Hey guys and possibly gals.

Have a bit of an issue here; I'm a windows network/system administrator and one of our clients has an IBM Server running AIX which I am not familiar with. I've used Linux in the past, so I'm not completely helpless. But before I headed out to the client I wanted to see if I could get my gameplan together.

Anyhow, the server that they have appears to have a bad hard disk. It's got a RAID array that has been mirrored and I need to know how I can test the disks to find out which of the bunch is bad. If someone can tell me the best command/way to check a disk in the AIX environment, I'm sure I can figure out the rest.

Any help will be greatly appreciated!

Thanks!

Eric

smit is the gui tool for managing EVERYTHING in AIX.

smitty gives you the text based version.

Both work well.

fsck will check the disk system, but you need to do a df from the command line and print out the output for the fsck to do you any good. fsck will report bad partition but if you don't know the what partitions are on what disk you'd be lost. Use smit or smitty and it should be self expanitory.

Good luck.

Thanks for the help, I'll give it a shot.

Ok, so I was able to get the list of disks. There are 2 SCSI drives, disk0 and disk1. I was able to generate an error report and it shows that disk0 has probably about 2 pages worth of disk operation errors, but doesn't specify what exactly. And there are also a handful for disk1.

Is there a way to more accurately find out what is wrong with said disk, and/or possibly run some sort of checkdisk like command or diagnostic?

I think ultimately we are looking at replacing one of said drives, but I will need to find out which drive is truly bad and for what reason.

Again, I appreciate the help.

Eric

fsck = file system check.

go to command line and type:
man fsck

This will show you the different options available. Possibly the fsck command by itself will work, but reading / printing out the man page should help you zero in on the problem by using the most appropriate options. I'm not near an AIX box now to provide the best options for you to use.

Ok, so the FSCK command allows you to check the different file systems that are mounted. I've done that and didn't really seem to get any helpful information.

Is there any other way to determine what is wrong with these hard disk(s)?

I don't know of any. Sounds like you have disks that are ready to fail. All the Aix servers I admin'd were produciton apps. When disks started throwing erros we replaced them. When they were under warranty IBM techs would help us get more detailed info on them. But you have to be paying for support for that.

The best way to know on which filesystem is the dirty blocks :

determine the filsystem name : lsvg (just to know the volume groups on the disks probably should be rootvg)
lsvg
rootvg
datavg

then : lsvg -l VolumeGroupName
This command shows you the list of the filesystems with their Logical Volume Name :
Logical Volume Type N of Partitions Online Filesystem
hd4 jfs 3 3 1 open/syncd /
hd2 jfs 147 147 1 open/syncd /usr
hd9var jfs 8 8 1 open/syncd /var
hd3 jfs 16 16 1 open/syncd /tmp
hd1 jfs 8 8 1 open/syncd /home

Then if you type :
lslv hd1 (to check /home)

LOGICAL VOLUME: hd1 VOLUME GROUP: rootvg
LV IDENTIFIER: 00425f7b497948fe.8 PERMISSION: read/write
VG STATE: active/complete LV STATE: opened/syncd
TYPE: jfs WRITE VERIFY: off
MAX LPs: 512 PP SIZE: 16 megabyte(s)
COPIES: 1 SCHED POLICY: parallel
LPs: 8 PPs: 8
STALE PPs: 0 BB POLICY: relocatable
INTER-POLICY: minimum RELOCATABLE: yes
INTRA-POLICY: center UPPER BOUND: 32
MOUNT POINT: /home LABEL: /home
MIRROR WRITE CONSISTENCY: on
EACH LP COPY ON A SEPARATE PV ?: yes

The Stale PP are dirty Blocks on the Filesystem !

You can do that by smitty vg / smitty lv but it's obvious