디스크의 큐 수치가 높아요

[조용훈]

안녕하세요.

시스로그에는 이런 메세지가 있습니다.

=======================================================

Apr 6 17:04:47 yesan vmunix: NOTICE: Cannot open module kazndrv on path /stand/

dlkm/mod.d

Apr 6 17:04:47 yesan vmunix:

Apr 6 17:03:26 yesan ftpd: Kosseal is not available, rc 40002.

Apr 6 17:04:47 yesan above message repeats 2 times

Apr 6 17:04:47 yesan vmunix: gio_mod_cdev_open: modld failed: retval = 0x2. Ret

urning ENODEV

Apr 6 17:04:47 yesan ftpd: init Kosseal not avail, rc 40002.

Apr 6 17:04:47 yesan vmunix: NOTICE: Cannot open module kazndrv on path /stand/

dlkm/mod.d

Apr 6 17:04:47 yesan vmunix:

Apr 6 17:04:47 yesan ftpd: Kosseal is not available, rc 40002.

===========================================



이것과 연관이 있는지는 모르지만



glance 에서 보니 다음과 같습니다.



===========================================

B3692A GlancePlus C.03.55.00 17:14: yesan 9000/800 Current Avg High

--------------------------------------------------------------------------------

CPU Util SSU U | 15% % 72%

Disk Util F | 1% 1% 30%

Mem Util S SU UB B | 99% 99% 99%

Swap Util U UR R | 55% 55% 55%

--------------------------------------------------------------------------------

IO BY DISK Users= 9

Idx Device Util Qlen KB/Sec Logl IO Phys IO

--------------------------------------------------------------------------------

1 0/0/2/0.6.0 0.0 / na/ na 1.

2 0/0/2/1.6.0 0.0 / na/ na .

3 0/5/0/0.1.1 0/ 0 0.0 0.0/ 0.0 na/ na 0.0/ 0.0

4 0/4/0/0.0.1 0/ 0 0.0 0.0/ 0.0 na/ na 0.0/ 0.0

5 0/5/0/0.1.7 0/ 0 0.0 0.0/ na/ na 0.0/ 0.

6 0/4/0/0.0.7 0/ 0 0.0 0.0/ 0.0 na/ na 0.0/ 0.0

7 0/4/0/0.0.6 / na/ na .

8 0/5/0/0.1.6 0/ 0 0.0 0.0/ 0.0 na/ na 0.0/ 0.0

9 0/5/0/0.1.2 4381. / na/ na .

10 0/4/0/0.0.2 0/ 0 0.0 0.0/ 0.0 na/ na 0.0/ 0.0

============================================



qlen 떨어지질 않고 계속 4381 상태 입니다.



디스크는 AutoRaid 가 구성 되어 있습니다.



어떤 조치를 취해야 하는 상황 인지요?

[이태곤]

안녕하세요

itrc에 다음과 같은내용이 있습니다. 참고하십시오

(결론적으로 December 2004 Support Plus 을 이용하여 patch하시거나

PHKL_31164와 DLKM 관련patch을 진행해보십시오)



Mar 21 11:54:00 ACL01 vmunix: gio_mod_cdev_open: modld failed:

retval = 0x2e. Returning ENODEV:

Mar 21 11:54:00 ACL01 EMS : ------ EMS Event Notification ------



Value: "error" for

Resource: "/system/kernel_resource/file_system/nflocks"

(Threshold: >= "80")

Mar 21 11:54:00 ACL01 EMS : ------ EMS Event Notification ------



Value: "error" for

Resource: "/system/kernel_resource/misc/ncallout" (Threshold:

>= "80")







KRM polls these values every 5 minutes. On the first polling after the

ioscan finishes the problem will correct itself.



Temporary Solution



TEXT

workaround

This problem can be avoided by using the kmsystem command to make

the "krm" kernel module be statically linked into the kernel rather

than dynamically loaded.



"kmsystem -l n krm"



The kernel will need to be built for this change to take effect.



Fix Text



fix

This fix is available in the December 2004 Support Plus media for 11.11.



Since the DLKM load will fail while an ioscan is running, this change

allows for that failure. krmond will return 0 usage for ncallout and

nflock for the first 12 DLKM load failures. This generally gives

ioscan an hour to complete.



===============

Oct 25 17:16:01 skicss01 vmunix: gio_mod_cdev_open: modld failed:

retval = 0x2e. Returning ENODEV



For 2), module name/info and reason of failure is not clear. The error

can happen due to gio lock failure or an error in modules' _load

function.

The gio lock error is timing related error, thus customer can simply

retry load later. For error in the module, the developer of the module

need to be involved.



Currently, we need to create and install instrumental to get

information

to identify the cause of the problem. If product code is enhanced to

print these information, that would help to isolate problem quickly.