Dear Experts
We have two R610 Del servers on CentOs5 (one as the main and the other as the spare), on our spare, we had some hardware problems with fans and one of the SSDs (we have two on each server) so we changed the damaged items and everything is ok now (we changed both of old 100GB SSDs with new 250GB ones), but when we restore backups and restart the spare server (main server is connected to the external RAID system but spare is not connected yet), CentOs5 give this warning during the boot sequence {The second disk "backup disk"(/dev/sdb) is not available. This disk holds the backup of the running system. you can continue here without any problems to the running system but there will be made no backups. so you should check your system and enable the second disk as soon as possible.}
what should I do to solve this problem?

Hello, and welcome here, at LQ,

without exact details hard to say anything. It's almost impossible to figure out what went wrong. Probably your backup was corrupted or was not restored properly. Or there is a problem with the raid controller or who knows?
So would be nice to look for logs and give us more information to be able to help you.
http://www.catb.org/~esr/faqs/smart-...html#beprecise

Thanks for your kind replay
Do you mean that if my backup is healthy, the second SSD should work without any error even when the size of the new SSDs is bigger, and no need to do any extra procedure to modify the new SSDs ( my First SSD is working properly and my server running now)?

Just checking, the Dell R610 was released in 2009 and it's end of service life was in October 2016. Centos 5 lived from 2007-2017. You can hardly be safe from any security threat.

If you're not missing large chunks of data, count yourself very lucky. If you are, give the faulty disk to professionals.

If this server is online, you need a backup, a new OS and preferably a new server if you're running a business. You could also consider migrating to the cloud, and save yourself capital layout & hassle.

Dear Sir
thanks for your kind replay, but my first SSD was restored completely from backup without a problem, and now this server is running as I told you without backup SSD, and the only difference that I see between these two servers is inside LVM manager on sdb and inside fdisk -l command output. on the main server, I have 3 partitions on sdb but on the spare (the one with the problem) I just have one partition on the sdb, and when I make partitions as my main server on the sdb of spare one the problem remains and even the LVM manager is not working.

Ok, if you're going to keep going that ancient system, replace the blown SSD and make regular offline backups of your data for when you get hacked. Hackers typically know their way around a system very well before they encrypt it, and the only backup that will survive mischief of that sort will be a disconnected one. In fact most of the hack may be intact in some of your backups.

I am trying to interpret this. So the server that is working correctly, it is running off external raid fs, does it see /dev/sdb? And you say LVM is not running right, how does that have anything to do with /dev/sdb? I've never seen the error message you see about backups not working, usually it isn't that detailed, like it says "Can't find /dev/sdb" and stops booting. So I think it might be some custom stuff?

If /dev/sdb is just used for backups, you should be able to comment it out of /etc/fstab and boot. BTW, I think you are using the hot swap drives. I've had some compatability issues with DELL hot swap on both servers and arrays. You can't plug any old drive in them.

Dear Sir
I know about the safety of the system, this server is not online and works inside an industrial plant, it is not possible to make any changes to the system because the provider of this system (Bruckner Co) does not support our country (IRAN) because of the sanctions, thanks for your hint.

Well, that situation narrows your choices all right.

I don't know what parts access you have but I would see a sound backup as a priority right now. I hope it works out for you.

Does your 610 have a PERC card? Disks are cheap, better to use PERC and be available (with backups as a failsafe) then be down and trying to figure how to boot everything. I usually 0+1 the os on 2 small drives, then data maybe raid 5 or 6 if the PERC has 6 support on data drives. And a healthy amount of monitoring.