Hi,My system is not booting and at the startup it is getting struck.In HMC error code is coming as 0000, I know the reason of failing.I have few queries on recovery, please answer:1. I have mksysb of the system from which I can restore the system but problem is my few application mount point was a part of rootvg and while taking the mksysb we have a practice to exclude those mount points.2. Now I know I can easily restore the system from mksysb but what about those mount points how can I restore them.3. what I thought is I have a spare disk so I will restore the mksysb on that and then attach these disk of rootvg having those application filesystem and restore them.4. but I am not sure about exact things need to be done.Please provide your viewThanks
mksysb is basically a "savevg rootvg", so there is no need to exclude such mount points because the backup process would stop on VG boundaries anyways. If you haven't included the mount points you will have to manually recreate them. Write a script which does that for you, complete with ownership, rights, filemodes, etc., so you could use it to restore them in case you need them again.
I hope this helps.
bakunin
Thanks for your reply bakunin, but you haven't answered my requirement.
See I have a mksysb which is not containing my application mount point as it was taken with excluding those mount point due to size contraints.
Now I want to recover the system with those mount points. System is not booting.
This is the scenario, now pls suggest how can I restore my system along with those mount points. I have a spare disks on which I can restore the mksysb, but then what abt those application mount point, how I will be able to recover them from those failed disk.
I have answered parts of your requirement, namely, how to restore mountpoints which haven't been backep up. The answer is: you can only restore what you have backed up before, so there is no way to retrieve them from the backup at all.
And i was telling you that this is a moot point and in fact nonsense: if the application data you mount into these mountpoints are not in the rootvg (which has to be the case, otherwise you would have to install the application anew anyways, mountpoints or not, because you have excluded it from the backup) you will only enlarge the backup by what it takes to store a bunch of directory entries - hardly more than a few bytes. I don't think this is any relevant constraint.
Sounds likely to be impossible. This is a SysAdmin forum, not the Uri Geller Academy. I am a technician, not a psychic.
Why, do you think, the disk has failed in first place? Because it is working?
If - but that is a big "if" - you get the disk to cooperate somehow: restore what you have from the mksysb to the new disk, boot from it, do an "importvg" of the old disk and have a look where the mountpoints are, then recreate them by hand. Maybe this will fail because you will try to import a rootvg into a system which already has one. I don't know and i can't recreate this problem here, because i have no system where i could try to import a foreign rootvg.
If this fails do the following: boot the system from an external medium into maintenance mode, there is an option of "open a maintenance shell" (or similar). Activate the volume group on the defect disk, then open such a maintenance shell and have a look. Write your mountpoints down, restore the system to the new disk from your mksysb and create the noted mountpoints.
I hope this helps.
bakunin
Sounds like you did the incredibly bad practice of putting all of your data in rootvg and then just excluding the filesystems from mksysb's using /etc/exclude.rootvg. Like bakunin said, if that is the case, the data cannot be recovered via MKSYSB. Maybe you have TSM, netbackup, etc? Maybe you don't understand what a MKSYSB is and your data actually resides on a different volume group?
If you are posting here about how to recover your down system, you must not have an IBM contract. This is the price you pay for not knowing how to recover your systems BEFORE they crash. Time to bite the bullet and pay IBM for support on this.
If you would share your knowledge with us, we might be able to help you recovering your system without restoring from backups - since it is apparently NOT a failed disk?
Anyways, you should be able to restore a plain rootvg onto your spare disk and than import the 'old' rootvg under another name / VGID ... AIX will rename the logical volumes but if you recreate the empty mountpoints, you should be able to mount the filesystems manually.
Kind regards
zxmaus