Running out of tape prematurely

[ITsmyfault]

Hi -
Using BackUp Exec for Netware v9 build 4172 with a DLT8000. OS is Netware 5.1

I have 2 different jobs that run sequentially each night. The first job is set to "overwrite", the 2nd is set to "append". The total job is around 50GB at the moment. A DLT8000 should give me ~80GB, but I am running out of tape (so says BUE) at around 44-45GB. I am not sure where to look for the issue. Some nights it works, some nights it tells me I'm out of tape. We are doing full backups every night, don't use pools or anything. Just a very simple backup job.

Anyone have a clue why BUE thinks I am running out of tape?

 

[NewJack73]

Is the backup tapes 40G uncompressed and 80G compressed? Also check to see if you have the backup set to run using compressing (I have hardware compression set on ours)

 

[ITsmyfault]

Hi - and thanks for the reply!

tapes are 40/80, compression is set to hardware. Block size is 64k.

 

[Qdog]

I am not familiar with BE for Netware. I am however thoroughly familiar w/BE for NT/2000 and I believe what I am about to say would be valid in either arena as the concepts are the same.

Check to see if software compression gets more on the tape. If it gets more or a lot more, you may need to upgrade or reapply firmware to your tape drive.

However, I would immediately check your tape drive statistics, particularly soft write errors. Each time a soft write error occurs, the tape drive fast forwards the tape to get past what it thinks is a bad spot on the tape. If you are getting more than a few of these errors, the tape drive is fast-forwarding, hence you have less tape to write data to. Check the same error statistics for your tapes as well. If it is just one tape, try a new tape. But if you see a lot of these errors for the tape drive, you need to try cleaning the drive at least twice and then test some backups. If the errors continue after that, consider getting the drive repaired/replaced as it indicates problems with the write heads.

I have seen the above situation too many times.

Hope this helps.

 

[ITsmyfault]

Hi - Great info! I am going to try software compression as I don't see errors on the tape.. (took a while to find that menu!)

Here's the stats from last night (when it ran out again.. with a brand new tape)

Media Totals Media Errors
Bytes Written: 31208.0 MB Seek: 0
Bytes Read: 0.0 MB Soft Read: 0
Total Mounts: 3 Hard Read: 0
Hours in use: 3 Soft Write: 0
Remaining bytes: n/a HardWrite: 0
Current Capacity: n/a

For anyone else looking for this, IMHO the easiest place to find it is on the Admin console on the server.
MEDIA-->MEDIA VIEW-->choose your media set-->choose your media
Enter on the media you want to look at. Arrow down to Statistics and hit enter.

 

[PalmStrike]

Hi there, I am having the same problem here, yet I have the compression set to Hardware, if not available then software. is this wrong then.

This backup drive(and server in fact) is only 6 months old, and was backing up fine with no problems till one day, I wasn't in, so didn't put in a tape for that day, since then it will not back up fully. I can back up the information store, or the Microsoft Exchange mailboxes, but not both, please help, even veritas can't seem to help. I know this is just a simple problem, as the previouse IT bloke used to just reset something and Bob's yer aunties live in lover, but this is driving mer mad.

 

[ITsmyfault]

odd bit to add.. after I posted my last bit with stat's from the server, I checked (just for S&G) the stats from the mgt console. My posted stats were from the tape the prior night and so do apply, but in the mgt console I have this:
Total Errors:
Bytes Written: 2691.5GB Soft Read: 113
BytesRead: 121.3GB Hard Read: 3
Total Mounts: 139 Soft Write: 3,812,089
Hours in use: 0 Hard Write: 0

Something seems to be going on.. have switched to SW compression and am going to clean the thing a couple times and see what gives. Hours in use is obviously wrong.. not sure what that means..

Anyone have a secret method to determining the best block size? Ours is at the dflt of 64k.

 

[Qdog]

ITsmyfault:

That is a boatload of softwrite errors!!! What is happening: your tape drive is fast-forwarding every time it encounters what it thinks is a bad spot on the tape. This occurs each time a soft-write error occurs.

These errors are precursors to a tape drive failure. Try cleaing the drive. If the errors don't stop, contact the tape hardware vendor/manufacturer and get it repaired or replaced. Because if cleaning the drive doesn't help, then you probably have damaged write heads.

Good luck!

 

[johnny99]

Just my small comment.
What I do at our shop is that I only care about the uncompressed capacity. Anything extra is nice but I don't count on it.
In our installation the total compression ration is 1,86 in the windows env. but we have volumes that compres 1:1 but others compress 1:22.
My "rule of thumb" is:
Windows overall compresion: 1:1,5
Unix overall compression: 1:2,5
OS/390 overall compression: 1:4,0

On a 40/80GB drive the part that says 80 is marketing, and as always take it for what it is: Marketing