Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations strongm on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

BackupExec 9.0, failures and reboots ( a lot of info here)

Status
Not open for further replies.

esmithbda

IS-IT--Management
Jun 10, 2003
304
US
I've recently taken over as the new IT/programming person at a small company. My background is pretty extensive in programming, and I'm trying to get back up to speed in terms of networking and server admin tasks.

We have a Dell PowerEdge 1400SC that is our primary domain controller (Win2K), Exchange Server, and our main backup machine. There are some other programs on there as well - mostly just in house finance software running against Access DBs.
There had been Symmantec's (sp?) NAV Corp Ed, but it was removed in hopes that it would solve some problems that I am about to get into. It still has the NAV for Exchange on it.

We had version 8.5 of Backup Exec on the machine and it got to the point where it was rebooting everyday.
Contrary to many posts I have seen on here, it wasn't rebooting only during backups, but instead just at arbitrary times (sometimes during backups, but for the most part not so). As I understand it, this system worked fine, and then at some point "just started doing this".
They (prior to me getting here, they have had a few consultants coming in and out taking a look at this) thought it was probably a conflict between the NAV and the backup software. They took the NAV off the system. They still had the problems.
They then upgraded the backup exec to 9.0, but just the eval version to see if that would help.

Sure enough, it would stay up a long time (over 27 days as opposed to rebooting every day).

Then right when I got here, the evaluation period expired and we needed the serial numbers (for Windows servers and for the Exchange Agent) for it to continue to work.
One of the consultants gave us a temp set of serial numbers to cover us until our new software came with the serials.
The backups worked fine and it stayed up for more than a day, but still it crashed within 3 days.
I had read (perhaps on here) that it could be the "write checksums to media" option needed to be unchecked (apparently some issue between the Dell motherboard and that options or something).
With that unchecked, we still got successful backups, and we still eventually got reboots after a day or two.

We then got our upgrade serials from Veritas. I put those in and it didn't like them because we had version 9.0 software intalled, and it wanted to see 8.5 - so I called Veritas and they gave us the correct full serial numbers (for Win Servers and for Exchange). During that process, they asked if we had any databases on that machine - being new, I wasn't sure off the top of my head and said no.

I put those new serials in and restarted the services. It now will fail on the backup and it also restarts the machine.
The failure doesn't list any errors, so I'm not sure where it went wrong - it does tell us that the report generated as xml isn't formatted correctly.

I read on here from "technome" that there could be a memory leak with VBE, so I setup the scheduler today to turn on our services at 9:20pm and then stop them at 1am (at 9:30PM it runs the inventory which takes 16 mins, and then at 10PM it runs the full backup which takes almost 2.5 hours).
I will see tomorrow (and the rest of the month I guess) if that helps the restart issue at all.

I looked at the logs and when we were getting backup failures before, it would do a successful inventory and then fail on the backup saying our eval period was up.
When we were successful, it would do the inventory and then do the backup - the backup was about 23 gigs.
The most current failures do the inventory successfully, and then the backups look like they have about 6 gigs and they complete in 1 min with a failure (but it doesn't tell us what that failure is in the log).
That made me think that it was missing some part of our system (23 gigs when it is right, and 6 gigs when wrong) - so I thought perhaps it is because we *do* have databases on this machine (MS SQL Server).
So I contacted the same Veritas person that helped us get the new serial numbers and told her that I actually do have databases on this machine.
She said that would involve a whole separate package and we would have to buy and license that. In the past we never had that, so I don't think that is our issue right now.

And then as I was at the server trying to work this out, a coworker walked by and told me that in the past, another problem that we sometimes have (it hasn't been up long enough lately for us to see this problem recently) is that it will run at 10pm and say that it failed... and then it will finish sometime between 11am and 4pm the next day and be successful - so you would see a log that says that it failed, and then at the end of it say that it was completed.

So I am pretty stumped on this one. We don't have the money nor the time to convert over to a different software product to do this for us, so we need to resolve this.

The reboots happen at pretty much any time - frequently late at night (after 2am) - nobody is in our ofice and the network traffic is pretty much nil, and the load on the machine is very low.
It was (in the past) also successfully completing the backup and then later at some point restarting. So it isn't a direct issue of the backup itself.

If anyone has *any* help at all on this, I'd be grateful - this is causing many headaches and I already have about 20 things aside from this that I need to resolve and would like to be able to move on from the backup issue.

Eric Smith, Meridian Corporate Services
 
It probably IS the database issue. Backup Exec 9.0 now creates it's own SQL Server database (the MSDE version) unless you already have SQL Server then it wants to be part of that.

-SQLBill
 
Well, just to make matters more interesting, it stayed up this past night and successfully backed up... no clue.

I'm guessing that it stayed up due to the bestart.bat and bestop.bat files that are scheduled to start and stop it only around backup times.

As for why it didn't fail this time, but did yesterday? No clue.

I hope it stays up this way, I have many other things that I need to get resolved now.

 
So is this still an issue for you? I have a server here doing the same thing and haven't been able to pinpoint exactly what might be causing the reboots. So was the stopping and starting the BE services the fix?

-GESwihart
 
It was an issue for us, yes.

I don't know if there is a way to do a search on all of the posts by a user on TechTips, but if there is - track what I have posted on here and probably 25% of my posts are issues with BE.

The start and stop scripts fixed it for us for awhile, but then we moved the mailserver to a different physical machine, but the BE software remained on the machine it had always been on.

This caused it to go into a state where no script would kill it and it would put up a fight if you tried to shut it down unless you killed its processes via the task manager and then one of them had to be done via the services panel (but not all would work from there).

We brought in an outside consultant that seems to be fairly confused with it as well - after two weeks now it looks like he has it backing up to an external disk and I think maybe the tapes are successful now too.
I know he did another upgrade as well - perhaps 9.1?


I have had so little luck/success with it that at this point I try to avoid it as much as possible.
 
my veritas has been failing since the time change (that may be just a coincidence) it was working relatively fine before that. the back up would normally take place at about 11 pm and end at 3 ish. but when I would come in in the morning it would not have been successful then I would try to manually do it and it would just run and run and it looked like it was backing up but then it would fail. the last error said: the media provided to continue back up already contains data from one or more sets created during this operation. the operation cannot be continued. we've tried different tapes to no avail. any suggestions?
 
esmithbda - had a similar problem. I know this sounds out there but it ended up being power supply in the server. Went thru replacing CPU, Ram, etc. Replaced the power supply and problem went away.
 

DELL recognize a problem with some of there POWER EDGE model. The 1st level support doent know about it, if you persist to have a 2nd level support they will be glad to change your motherboard. I experienced it!
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top