Unsolved

This post is more than 5 years old

19 Posts

1435

July 11th, 2006 11:00

Tapes marked full prematurely

Hi,

I am having Networker 7.3 server loaded on Windows 2003. I have Windows & Solaris storage nodes (SAN) and Windows and Solaris clients (LAN) also. When I am taking backup of Solaris storage nodes, the backup is working fine and the percentage of tape filling is proper. But when I am taking the backup of Windows Storage Nodes (SAN) and Windows & Solaris clients (LAN) through Networker server the tapes are getting marked full at lower capacities (say 400MB, 2GB, 10GB, etc.) and getting the alert "waiting for one writable media". I am taking the full backups of all the servers. Pls help me to solve this problem

Regards,

Shreejesh.

6 Operator

 • 

14.4K Posts

 • 

56.2K Points

July 11th, 2006 11:00

If you check the logs what is the error message you get before tape has been marked full? Given that this is windows 2003 you may wish to make sure that:
a) TURs are disabled (has been discussed in few threads)
b) for the block sizes above 64KB you will need MS patch
c) latest drive drivers is installed
d) RSM is disabled

6 Operator

 • 

14.4K Posts

 • 

56.2K Points

July 13th, 2006 12:00

Strange indeed. If you try - for a test - to use NT backup do you get the same issue? If not and if you are able to fill properly tapes using NT backup I would advise to involve your support on this one for further tests and environment inspection (if you have enough resources and time to play, try to configure win2k storage node and check if you get the same - do not use standard build usually used by your company, but rather build this machine from scratch without anything else on it except latest patch level and NW and then try to use it just to verify if you get the same issue).

19 Posts

July 13th, 2006 12:00

Hi Hrvoje,

I have tried the following activities.
1. Disabled TRU and RSM service.
2. Updated the Windows 2003 Networker server and storage nodes.
3. Updated the device drivers.

But the problem remains the same. I am facing this problem when taking backup through Win2K3 Networker server and through Win2K3 storage nodes. The backup through solaris storage nades is proper. Please guide to resolve this.

Regards,

Shreejesh.

19 Posts

July 17th, 2006 20:00

Is it possible that the tapes may have been labeled on the Solaris storage node and then loaded on the Windows storage node or server to perform the backup. The different default block sizes (unix 512kb / Win 64kb) can cause some strange results. Try mounting a tape and then verify block size under device details and/or delete a volume and relabel on the windows storage node and then start a backup.

6 Operator

 • 

14.4K Posts

 • 

56.2K Points

July 18th, 2006 01:00

Block size problem will cause error with read - unlikely with write. My understanding is that everything seems to be fine until certain moment (however, if that saveset would be first one on the tape and it always fails for first saveset then I would agree that this is bs as verification would fail at this point). The size will depend on following:
- media type defined in NetWorker
- environment variables (if used)
- HW supported block size

For example, LTO2 will have default bs 64KB so no errors should happen under default setup.

22 Posts

July 21st, 2006 06:00

Hi there,

Recently I have the same problem, I just updated the tape library firmware and the tape drive(s) firmware, and eveything is back to normal...
No Events found!

Top