Anantha,
I was hoping to see a lot less trace records than that. Was DTrace
running the whole time or did you start it just before you saw the
problem?
Can you sieve thru the trace to see if you can see any subsequent
firings whose timestamp differences are big? (e.g. > 1s).
You can try this awk script:
awk '{ if ($4 - lastts > 1000000000) { print last,($4 - lastts);
print $0 } last=$0; lastts = $4}' < trace_file
This should zoom down on the exact function that is stalling the write.
It does seem from your mdb traces that the issue is something to do
with the intent log.
On 9/13/06, Anantha N. Srirama <[EMAIL PROTECTED]> wrote:
I ran the DTrace script and the resulting output is rather large (1 million
lines and 65MB), so I won't burden this forum with that much data. Here are the
top 100 lines from the DTrace output. Let me know if you need the full output
and I'll figure out a way for the group to get it.
dtrace: description 'fbt:zfs::' matched 2404 probes
CPU FUNCTION
520 -> zfs_lookup 2929705866442880
520 -> zfs_zaccess 2929705866448160
520 -> zfs_zaccess_common 2929705866451840
520 -> zfs_acl_node_read 2929705866455040
...
--
Just me,
Wire ...
_______________________________________________
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
