Hi,
I am very new to AIX, i have a development and a production server with AIX 6.1. I am using following command to backup my system
/usr/bin/mksysb '-m' '-V' '-i' /dev/rmt0
You can what i have in my root volume group in the atached file and and you can also see the backup information of the backup which i take on the tape.
My problem is, backup takes more than one hour to complete and the size of backup you can see is 19G. Why my backup size is this much big and it is taking this much time? On my production server, it takes 3 hours to complete and backup size is around 40G. Why is it so?
I just want to backup my System so that i can recover it in case of any failure.
Thanks
I am taking backup with verbose and will be uploading next time, dafinately i will be using -v option for this..
Can you explain this question "Are there any errors on rmt0 in errpt?"
I didnt see any error message when it checks tape drives readbility
I am first time doing this type of sys admin guy. I am a DBA actualy
Please see result of command you asked me to run in attached file
2) change the block_size of rmt0 to 1024 if it is listed as 512 and rerurn the backup. add the -b 200 option to mksysb.
chdev -l rmt0 -a block_size=1024
3) if the compress attribute is no, then check it to yes and re-run the backup.
chdev -l rmt0 -a compress=yes
4) if 2 or 3 don't apply then I would suggest that you exclude the /ml-fixes file system from your backup. This should reduce the backup time by almost 50%. You can always backup that file system with another method without slowing down the mksysb backup or worse case a system restore. re-run the backup and report the times if this is ok.
echo /ml-fixes >> /etc/exclude.rootvg
Hi,
Thanks for your suggestions. I would be applyin what you just said but i didn't understand your phrase "ok. your backup is only ~4.5G which isnt to bad."
My curren tbackup size is 19.3 G in compressed format and 27.6 G.
Please see the output of the command you mentioned in point 1st, in the attached file.
I have excluded /ml-fixes. And it is taking around 50 minutes to backup.
But my main concern is the size of my backup which is still 19.3 in compressed forat which i can see if i execute the command
The /dev/rmt0.1 is the non-rewind device of the tape drive rmt0 and has nothing to do with your problems as when the tape is ejected it will be rewound anyway. I will have a look at what you have said in the past and see if there is something I have missed.
Not knowing anything about the different environments I cannot comment on why the mksysb from one system takes a lot longer than the other. So many factors can effect the way a mksysb is run. This could be as simple as the SCSI device the tape drive is attached to or the data rate at which the data is written to the tape. All you need to worry about is whether the mksysb completes without errors!
Also a check in the error log to see if there are any failures against the rmt0 tape drive.
Try the command lsvg rootvg and check the total used PPs and post the result here.
Another important fact to take into account is this is a mksysb so it is not wise to exclude anything as the system might be unstable when re-built.
My concern is the size of the backup, not the time it is taking.
See the result bellow of command "listvgbackup -l -f '/dev/rmt0'" and have a look at the size of backup in output
like I said in my previous post, the backup size is 4.5 GB.
The "BACKUP SIZE" you see in the output is the total size (unused + data) to restore the system. This includes the paging space size, dump spaces, etc... which are not part of the mksysb backup. This is what you would need at minimum if you were to do a direct clone of the system without shrinking file systems or modifying the image.data file.