff814f75e4&gt;] ? panic+0xa0/0x168 [ 4848.463569] [&lt;ffffffffa0b34f7b&gt;] ? lbug_with_loc+0x9b/0xb0 [libcfs] [ 4848.463582] [&lt;ffffffffa059008d&gt;] ? lov_attr_get_raid0+0x2fd/0x340 [lov] [ 4848.463590] [&lt;ffffffffa058debe&gt;] ? lov_attr_get+0x1e/0x60 [lov] [ 4848.463611] [&lt;ffffffffa0cabb48&gt;] ? cl_object_attr_get+0x78/0x180 [obdclass] [ 4848.463625] [&lt;ffffffffa0ab7636&gt;] ? ccc_lock_state+0x196/0x3b0 [lustre] [ 4848.463643] [&lt;ffffffffa0cb2da8&gt;] ? cl_lock_state_signal+0x68/0x160 [obdclass] [ 4848.463660] [&lt;ffffffffa0cb2f24&gt;] ? cl_lock_state_set+0x84/0x1d0 [obdclass] [ 4848.463677] [&lt;ffffffffa0cb3c82&gt;] ? cl_wait_try+0x122/0x310 [obdclass] [ 4848.463694] [&lt;ffffffffa0cb7d95&gt;] ? cl_wait+0x65/0x270 [obdclass] [ 4848.463711] [&lt;ffffffffa0cbe388&gt;] ? cl_io_lock+0x498/0x5b0 [obdclass] [ 4848.463727] [&lt;ffffffffa0cbe542&gt;] ? cl_io_loop+0xa2/0x1b0 [obdclass] [ 4848.463736] [&lt;ffffffffa0a6c22b&gt;] ? ll_file_io_generic+0x42b/0x550 [lustre] [ 4848.463746] [&lt;ffffffffa0a6d11c&gt;] ? ll_file_aio_write+0x13c/0x2c0 [lustre] [ 4848.463755] [&lt;ffffffffa0a6d409&gt;] ? ll_file_write+0x169/0x2a0 [lustre] [ 4848.463757] [&lt;ffffffff8117b2e8&gt;] ? vfs_write+0xb8/0x1a0 [ 4848.463759] [&lt;ffffffff810d5192&gt;] ? audit_syscall_entry+0x272/0x2a0 [ 4848.463760] [&lt;ffffffff8117bbb1&gt;] ? sys_write+0x51/0x90 [ 4848.463763] [&lt;ffffffff8100b0f2&gt;] ? system_call_fastpath+0x16/0x1b </pre> </div></div> LU-2117

sanityn test 18 lov_attr_get_raid0()) ASSERTION( atomic_read(&lsm->lsm_refc) > 1 )

Bug Blocker Resolved Fixed WC Triage Oleg Drokin Tue, 9 Oct 2012 01:43:24 +0000 Tue, 27 Nov 2012 15:58:31 +0000 Tue, 6 Nov 2012 01:52:01 +0000 0 4 <p>Had this happen again after 22 hours of running sanityn.sh in a loop</p> <p>Probably same as</p> <p>The root cause of this problem is that when a cl_lock belonging to other object is requested(this happens for mmap case), the stripe's refcount may not be greater than 1 if there is no pending IO for it.</p> <p>Probably this assertion is too restricted here. Anyway, the layout can't go away if there exists locks. So I think we can fix this by removing this assertion.</p> <p>This issue was imported by the layout lock patch on the client side, so it will be fixed in <a href="https://jira.whamcloud.com/browse/LU-1876" title="Layout Lock Server Patch Landings to Master" class="issue-link" data-issue-key="LU-1876"><del>LU-1876</del></a>(patch <a href="http://review.whamcloud.com/4416" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/4416</a>)</p> Duplicate Related Development Rank 1|hzv9qn: Rank (Obsolete) 5112 Severity