I just did an upgrade on the lustre servers from 1.6.6 to 1.6.7 using the patched kernels. There was no quota enabled in the older version and I was trying to enable quota in the new version. Just to be sure, I did a "e2fsck -f" on mgs, mdt and all OSTs and there was no filesystem errors reported. I got the following errors while trying to run a "lfs quotacheck -ug /lustre": Mar 3 16:01:49 lustre4 kernel: LustreError: 10001:0:(client.c:547:ptlrpc_prep_req_pool()) ASSERTION(imp != LP_POISON) failed Mar 3 16:01:49 lustre4 kernel: LustreError: 10001:0:(client.c:547:ptlrpc_prep_req_pool()) LBUG Mar 3 16:01:49 lustre4 kernel: Lustre: 10001:0:(linux-debug.c:222:libcfs_debug_dumpstack()) showing stack for process 10001 Mar 3 16:01:49 lustre4 kernel: quotacheck R running task 0 10001 1 10000 (L-TLB) Mar 3 16:01:49 lustre4 kernel: 0000000000000000 0000000000000001 0000000000000086 0000000000000001 Mar 3 16:01:49 lustre4 kernel: ffff810109e8b9d0 ffffffff8002e15a 0000000000000046 ffff810109e8ba60 Mar 3 16:01:49 lustre4 kernel: ffffffff887923b0 ffffffff8049a5c0 0000000000000060 ffffffff887ad930 Mar 3 16:01:49 lustre4 kernel: Call Trace: Mar 3 16:01:49 lustre4 kernel: [<ffffffff8002e15a>] __wake_up+0x38/0x4f Mar 3 16:01:49 lustre4 kernel: [<ffffffff8009ddc3>] autoremove_wake_function+0x9/0x2e Mar 3 16:01:49 lustre4 kernel: [<ffffffff800891f6>] __wake_up_common+0x3e/0x68 Mar 3 16:01:49 lustre4 kernel: [<ffffffff8008fd2b>] vprintk+0x290/0x2dc Mar 3 16:01:49 lustre4 last message repeated 2 times Mar 3 16:01:49 lustre4 kernel: [<ffffffff8008fdc9>] printk+0x52/0xbd Mar 3 16:01:49 lustre4 kernel: [<ffffffff800891f6>] __wake_up_common+0x3e/0x68 Mar 3 16:01:49 lustre4 kernel: [<ffffffff800891f6>] __wake_up_common+0x3e/0x68 Mar 3 16:01:49 lustre4 kernel: [<ffffffff8002e15a>] __wake_up+0x38/0x4f Mar 3 16:01:49 lustre4 kernel: [<ffffffff800a54eb>] kallsyms_lookup+0xc2/0x17b Mar 3 16:01:49 lustre4 last message repeated 3 times Mar 3 16:01:49 lustre4 kernel: [<ffffffff8006b77d>] printk_address+0x9f/0xab Mar 3 16:01:49 lustre4 kernel: [<ffffffff8008fd00>] vprintk+0x265/0x2dc Mar 3 16:01:49 lustre4 kernel: [<ffffffff8008fdc9>] printk+0x52/0xbd Mar 3 16:01:49 lustre4 kernel: [<ffffffff800a3072>] module_text_address+0x33/0x3c Mar 3 16:01:49 lustre4 kernel: [<ffffffff8009c34c>] kernel_text_address+0x1a/0x26 Mar 3 16:01:49 lustre4 kernel: [<ffffffff8006b463>] dump_trace+0x211/0x23a Mar 3 16:01:49 lustre4 kernel: [<ffffffff8006b4c0>] show_trace+0x34/0x47 Mar 3 16:01:49 lustre4 kernel: [<ffffffff8006b5c5>] _show_stack+0xdb/0xea Mar 3 16:01:49 lustre4 kernel: [<ffffffff88787ada>] :libcfs:lbug_with_loc+0x7a/0xd0 Mar 3 16:01:49 lustre4 kernel: [<ffffffff8878fc40>] :libcfs:tracefile_init+0x0/0x110 Mar 3 16:01:49 lustre4 kernel: [<ffffffff888d2739>] :ptlrpc:ptlrpc_prep_req_pool+0xc9/0x6b0 Mar 3 16:01:49 lustre4 kernel: [<ffffffff8002ca21>] mntput_no_expire+0x19/0x89 Mar 3 16:01:49 lustre4 kernel: [<ffffffff888d2d31>] :ptlrpc:ptlrpc_prep_req+0x11/0x20 Mar 3 16:01:49 lustre4 kernel: [<ffffffff889db43e>] :lquota:target_quotacheck_thread+0x15e/0x3e0 Mar 3 16:01:49 lustre4 kernel: [<ffffffff8004dc3f>] lock_hrtimer_base+0x26/0x4c Mar 3 16:01:49 lustre4 kernel: [<ffffffff80034df7>] mm_release+0x21/0x8e Mar 3 16:01:49 lustre4 kernel: [<ffffffff800157de>] do_exit+0x8e5/0x8f1 Mar 3 16:01:49 lustre4 kernel: [<ffffffff8005dfb1>] child_rip+0xa/0x11 Mar 3 16:01:49 lustre4 kernel: [<ffffffff889db2e0>] :lquota:target_quotacheck_thread+0x0/0x3e0 Mar 3 16:01:49 lustre4 kernel: [<ffffffff8005dfa7>] child_rip+0x0/0x11 Mar 3 16:01:49 lustre4 kernel: Mar 3 16:01:49 lustre4 kernel: LustreError: dumping log to /tmp/lustre-log.1236117709.10001 Mar 3 16:02:03 lustre4 kernel: LustreError: 9999:0:(client.c:547:ptlrpc_prep_req_pool()) ASSERTION(imp != LP_POISON) failed Mar 3 16:02:03 lustre4 kernel: LustreError: 9999:0:(client.c:547:ptlrpc_prep_req_pool()) LBUG Mar 3 16:02:03 lustre4 kernel: Lustre: 9999:0:(linux-debug.c:222:libcfs_debug_dumpstack()) showing stack for process 9999 Mar 3 16:02:03 lustre4 kernel: quotacheck R running task 0 9999 1 10000 9997 (L-TLB) Mar 3 16:02:03 lustre4 kernel: 0000000000000000 0000000000000001 0000000000000086 0000000000000001 Mar 3 16:02:03 lustre4 kernel: ffff81010a5119d0 ffffffff8002e15a 0000000000000046 ffff81010a511a60 Mar 3 16:02:03 lustre4 kernel: ffffffff887923b0 ffffffff8049a5be 000000000000005e ffffffff887ad930 Mar 3 16:02:03 lustre4 kernel: Call Trace: Mar 3 16:02:03 lustre4 kernel: [<ffffffff8002e15a>] __wake_up+0x38/0x4f Mar 3 16:02:03 lustre4 kernel: [<ffffffff8009ddc3>] autoremove_wake_function+0x9/0x2e Mar 3 16:02:03 lustre4 kernel: [<ffffffff800891f6>] __wake_up_common+0x3e/0x68 Mar 3 16:02:03 lustre4 kernel: [<ffffffff8008fd2b>] vprintk+0x290/0x2dc Mar 3 16:02:03 lustre4 last message repeated 2 times Mar 3 16:02:03 lustre4 kernel: [<ffffffff8008fdc9>] printk+0x52/0xbd Mar 3 16:02:03 lustre4 kernel: [<ffffffff800891f6>] __wake_up_common+0x3e/0x68 Mar 3 16:02:03 lustre4 kernel: [<ffffffff800891f6>] __wake_up_common+0x3e/0x68 Mar 3 16:02:03 lustre4 kernel: [<ffffffff8002e15a>] __wake_up+0x38/0x4f Mar 3 16:02:03 lustre4 kernel: [<ffffffff800a54eb>] kallsyms_lookup+0xc2/0x17b Mar 3 16:02:03 lustre4 last message repeated 3 times Mar 3 16:02:03 lustre4 kernel: [<ffffffff8006b77d>] printk_address+0x9f/0xab Mar 3 16:02:03 lustre4 kernel: [<ffffffff8008fd00>] vprintk+0x265/0x2dc Mar 3 16:02:03 lustre4 kernel: [<ffffffff8008fdc9>] printk+0x52/0xbd Mar 3 16:02:03 lustre4 kernel: [<ffffffff800a3072>] module_text_address+0x33/0x3c Mar 3 16:02:03 lustre4 kernel: [<ffffffff8009c34c>] kernel_text_address+0x1a/0x26 Mar 3 16:02:03 lustre4 kernel: [<ffffffff8006b463>] dump_trace+0x211/0x23a Mar 3 16:02:03 lustre4 kernel: [<ffffffff8006b4c0>] show_trace+0x34/0x47 Mar 3 16:02:03 lustre4 kernel: [<ffffffff8006b5c5>] _show_stack+0xdb/0xea Mar 3 16:02:03 lustre4 kernel: [<ffffffff88787ada>] :libcfs:lbug_with_loc+0x7a/0xd0 Mar 3 16:02:03 lustre4 kernel: [<ffffffff8878fc40>] :libcfs:tracefile_init+0x0/0x110 Mar 3 16:02:03 lustre4 kernel: [<ffffffff888d2739>] :ptlrpc:ptlrpc_prep_req_pool+0xc9/0x6b0 Mar 3 16:02:03 lustre4 kernel: [<ffffffff8002ca21>] mntput_no_expire+0x19/0x89 Mar 3 16:02:03 lustre4 kernel: [<ffffffff888d2d31>] :ptlrpc:ptlrpc_prep_req+0x11/0x20 Mar 3 16:02:03 lustre4 kernel: [<ffffffff889db43e>] :lquota:target_quotacheck_thread+0x15e/0x3e0 Mar 3 16:02:03 lustre4 kernel: [<ffffffff8005dfb1>] child_rip+0xa/0x11 Mar 3 16:02:03 lustre4 kernel: [<ffffffff889db2e0>] :lquota:target_quotacheck_thread+0x0/0x3e0 Mar 3 16:02:03 lustre4 kernel: [<ffffffff8005dfa7>] child_rip+0x0/0x11 Mar 3 16:02:03 lustre4 kernel: Mar 3 16:02:03 lustre4 kernel: LustreError: dumping log to /tmp/lustre-log.1236117723.9999 Thanks Nirmal
Johann Lombardi
2009-Mar-06 23:55 UTC
[Lustre-discuss] Possible quota bug in version 1.6.7
On Tue, Mar 03, 2009 at 05:21:24PM -0600, Nirmal Seenu wrote:> I just did an upgrade on the lustre servers from 1.6.6 to 1.6.7 using > the patched kernels. There was no quota enabled in the older version and > I was trying to enable quota in the new version. > > Just to be sure, I did a "e2fsck -f" on mgs, mdt and all OSTs and there > was no filesystem errors reported. I got the following errors while > trying to run a "lfs quotacheck -ug /lustre": > > Mar 3 16:01:49 lustre4 kernel: LustreError: > 10001:0:(client.c:547:ptlrpc_prep_req_pool()) ASSERTION(imp != > LP_POISON) failedBugzilla ticket #18126 has a patch to fix this problem. Cheers, Johann