Search results for query: *

I recommend you work on a slower but steady process, taking a leap from 5.3 to 7.1, might come as a big surprise. Go with 6.1 1st then after a week or atleast after couple of days jump from 6.1 to 7.1 Those files should not be affected. SARFARAZ AHMED SYED, Sr. Systems Engineer

Why did you added cd /usr2/HHM/pub/SMM/transfer/request/ < - - - - - - - - - - - - Added this line You don't need that. SARFARAZ AHMED SYED, Sr. Systems Engineer

Ok, this is a very basic script. How often the file is written in the directory? If you think the file is generated every 10 minutes you can add a crontab. #!/bin/ksh DIR="/usr2/HHM/pub/SMM/transfer/request" ls -ltr $DIR | tail -1 |awk '{print $9}' |while read i do uuencode $i filename |mailx...

Did you resolved it already? Seems like you corrected it.

If your rootvg is spread across three disks, 1st off its NOT a good practice. You need to get all those LV on to a single disk. Say your 4th disk is large enough to accommodate all LV's, then you can either run migratepv orrun mklvcopy. Note: Migratepv will migrate the LV from one PV to...

Once you install MPIO, it should automatically change the reserve policy to no_reserve. But remember, it is always better (I rather say best) to have no_reserve on disks (this really matters on clusters, but always good to have). I need more details, send me the output of lsdev -Ccdisk lslpp...

Before to answer your question as YES, 1st thing, is your FC card NPIV capable? (like 8 gig card and SAN Switch support it as well). How to check? login in to VIO as padmin and run the below $lsnports If you see any FCS's here that means its capable, and if fabric is showing value '1', then...

You are doing it wrong. The disk are part of Controller (few hdisks and a CDROM), you can dedicate the Controller it one System and cannot be seen on multiple server. Ok, Here is what you do, bring up the VIO on hdisk0 Create a VG (name it what ever you want - but a relevant name) on hdisk1 #...

Can you paste the .profile output?

Hmmm, What version of AIX OS are you running? Check for "account_locked" attribute of the user unortjea, is it true or false. If it's true change it to false. SARFARAZ AHMED SYED, Sr. Systems Engineer

Are you sure they were mounted? Is that oracle DB server with raw volumes? as I am not seeing any mount points. If so, you don't have to do anything, oracle will take care of it.

You had only one fibre card or all the available cards failed? If you had only one card then, you need reboot the system, select SMS menu and option no 4, at that point the WWNS are flashed on the Fabric (Switch). If you have redundant fibre cards then, you need to replace the failed card...

It all depends on the overall architecture of the system. One just cannot held disk for at fault, you have to look from all aspects, what is the memory? what kind of disk were you using? did you correctly set the queue_depth value on the disk? how is the disk coming to system (NPIV or vscsi)...

Yes, I do see it is throwing some errors for disk, ok do this Paste the output of the below errpt -aj F7DDA124 errpt -aj 26120107

What do you see at the error report? Are hdisk1 and hdisk2 internal disks? or SAN connected? If internal check the disk in diag. If external check the path and connect with Storage team to determine the root cause.

How are you connecting? which protocol are you using? ftp, sftp, telnet??

So, whats the deal? you saw what you needed to right?

Try running istat <cifs file system name>

To shoot a mksysb (backup) on NIM a server DOESN'T have to be a NIM Client. Let us 1st concentrate on what is needed to successfully take mksysb. 1. If you are taking locally you need enough disk space. 2. If you want to take it remotely. a. On tape. b. On any NFS share (generally NIM...

If it doesn't have a tape then why are you running it on tape? If you have a NIM server, export a file system (the one you keep the local handy mksysbs). nfs mount that file system, where you want to take the backup. run # mksysb -i nfsmount/hostname.mksysb you can use -e if you want to...