I have two identical clustered servers with identical compaq 35/70 drives and identical scsi cards. Backed up data on both drives is about 35gb. The problem is the first cluster takes about 17-18hrs to backup and the second cluster takes about 11-12hrs. I can't understand why there is such a big difference.
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!
Slow backup on NT cluster server
- Thread starter accobra
- Start date
Are you running the clusters as active-active, or active-passive?
Because of the pathetic non-clustered nature of a cluster aware BENT installation, it runs like a pig when you have BENT running on a cluster member that is not hosting the virtual server with the data being backed up. It also runs slower than non cluster aware installations when backing up locally hosted data.
If you are running active-active on your clusters & the distribution of data is not even between the cluster members, the cluster with BENT running on the server hosting the larger portion of the data will run much faster.
Because of the pathetic non-clustered nature of a cluster aware BENT installation, it runs like a pig when you have BENT running on a cluster member that is not hosting the virtual server with the data being backed up. It also runs slower than non cluster aware installations when backing up locally hosted data.
If you are running active-active on your clusters & the distribution of data is not even between the cluster members, the cluster with BENT running on the server hosting the larger portion of the data will run much faster.
ACCOBRA:
--------
- what you might want to do is check the following:
- check the buffer and block size settings in the devices tab within the backup console. You would want the block size to be 32K and the same for buffer.
- i very much doubt it has anything to do with backup exec be it installed as cluster-aware or not.
- other factors you might want to check are:
- SCSI drivers > update them
- check to see if the clustered resources you are backing up are fragmented.
- check the system logs and see if you are getting any I/O issues (ID 7,9,11 and 15's in the system logs) for your tape drive and also for your HDD's.
- disable you virus checker for one test Backup and see if the thruput increases.
- are you using OFO (open file option) when running your backup. If so, OFO requires a certain amount of minimum quiet time on your HDD's before it snaps your volumes, so for your second cluster with the longer backup time there could of been more activity compared to the first.
- what are you backing up ie. if you are backing up Exchange mailboxes, your thruput will drop.
- check your NIC drivers , update if required
- if backup exec is installed outside the cluster, try put the NIC on the cluster node and on the Media server on 1/2 duplex and see what result that has
--------
- what you might want to do is check the following:
- check the buffer and block size settings in the devices tab within the backup console. You would want the block size to be 32K and the same for buffer.
- i very much doubt it has anything to do with backup exec be it installed as cluster-aware or not.
- other factors you might want to check are:
- SCSI drivers > update them
- check to see if the clustered resources you are backing up are fragmented.
- check the system logs and see if you are getting any I/O issues (ID 7,9,11 and 15's in the system logs) for your tape drive and also for your HDD's.
- disable you virus checker for one test Backup and see if the thruput increases.
- are you using OFO (open file option) when running your backup. If so, OFO requires a certain amount of minimum quiet time on your HDD's before it snaps your volumes, so for your second cluster with the longer backup time there could of been more activity compared to the first.
- what are you backing up ie. if you are backing up Exchange mailboxes, your thruput will drop.
- check your NIC drivers , update if required
- if backup exec is installed outside the cluster, try put the NIC on the cluster node and on the Media server on 1/2 duplex and see what result that has
- Status
Similar threads
- Locked
- Question