Component -------------------Summary Description------------------- Qty
PD Disks experiencing a high level of I/O per second 93
Component --Identifier-- ---------Detailed Description----------
PD disk:100 Disk is experiencing a high level of I/O per second:
789.0
pd Suggested Action 3
This check samples the I/O per second (IOPS) information in statpd to see if any Drives are being
overworked, and then it samples again after five seconds. This does not necessarily indicate a problem,
but it could negatively affect system performance. The IOPS thresholds currently set for this condition are:
• NL drives < 75 IOPS
• FC 10K RPM drives < 150 IOPS
• FC 15K RPM drives < 200 IOPS
• SSD < 12000 IOPS
Operations such as servicemag and tunevv can cause this condition. If the IOPS rate is very high
and/or a large number of Drives are experiencing very heavy I/O, examine the system further using
statistical monitoring commands/utilities such as statpd, the SSMC (GUI) and System Reporter. The
following example reports Drives whose total I/O is 150/sec or more.
cli%
statpd -filt curs,t,iops,150
14:51:49 11/03/09 r/w I/O per second KBytes per sec ... Idle %
ID Port Cur Avg Max Cur Avg Max ... Cur Avg
100 3:2:1 t 658 664 666 172563 174007 174618 ... 6 6
pd Example 4
Component --Identifier-- -------Detailed Description----------
PD disk:3 Detailed State: old_firmware
pd Suggested Action 4
The identified Drive does not have firmware that the storage system considers current. When a Drive is
replaced, the servicemag operation should upgrade the Drive's firmware. When Drives are installed or
added to a system, the admithw command can perform the firmware upgrade. Check the state of the
Drive using CLI commands such as showpd -s , showpd -i , and showfirmwaredb.
84 Troubleshooting
To Next Page
Loading...
