I'm running Networker 7.6.1.7 on a Sun T2000 running Solaris 10 with five
Storage Nodes and 25 Dedicated Storage Nodes. We use two Brocade switches to
connect the backup fabric. I have four Quantum i6000 libraries with 48, 47, 16
and 8 drives (IBM LTO-3, -4 and -5 drives). I was using Quantum media when the
"prematurely full" messages began back in April 2011. We are now using TDK
media and the problem has become worse. Due to our requirements, we primarily
use new tape media. We've replaced about eight tape drives, and removed about
25 tapes from circulation. We've had several hundred tape drives "marked full
prematurely". This problem appears primarily on our two libraries that have the
highest tape usage (3000+/month between the two of them).
I am told that Quantum and IBM have an issue with their firmware in how it
handles tape errors. The new TDK media has a problem with tape tension (like
most new media) and "dust" from the manufacturing process. Apparently, the
firmware is a little too sensitive to tape errors that aren't really errors and
the "dust" requires the tape drives to be cleaned much more often (several
times a month). And we all know how enthusiastic Networker is on marking tapes
as bad on drive errors. Our short-term fix is our script that ejects tapes now
changes all tapes that are marked prematurely full back to appendable. On the
fourth such error, we eject the tape and discard it. We have removed about six
tapes this way. The remainder of the tapes have been reused without error.
I will say that Quantum has been very responsive in determining a root cause to
this issue.
YMMV
Paul Gilles
NAVTEQ
-----Original Message-----
From: EMC NetWorker discussion [mailto:NETWORKER AT listserv.temple DOT edu] On
Behalf Of STANLEY R. HORWITZ
Sent: Friday, July 29, 2011 9:19 AM
To: NETWORKER AT listserv.temple DOT edu
Subject: [Networker] Tape prematurely marked full
Perhaps some of you can help me uncover a mystery. I have a Spectra T50e
tape library with two LTO-5 tape drives with the hardware encryption
feature enabled. This tape library is used exclusively to do group-based
cloning in NetWorker 7.6 SP1 Power Edition on a Red Hat Linux 64-bit Dell
server. The drives are connected to the NetWorker server via fibre channel.
Last night, one of the three tapes that received clone data was marked
full much too soon. All I can find in the /nsr/logs/daemon.raw file is a
notice that this tape was marked full prematurely. This was the first tape
to be written to that night. I did not notice any problems with the other
two tapes that were used last night.
I am wondering why that occurred. Could this particular tape be defective?
Nothing out of the ordinary showed up in the /var/log/messages file last
night. The tape in question is only about two months old and it has no
aborted safe sets on it.
To sign off this list, send email to listserv AT listserv.temple DOT edu and
type "signoff networker" in the body of the email. Please write to
networker-request AT listserv.temple DOT edu if you have any problems with this
list. You can access the archives at
http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER
The information contained in this communication may be CONFIDENTIAL and is
intended only for the use of the recipient(s) named above. If you are not the
intended recipient, you are hereby notified that any dissemination,
distribution, or copying of this communication, or any of its contents, is
strictly prohibited. If you have received this communication in error, please
notify the sender and delete/destroy the original message and any copy of it
from your computer or paper files.
To sign off this list, send email to listserv AT listserv.temple DOT edu and
type "signoff networker" in the body of the email. Please write to
networker-request AT listserv.temple DOT edu if you have any problems with this
list. You can access the archives at
http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER
|
