I have a pretty standard ZFS boot AMD64 desktop. A the moment, most ZFS related commands are hanging (can''t be killed) . Running ''truss share'' the last few lines I see are: ioctl(3, ZFS_IOC_OBJSET_STATS, 0x08044860) = 0 brk(0x080C1000) = 0 ioctl(3, ZFS_IOC_POOL_STATS, 0x080447E0) = 0 ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08045880) = 0 ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08045880) = 0 ioctl(3, ZFS_IOC_OBJSET_STATS, 0x080437E0) = 0 ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08044800) Err#3 ESRCH ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08045880) = 0 ioctl(3, ZFS_IOC_OBJSET_STATS, 0x080437E0) = 0 ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08044800) Err#3 ESRCH ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08045880) = 0 ioctl(3, ZFS_IOC_OBJSET_STATS, 0x080437E0) = 0 ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08044800) = 0 ioctl(3, ZFS_IOC_OBJSET_STATS, 0x08042760) = 0 ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08043780) Err#3 ESRCH ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08044800) = 0 ioctl(3, ZFS_IOC_OBJSET_STATS, 0x08042760) = 0 ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08043780) Err#3 ESRCH ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08044800) = 0 ioctl(3, ZFS_IOC_OBJSET_STATS, 0x08042760) = 0 ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08043780) Err#3 ESRCH ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08044800) = 0 ioctl(3, ZFS_IOC_OBJSET_STATS, 0x08042760) = 0 ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08043780) Err#3 ESRCH ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08044800) Err#3 ESRCH ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08045880) = 0 ioctl(3, ZFS_IOC_OBJSET_STATS, 0x080437E0) = 0 ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08045880) = 0 ioctl(3, ZFS_IOC_OBJSET_STATS, 0x080437E0) = 0 ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08044800) = 0 ioctl(3, ZFS_IOC_OBJSET_STATS, 0x08042760) = 0 ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08043780) Err#3 ESRCH ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08044800) Err#3 ESRCH ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08045880) = 0 ioctl(3, ZFS_IOC_OBJSET_STATS, 0x080437E0) = 0 ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08044800) Err#3 ESRCH ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08045880) = 0 ioctl(3, ZFS_IOC_OBJSET_STATS, 0x080437E0) = 0 ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08044800) = 0 ioctl(3, ZFS_IOC_OBJSET_STATS, 0x08042760) = 0 ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08043780) Err#3 ESRCH ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08044800) Err#3 ESRCH ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08045880) = 0 ioctl(3, ZFS_IOC_OBJSET_STATS, 0x080437E0) = 0 ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08045880) = 0 ioctl(3, ZFS_IOC_OBJSET_STATS, 0x080437E0) = 0 ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08044800) Err#3 ESRCH ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08045880) = 0 ioctl(3, ZFS_IOC_DATASET_LIST_NEXT, 0x08045880) Err#3 ESRCH Where fd 3 came from open("/dev/zfs", O_RDWR) = 3
Ian Collins wrote:> I have a pretty standard ZFS boot AMD64 desktop. A the moment, most ZFS > related commands are hanging (can''t be killed) . Running ''truss share'' > the last few lines I see are: >Can you provide a kernel thread list report? You can use mdb -k to get that. Once in mdb type ::threadlist -v -Mark
Mark Shellenbaum wrote:> Ian Collins wrote: >> I have a pretty standard ZFS boot AMD64 desktop. A the moment, most ZFS >> related commands are hanging (can''t be killed) . Running ''truss share'' >> the last few lines I see are: >> > > Can you provide a kernel thread list report? You can use mdb -k to > get that. Once in mdb type ::threadlist -v >Here you go:> ::threadlist -vADDR PROC LWP CLS PRI WCHAN fffffffffbc29420 fffffffffbc287b0 fffffffffbc2b010 0 96 0 PC: _resume_from_idle+0xf1 CMD: sched stack pointer for thread fffffffffbc29420: fffffffffbc48d60 [ fffffffffbc48d60 _resume_from_idle+0xf1() ] swtch+0x17f() sched+0x423() main+0x419() _locore_start+0x92() ffffff0007a05c80 fffffffffbc287b0 0 0 -1 0 PC: _resume_from_idle+0xf1 THREAD: idle() stack pointer for thread ffffff0007a05c80: ffffff0007a05c00 [ ffffff0007a05c00 _resume_from_idle+0xf1() ] cpu_idle+0xe9() idle+0x10e() thread_start+8() ffffff0007a0bc80 fffffffffbc287b0 0 0 60 fffffffffbcaa8b0 PC: _resume_from_idle+0xf1 THREAD: thread_reaper() stack pointer for thread ffffff0007a0bc80: ffffff0007a0bb90 [ ffffff0007a0bb90 _resume_from_idle+0xf1() ] swtch+0x17f() cv_wait+0x61() thread_reaper+0xb7() thread_start+8() ffffff0007a11c80 fffffffffbc287b0 0 0 60 ffffff01cce37e90 PC: _resume_from_idle+0xf1 TASKQ: kmem_move_taskq stack pointer for thread ffffff0007a11c80: ffffff0007a11b60 [ ffffff0007a11b60 _resume_from_idle+0xf1() ] swtch+0x17f() cv_wait+0x61() taskq_thread+0x11e() thread_start+8() ffffff0007a17c80 fffffffffbc287b0 0 0 60 ffffff01cce37db8 PC: _resume_from_idle+0xf1 TASKQ: kmem_taskq stack pointer for thread ffffff0007a17c80: ffffff0007a17b60 [ ffffff0007a17b60 _resume_from_idle+0xf1() ] swtch+0x17f() cv_wait+0x61() taskq_thread+0x11e() thread_start+8() ffffff0007a1dc80 fffffffffbc287b0 0 0 60 ffffff01cce37ce0 PC: _resume_from_idle+0xf1 TASKQ: pseudo_nexus_enum_tq stack pointer for thread ffffff0007a1dc80: ffffff0007a1db60 [ ffffff0007a1db60 _resume_from_idle+0xf1() ] swtch+0x17f() cv_wait+0x61() taskq_thread+0x11e() thread_start+8() ffffff0007a23c80 fffffffffbc287b0 0 0 60 ffffff01cce37c08 PC: _resume_from_idle+0xf1 TASKQ: scsi_vhci_nexus_enum_tq stack pointer for thread ffffff0007a23c80: ffffff0007a23b60 [ ffffff0007a23b60 _resume_from_idle+0xf1() ] swtch+0x17f() cv_wait+0x61() Ian
Ian Collins wrote:> Mark Shellenbaum wrote: >> Ian Collins wrote: >>> I have a pretty standard ZFS boot AMD64 desktop. A the moment, most ZFS >>> related commands are hanging (can''t be killed) . Running ''truss share'' >>> the last few lines I see are: >>> >> Can you provide a kernel thread list report? You can use mdb -k to >> get that. Once in mdb type ::threadlist -v >> > Here you go: >That doesn''t appear to be a complete list. We need to see the ZFS threads.>> ::threadlist -v > ADDR PROC LWP CLS PRI WCHAN > fffffffffbc29420 fffffffffbc287b0 fffffffffbc2b010 0 96 0 > PC: _resume_from_idle+0xf1 CMD: sched > stack pointer for thread fffffffffbc29420: fffffffffbc48d60 > [ fffffffffbc48d60 _resume_from_idle+0xf1() ] > swtch+0x17f() > sched+0x423() > main+0x419() > _locore_start+0x92() > > ffffff0007a05c80 fffffffffbc287b0 0 0 -1 0 > PC: _resume_from_idle+0xf1 THREAD: idle() > stack pointer for thread ffffff0007a05c80: ffffff0007a05c00 > [ ffffff0007a05c00 _resume_from_idle+0xf1() ] > cpu_idle+0xe9() > idle+0x10e() > thread_start+8() > > ffffff0007a0bc80 fffffffffbc287b0 0 0 60 fffffffffbcaa8b0 > PC: _resume_from_idle+0xf1 THREAD: thread_reaper() > stack pointer for thread ffffff0007a0bc80: ffffff0007a0bb90 > [ ffffff0007a0bb90 _resume_from_idle+0xf1() ] > swtch+0x17f() > cv_wait+0x61() > thread_reaper+0xb7() > thread_start+8() > > ffffff0007a11c80 fffffffffbc287b0 0 0 60 ffffff01cce37e90 > PC: _resume_from_idle+0xf1 TASKQ: kmem_move_taskq > stack pointer for thread ffffff0007a11c80: ffffff0007a11b60 > [ ffffff0007a11b60 _resume_from_idle+0xf1() ] > swtch+0x17f() > cv_wait+0x61() > taskq_thread+0x11e() > thread_start+8() > > ffffff0007a17c80 fffffffffbc287b0 0 0 60 ffffff01cce37db8 > PC: _resume_from_idle+0xf1 TASKQ: kmem_taskq > stack pointer for thread ffffff0007a17c80: ffffff0007a17b60 > [ ffffff0007a17b60 _resume_from_idle+0xf1() ] > swtch+0x17f() > cv_wait+0x61() > taskq_thread+0x11e() > thread_start+8() > > ffffff0007a1dc80 fffffffffbc287b0 0 0 60 ffffff01cce37ce0 > PC: _resume_from_idle+0xf1 TASKQ: pseudo_nexus_enum_tq > stack pointer for thread ffffff0007a1dc80: ffffff0007a1db60 > [ ffffff0007a1db60 _resume_from_idle+0xf1() ] > swtch+0x17f() > cv_wait+0x61() > taskq_thread+0x11e() > thread_start+8() > > ffffff0007a23c80 fffffffffbc287b0 0 0 60 ffffff01cce37c08 > PC: _resume_from_idle+0xf1 TASKQ: scsi_vhci_nexus_enum_tq > stack pointer for thread ffffff0007a23c80: ffffff0007a23b60 > [ ffffff0007a23b60 _resume_from_idle+0xf1() ] > swtch+0x17f() > cv_wait+0x61() > > Ian > > _______________________________________________ > zfs-discuss mailing list > zfs-discuss at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/zfs-discuss