有一台blade2000的机器，平时用的时候总感觉不对劲，有是登陆或执行命令时出现停顿。
看了一下messages,显示：
NOTICE: Qlogic qlc(0): Loop ONLINE
Mar 16 15:22:07 blade2a qlc: [ID 686697 kern.info] NOTICE: Qlogic qlc(0): Loop ONLINE
Mar 17 02:01:15 blade2a qlc: [ID 686697 kern.info] NOTICE: Qlogic qlc(0): Loop OFFLINE
Mar 17 02:01:15 blade2a qlc: [ID 686697 kern.info] NOTICE: Qlogic qlc(0): Loop ONLINE
Mar 17 03:30:02 blade2a qlc: [ID 686697 kern.info] NOTICE: Qlogic qlc(0): Loop OFFLINE
Mar 17 03:30:02 blade2a qlc: [ID 686697 kern.info] NOTICE: Qlogic qlc(0): Loop ONLINE
Mar 17 04:55:47 blade2a qlc: [ID 686697 kern.info] NOTICE: Qlogic qlc(0): Loop OFFLINE
Mar 17 04:55:47 blade2a qlc: [ID 686697 kern.info] NOTICE: Qlogic qlc(0): Loop ONLINE
Mar 17 09:20:24 blade2a last message repeated 1 time
Mar 17 15:10:04 blade2a qlc: [ID 686697 kern.info] NOTICE: Qlogic qlc(0): Loop ONLINE
Mar 18 02:01:15 blade2a last message repeated 1 time

到google上搜了一下，看到一篇文章，有一个人也遇到类似的问题，贴出了一写解决方法：
I called SUN and was told I need
patch 111097-12 along with the SUNWsan (San Foundation Kit) package and
patches 111095-13 (or newer) fctl/fp/fcp/usoc driver patch
111096-06 (or newer) fcip driver and adb macros patch
111412-11 (or newer) mpxio driver patch

along with these (This is from the 111097 README file)

我就奇怪了：我也没用Storage Array之类的东西，装了它就能解决吗？

唉，没人帮忙，还是自己查吧

找到了，还是要自力更生
http://sunsolve.sun.com/pub-cgi/retrieve.p...&zone_110=47942

What is happening is powerd is spinning down the disks. The time that it seems to hang is when the disks are spinning back up.

There are two ways to fix this problem.

1. To disable power management (powerd) altogether, perform the following:

#cd /etc/rc2.d

#mv S85power s85power

#init 6


2. Start the Power Manager GUI (DTpower)

Select the "More" option

Select the "Disks" option and change the setting to "Always On"

Change "Device Idle Time" to "Always On"

Click "OK", The new settings should not spin down the disks.

果然不再有延迟了

学习学习.