The connection to the data to backup has been lost. Backup set cancel 1

[TecGuy]

Greetings Everyone,

I am running into a problem with Veritas Backup Exec 8.6 for NT and Windows 2000. This server was backing up fine last week. I know nothing has changed as far as Hardware goes. The only thing changed may be more files add to this machine. (Its our main file server) I have set it to do Detailed logging but still it only shows.

The connection to the data to backup has been lost. Backup set canceled.
^ ^ ^ ^ ^

Its backing up at 33.9 MB when the job fails. The weird part is it backups up the C: drive no problem.. It then starts on the D: Drive and crashes at random spots on the d: drive.

Anyone run into this.. Any other things I should try. I am currently skipping the ACT DB directorys and Heavy Directory structures as well as MAS 90 Directory. All this to try and see if the problem is from one of these.


HELP!

Any Ideas..

 

[TecGuy]

Just an Update to this..

I changed the backup job to just backup User Drives and Project Shares.. ETc.. and still it hangs with connection lost error message.

Hmmmm..

 

[TecGuy]

Getting The following in event viewer.

Event ID: 57745
Source: Backup Exec.

Job "Full Backup" ended in an error state. Examine recent event log entries and the job log for details.

 

[SQLBill]

Have you tried 'breaking' your backup job into two parts? One job to backup the C drive files and one to backup the D drive files?

-Bill

 

[TecGuy]

I have broke the Job into 6 parts.. The c: drive as 1. Then different directories such as User shares, Group Shares, Project Shares etc..

I found that it crashed on 2 of the 6 parts. I'll be taking a deeper look into those as we speak..

More information on those failed jobs to follow..

 

[SQLBill]

Check permissions on the files that caused the crashes. It's possible the backup didn't have permission to access those files. Also, are you using Open File Option? Do the failed parts have open files and the others don't?

-Bill

 

[TecGuy]

This Job before I split it, would backup a random amount before crashing. That leads me to think that its not a permissions issue.

Its backedup 3gb then crashed.. 5gb then crashed.. its done 15gb. 13gb all crashing before the full backup job is complete. This drive has about 23 GB of used space.

I've already uped the agent timeouts.

We are not useing the Open file Option. We maybe very soon. We are looking at purhcasing that option as well as the IDR option.

The job logs dont tell you anything.. Just ..

The connection to the data to backup has been lost. Backup set canceled.
^ ^ ^ ^ ^

You would think there would be a little more information.

 

[cybersaur]

Assuming D: is an internal HDD and not a network resource, is D: a different type of drive from C:? I'm thinking you may have a driver issue. Especially if D: is a SCSI device (different from C.
Something to consider.

Trent
Network Admin

 

[ActionJacks]

Try

http://seer.support.veritas.com/docs/240501.htm

its got some details on increasing backup exec's send and receive timeouts.

-Action

 

[TecGuy]

Thank you.

I appreciate the link. I have however already increased the time on the agents.

As far as the Drives. They are part of a Raid 5 configuration. C and D both. All SCSI. from the same drives. D is local not a remote connection.

 

[l1nder]

I had this problem and resolved it by setting my NIC card from "auto-negotiate" to "100MB/FULL" -- assuming that is correct for your backbone.

 

[leongodwin]

I also have exactly the same problem. I have 10 servers that backup fine and 1 that fails with:

The connection to the data to backup has been lost. Backup set canceled.

Before it fails throughput drops dramtically. I have tried reinstalling the agent, and seperating the probelm server into its own job. Backup takes place across a fibre HBA, but i do not think this is the source of the problem as I can manually move data through the fibre card at high speeds, and all the other servers have exactly the same config.

I have also modified the tiomeouts, and still no joy.

 

[Adm793]

Hey Techguy,

I am still new to Backup exec, but I after a few weeks reponding back and forth with their techs, I am still having the same problems as you have stated here. I am curios if you were ever able to resolve the issue.

I have four servers that I am backing up and the jobs only fail on my file server, and only on two of the three drives. Always on random files and stating that a remote connection was lost. I have previosly tried all solutions listed within this forum with no avail, along with many more from veritas. I am having no luck.

Any help would be greatly aprreciated.

Thankyou

 

[TecGuy]

Greetings Everyone,

yes this has been resolved.

To correct the problem set all your Nic's and Switch (dont forget the switch) to 100mb full duplex.

I was changing the Nic's and forgot to change the Switch. This resolved my problems and backup runs great.. No more crashes.

Again,

Change the NIC and Switch that the Servers connect to. TO 100mb Full duplex.

Problem Resovled.

Hope that helps.
Thanks,
TecGUY