Hello,
When issuing a "MOUNT DATABASE" command, we are seeing that it takes longer than expected.
I captured the following diagnostics:
1> sp_monitor enable
2> go
Drop and recreate control table 'tempdb.dbo.mon_config'
(return status = 0)
1> sp_monitor "connection", "diskio"
2> go
spid LoginName ElapsedTime CPU_Time Physical_Reads LocksHeld SQLText
---- --------- ----------- -------- -------------- --------- -------
(0 rows affected)
(return status = 0)
1> sp_monitor "event", "276"
2> go
WaitTime Description
-------- -------------------------------------------
41077 waiting for regular buffer read to complete
(1 row affected)
(return status = 0)
1> sp_monitor "event", "276"
2> go
WaitTime Description
-------- -------------------------------------------
57733 waiting for regular buffer read to complete
58 waiting on run queue after sleep
(2 rows affected)
(return status = 0)
1> dbcc traceon(3604)
2> go
DBCC execution completed. If DBCC printed error messages, contact a user with System Administrator (SA) role.
1> dbcc stacktrace(276)
2> go
Start dbcc stacktrace...
pc: 0x000000000151621c upsleepgeneric+0x4cc()
pc: 0x0000000000d08b3d bufread+0x26d()
pc: 0x0000000000a8db33 getpage_with_validation+0x733()
pc: 0x0000000000dd2e5f pg_get_alloc_page+0x7f()
[Handler pc: 0x0x00000000011a4d23 std_handle installed by the following function:-]
pc: 0x000000000139b3e6 rec_reset_allocnode+0x126()
[Handler pc: 0x0x00000000013a0e24 rec_handle installed by the following function:-]
pc: 0x0000000001392e1b recovery+0x7fb()
[Handler pc: 0x0x00000000011a4c3f hdl_backout_msg installed by the following function:-]
pc: 0x00000000014348f3 umt_recover_offline+0x193()
pc: 0x0000000001434b02 umt_recover_dbs+0x57()
[Handler pc: 0x0x000000000142da50 umt_handle installed by the following function:-]
pc: 0x000000000143538b mountdb+0x3ed()
pc: 0x0000000001109323 s_execute+0x74b3()
[Handler pc: 0x0x00000000011a46f0 hdl_stack installed by the following function:-]
[Handler pc: 0x0x0000000001140670 s_handle installed by the following function:-]
pc: 0x00000000011432af sequencer+0x86f()
pc: 0x00000000007b427c tdsrecv_language+0x1ec()
[Handler pc: 0x0x000000000143d230 ut_handle installed by the following function:-]
pc: 0x00000000007d4490 conn_hdlr+0x1080()
dbcc stacktrace finished.
DBCC execution completed. If DBCC printed error messages, contact a user with System Administrator (SA) role.
I saw the WaitTime continue to increase to over 100,000. Is this telling us that ASE is waiting for a network/disk read to complete (the database devices are mounted over NFS)?
Or does this tell us there are ASE configuration issues we should look into? If so, any hints on what else we should look at?
Thanks,
Neal