With both drives "going full" at the same time, it sounds like the
connection to the drives is failing, through either a "stall" somewhere in
the fabric or router, or possibly SCSI resets issued by a server on the
SAN.  Some things to check:

- is SAN Switch zoned such that tape drives are visible to ONLY the (3)
HBA's that need to see them?
- is Removable Storage Manager service disabled on the Windows Storage
Nodes?
- is SCSI Reserve/Release set in st.conf on Solaris host?
- latest models of QLogic HBA's have a field named "Target Reset" - this
should be disabled on HBA's for tape - SANBlade Control is config utility I
was using on Windows.  Also allows for Persistent Binding, though this
shouldn't be cause of your current problem, only at boot-up.

HTH

On Fri, 2 May 2003 23:52:48 -0400, Tremblay, Frederic
<Frederic.Tremblay AT SPACE.GC DOT CA> wrote:

>Dear Networker-l,
>We hae some problems with our new installation. Here's the configuration:
>
>Server:
>Sun Enterprise 450
>Solaris 8
>Legato 6.1.3 power edition
>ADIC scalar 10K library (SCSI attached) with 4 SDLT320 tape drives (SAN
>attached thru a SNC Pathlight bridge).
>
>
>Clients:
>2 Dell powerEdge 2550
>2 x QLogic PCI card (one configured as primary path for the disks and the
>other for the tapes)
>PV660 SAN
>6.1.3 client configured as storage node.
>Latest drivers of SDLT320.
>
>The tapes are not ugons full capacity. We are getting those messages during
>the backup
>
>May  2 23:29:05 libra root: [ID 702911 daemon.notice] NetWorker media:
>(warning) rd=eureka.csa.space.gc.ca:\\.\Tape2 writing: The semaphore
timeout
>period has expired., at file 13 record 5792
>May  2 23:29:05 libra root: [ID 702911 daemon.notice] NetWorker media:
>(notice) sdlt320 tape 000002S on rd=eureka.csa.space.gc.ca:\\.\Tape2 is
full
>May  2 23:29:05 libra root: [ID 702911 daemon.notice] NetWorker media:
>(notice) sdlt320 tape 000002S used 23 GB of 160 GB capacity
>May  2 23:30:04 libra root: [ID 702911 daemon.notice] NetWorker media:
>(warning) rd=eureka.csa.space.gc.ca:\\.\Tape1 writing: The semaphore
timeout
>period has expired., at file 15 record 13008
>May  2 23:30:04 libra root: [ID 702911 daemon.notice] NetWorker media:
>(notice) sdlt320 tape 000003S used 28 GB of 160 GB capacity
>May  2 23:30:04 libra root: [ID 702911 daemon.notice] NetWorker media:
>(notice) sdlt320 tape 000003S on rd=eureka.csa.space.gc.ca:\\.\Tape1 is
full
>
>Any ideas or suggestions?
>
>Frederic Tremblay
>> Canadian Space Agency

--
Note: To sign off this list, send a "signoff networker" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=