On 3 Aug 2011, at 23:29, Jordan Brown wrote:
> Please try this and report the output:
>
> # mdb -k
> > ::pgrep smbd | ::walk thread | ::stacks
>
> ("c" will tell the pager to just let all of the output come out without
> stopping.)
> ::pgrep smbd | ::walk thread | ::stacks
THREAD STATE SOBJ COUNT
ffffff04006d1120 STOPPED <NONE> 11
swtch+0x145
stop+0x82e
holdlwp+0x6f
post_syscall+0x43d
syscall_exit+0x59
ffffff03fb58c0a0 STOPPED <NONE> 3
swtch+0x145
stop+0x82e
holdlwp+0x6f
trap+0x13ab
sys_rtt_common+0x68
ffffff05230cd100 SLEEP CV 1
swtch+0x145
cv_wait+0x61
exitlwps+0x1cb
psig+0x4b1
trap+0x1408
dtrace_user_probe+0x18a
ffffff0567127880 SLEEP CV 1
swtch+0x145
cv_wait+0x61
smb_server_spooldoc+0x6f
smb_drv_ioctl+0xd4
cdev_ioctl+0x45
spec_ioctl+0x5a
fop_ioctl+0x7b
ioctl+0x18e
_sys_sysenter_post_swapgs+0x149
ffffff042e5f4860 STOPPED <NONE> 1
swtch+0x145
stop+0x82e
lwp_exit+0xc2
psig+0x589
post_syscall+0x49e
syscall_exit+0x59
I'm not sure if the dtrace_user_probe on the middle thread there is important;
I *was* dtracing smbd at times before it got into this state, but dtrace is
certainly not running now.
To me it looks more like smb_server_spooldoc is waiting on something that isn't
happening.
Chris
_______________________________________________
cifs-discuss mailing list
cifs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/cifs-discuss
