I have found the following bug which I can easily reproduce. Right now I'm live debugging it with a sun developer. Here is a link to the bug in bugtracker (contains ::threadlist -v output as attachment) and some mdb output:
http://defect.opensolaris.org/bz/show_bug.cgi?id=3690 [EMAIL PROTECTED]:~# echo "dc0e7600::findstack -v" | mdb -k stack pointer for thread dc0e7600: da867d14 da867d44 swtch+0x195() da867d54 cv_wait+0x53(daa759aa, daa75968, ffffffff, 0) da867d94 txg_wait_open+0x90(daa75800, a511, 0, 20000) da867dd4 dmu_tx_wait+0xdd(d61ec020) da867e64 zfs_write+0x35f(d6b76ec0, da867efc, 0, da5fe898, 0, dab874d8) da867eb4 fop_write+0x4a(d6b76ec0, da867efc, 0, da5fe898, 0, 1) da867f44 write+0x2bb(4, 8080000, 20000, da867fac, fe800c57, 0) da867f84 dtrace_systrace_syscall+0xc8() da867fac sys_sysenter+0x106() [EMAIL PROTECTED]:~# echo "daa75800::print dsl_pool_t dp_tx" | mdb -k { dp_tx.tx_cpu = 0xdb523100 dp_tx.tx_sync_lock = { _opaque = [ 0, 0 ] } dp_tx.tx_suspend = { _opaque = [ 0 ] } dp_tx.tx_open_txg = 0xa510 dp_tx.tx_quiesced_txg = 0xa50f dp_tx.tx_syncing_txg = 0xa50e dp_tx.tx_synced_txg = 0xa50d dp_tx.tx_sync_txg_waiting = 0xa38c dp_tx.tx_quiesce_txg_waiting = 0xa511 dp_tx.tx_sync_more_cv = { _opaque = 0 } dp_tx.tx_sync_done_cv = { _opaque = 0 } dp_tx.tx_quiesce_more_cv = { _opaque = 0x1 } dp_tx.tx_quiesce_done_cv = { _opaque = 0x1 } dp_tx.tx_timeout_cv = { _opaque = 0 } dp_tx.tx_exit_cv = { _opaque = 0 } dp_tx.tx_threads = 0x2 dp_tx.tx_exiting = 0 dp_tx.tx_sync_thread = 0xdbfbbde0 dp_tx.tx_quiesce_thread = 0xdbe84de0 dp_tx.tx_timelimit_thread = 0 } Regards, Björn -- This message posted from opensolaris.org _______________________________________________ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
