Can you make the crash dumps available?  I only see text output attached to
the bug.

Any particular reason that you're creating a pool on top of an iscsi device
which is backed by a zvol on a pool on the same machine?  Have you tried it
using 2 different machines?

--matt


On Wed, Feb 12, 2014 at 4:26 AM, Franz Schober <[email protected]>wrote:

> Hi,
>
> I have an in a few cmd lines reproducible kernel panic on a zfs send.
>
> At first the zfs send and the whole zpool blocks for some minutes, then
> the system panics.
>
> The stack trace of the zfs send process while blocking (The system is
> OmniOS r151.006y here):
>
> stack pointer for thread ffffff014a8283c0: ffffff0004cfb810
> [ ffffff0004cfb810 _resume_from_idle+0xf1() ]
> ffffff0004cfb840 swtch+0x141()
> ffffff0004cfb880 cv_wait+0x70(ffffff014f7125d2, ffffff014f712598)
> ffffff0004cfb8d0 txg_wait_synced+0x83(ffffff014f712400, e2e)
> ffffff0004cfb9e0 dsl_sync_task+0x187(ffffff0148ecc080, fffffffff79ca0f0,
> fffffffff79ca190, ffffff0004cfbaf0, 1)
> ffffff0004cfbb50 dsl_dataset_user_release_tmp+0xa5(ffffff014f712400, 32,
> ffffff0150774b90)
> ffffff0004cfbb90 dsl_dataset_user_release_onexit+0xa2(ffffff0150774a80)
> ffffff0004cfbbd0 zfs_onexit_destroy+0x43(ffffff0154991a58)
> ffffff0004cfbc00 zfs_ctldev_destroy+0x18(ffffff0154991a58, 4)
> ffffff0004cfbc60 zfsdev_close+0x89(ac00000004, 403, 2, ffffff014e6b8ad8)
> ffffff0004cfbc90 dev_close+0x31(ac00000004, 403, 2, ffffff014e6b8ad8)
> ffffff0004cfbce0 device_close+0xd8(ffffff0154bab880, 403,
> ffffff014e6b8ad8)
> ffffff0004cfbd70 spec_close+0x17b(ffffff0154bab880, 403, 1, 0,
> ffffff014e6b8ad8, 0)
> ffffff0004cfbdf0 fop_close+0x61(ffffff0154bab880, 403, 1, 0,
> ffffff014e6b8ad8, 0)
> ffffff0004cfbe30 closef+0x5e(ffffff014f360950)
> ffffff0004cfbea0 closeandsetf+0x398(8, 0)
> ffffff0004cfbec0 close+0x13(8)
> ffffff0004cfbf10 _sys_sysenter_post_swapgs+0x149()
>
> After that the kernel panics with:
>
> ffffff00062e1c40 PANIC    <NONE>                  1
>                  0xffffff01492e7040
>                  apic_send_ipi+0x73
>                  send_dirint+0x18
>                  poke_cpu+0x2a
>                  cpu_wakeup+0x9f
>                  apix_do_softint_prolog+0x59
>                  0
>                  panic
>                  mutex_vector_enter+0x367
>                  idm_conn_event+0x35
>                  idm_ini_conn_disconnect+0x18
>                  iscsi_timeout_checks+0x1eb
>                  iscsi_wd_thread+0x28
>                  iscsi_threads_entry+0x16
>                  taskq_thread+0x2d0
>                  thread_start+8
>
> (crash file is available in https://www.illumos.org/
> issues/4589#change-11778)
>
> Here are the lines to reproduce this issue:
>
> zfs create -V 2G rpool/vol1
> stmfadm create-lu /dev/zvol/rdsk/rpool/vol1
> stmfadm add-view 600144F000000000000052F8B1390001 #your lun here
> svcadm enable -s stmf
> svcadm enable -s iscsi/target itadm create-target iscsiadm add
> discovery-address 10.1.0.155:3260 iscsiadm modify discovery --sendtargets
> enable zpool create test c3t600144F000000000000052F8B1390001d0 #your
> devicename here
>
> zfs create test/testds dd if=/dev/zero of=/test/testds/file1 bs=1M
> count=50 dd if=/dev/zero of=/test/testds/file2 bs=1M count=50 zfs snapshot
> test/testds@test zfs send -R test/testds@test | pv > /dev/null
>
>
> On OmniOS r151008j the system crashed already on the zpool create at first
> test,
>
> ffffff0008c84c40 PANIC <NONE> 1
> param_preset
> mutex_panic+0x73
> mutex_vector_enter+0x367
> idm_conn_event+0x35
> idm_ini_conn_disconnect+0x18
> iscsi_timeout_checks+0x1eb
> iscsi_wd_thread+0x28
> iscsi_threads_entry+0x16
> taskq_thread+0x2d0
> thread_start+8
>
> (crash file is also available in the issue tracker mentioned above)
> on a second test the result was the same as in r151.006y.
> Thank you very much for your big help.
>
> Thx,
> Franz
>
> _______________________________________________
> developer mailing list
> [email protected]
> http://lists.open-zfs.org/mailman/listinfo/developer
>
_______________________________________________
developer mailing list
[email protected]
http://lists.open-zfs.org/mailman/listinfo/developer

Reply via email to