How to isolate a bad dimm by command on Solaris 10 host?


 
Thread Tools Search this Thread
Operating Systems Solaris How to isolate a bad dimm by command on Solaris 10 host?
# 1  
Old 06-11-2013
How to isolate a bad dimm by command on Solaris 10 host?

Hello,

I have a HP ProLiant DL385 ( X86 ) running Solaris 10 on it.

Our hardware team passwd by server last night and noticed an amber light to indicate a possible bad dimm.

/var/adm/messages, dmesg, prtdiag -v, all shows nothing.

/opt/HPQhealth/sbin/hpasmcli indicated I have a bad dimm at module 2:

Cartridge #: 0
Module #: 2
Present: Yes
Form Factor: 9h
Memory Type: 12h
Size: 2048 MB
Speed: 400 MHz
Status: Dimm is degraded

Since this server has 8 memory and 2 CPUs, customer is requesting to isolate this bad dimm and keep the server running for production applications.

System Configuration: HP ProLiant DL385 G1
BIOS Configuration: HP A05 03/01/2006
==== Processor Sockets ====================================
Version Location Tag
-------------------------------- --------------------------
Opteron Node 1
Opteron Node 2
==== Memory Device Sockets ================================
Type Status Set Device Locator Bank Locator
------- ------ --- ------------------- --------------------
DDR in use 1 DIMM 01
DDR in use 1 DIMM 02 --> I think this is the bad dimm
DDR in use 2 DIMM 03
DDR in use 2 DIMM 04
DDR in use 3 DIMM 05
DDR in use 3 DIMM 06
DDR in use 4 DIMM 07
DDR in use 4 DIMM 08


Does anyone know how to isolate DIMM 02 online by either Solaris 10 command or HP command?

Thank you very much,

SC
# 2  
Old 06-11-2013
what do you mean with "isolate"?
# 3  
Old 06-11-2013
Hello,

What I meant is to "isolate" the bad dimm from OS Level, so the OS will not see/use this bad dimm and continue to run with 7 other good dimm ( without bringing system down ).

SC
# 4  
Old 06-11-2013
i don't think this is possible in a x86 system... the only command comming to my mind for dynamic reconfiguration is cfgadm.
# 5  
Old 06-11-2013
Thank you very much. Yeah, when I was running "cfgadm -al", I can't even see memory info like what usually will shows up on a SPARC host.

Seems like HP has there own tool called "ProLiant Memory Configuration Tool", which I don't have on my server either for configuration.

Thank you very much.
# 6  
Old 06-11-2013
Quote:
Originally Posted by sunnychen98
What I meant is to "isolate" the bad dimm from OS Level, so the OS will not see/use this bad dimm and continue to run with 7 other good dimm ( without bringing system down ).
Even if a command allows disabling that dimm, that would be impossible to run with 7 of them.
Six would be the best option as they must be installed in pairs.
Login or Register to Ask a Question

Previous Thread | Next Thread

9 More Discussions You Might Find Interesting

1. Solaris

Migration of Solaris 10 on physical host to Solaris Zones

Hi All Kindly let me know how can I move Solaris 10 OS running update 10 on physical machine to another machine solaris zone running Solaris 10 update 11 (2 Replies)
Discussion started by: amity
2 Replies

2. Shell Programming and Scripting

Store and isolate bad pages from a file to new file

I have a file like below . The good pages must have 3 conditions : The pages that containing page total only must have 50 lines. The pages that containing customer total only must have 53 lines. The last page of Customer Total should be the last page. How can I accomplish separating good... (1 Reply)
Discussion started by: ehabaziz2001
1 Replies

3. IP Networking

ping can not recognize host but host command can

Hi, I have a weird problem. when ever I do ping command like for example ping unix.comI get the following message: # ping unix.com ping: unknown host unix.com but when I use host the computer is able to know the host. # host unix.com unix.com has address 81.17.242.186 unix.com mail is... (2 Replies)
Discussion started by: programAngel
2 Replies

4. Solaris

Solaris 8.2 Bad magic number

I'll keep it fairly straight forward. I work with a Solaris server and magically today it decided to take a dump on me. At first it give a long list of files that couldn't be acessed before terminating the boot process and returning to the 'ok' prompt. Booting in single-user mode allowed me to run... (4 Replies)
Discussion started by: Aon
4 Replies

5. Solaris

Solaris x86: Bad PBR sig

Hello everyone, I have a Sun x4600 (Solaris 10 x86) server.During boot up "Bad PBR sig" error comes soon after post (before GRUB). Unless the return key is pressed the server never boots up. Once the return key is pressed the boot process continues without any problem. Can you please help me... (1 Reply)
Discussion started by: pingmeback
1 Replies

6. HP-UX

Memory dimm status

Hi, How to check the memory dimm status from OS leavel!! My system model HP N4000. Thanks, ARumugam. (2 Replies)
Discussion started by: arumsun
2 Replies

7. Solaris

PING - Unknown host 127.0.0.1, Unknown host localhost - Solaris 10

Hello, I have a problem - I created a chrooted jail for one user. When I'm logged in as root, everything work fine, but when I'm logged in as a chrooted user - I have many problems: 1. When I execute the command ping, I get weird results: bash-3.00$ usr/sbin/ping localhost ... (4 Replies)
Discussion started by: Przemek
4 Replies

8. Solaris

solaris error BAD SUPER BLOCK

I want mount a disk. I have this error. I'm trying to correct with the superblock but i have the same error. Look my procedure. bash-2.03# fsck -F ufs /dev/rdsk/c0t1d0s0 Alternate super block location: 9423392. ** /dev/rdsk/c0t1d0s0 BAD SUPER BLOCK: MAGIC NUMBER WRONG USE AN ALTERNATE... (1 Reply)
Discussion started by: simquest
1 Replies

9. Solaris

PBR Bad in sun solaris 10 intel version

Hai while booting sun solaris 10 in a intel machine,its showing PBR bad and its showing disk boot failure.but the is detecting.please guide me (1 Reply)
Discussion started by: subbiahvin
1 Replies
Login or Register to Ask a Question