This will sound bizarre, but:  Make sure you have plenty of NEW cleaning
carts in the library, and check the library to see how frequently the drives
are getting cleaned.   I have seen similar random failures where someone
reinserted a used-up cleaning cart.  Long story why, but eventually it
causes a timeout in ACSLS.

If that doesn't help, call STK/Sun and have the CE run diagnostics on the
library.  One time I had similar problems and fortunately a really, really
good STK CE.  He found that a cable had worn a bare spot, and the librayr
was getting shocked intermittently, depending on how the robot moved and
touced that cable.  Replacing the cable fixed it.

If the CE finds nothing, you'll have to open a problem with ACSLS support,
have them run a trace to find out what's happening.

Good luck!

Wanda
On Thu, Mar 4, 2010 at 12:18 PM, David McClelland <tsm AT networkc.co DOT 
uk>wrote:

> Hi Guys,
>
>
>
> I'm hoping someone with a little more ACSLS experience than I might be able
> to point me in the right direction with this one.
>
>
>
> Fundamentally, I'm witnessing seemingly random failures when mounting
> volumes in an ACSLS-managed STK L700 library. A cart which fails to mount
> in
> one operation will mount successfully two minutes later. A drive which
> failed mount a tape will also successfully mount a tape at the next
> attempt.
> From tracking this over a period of time, there appears to be no obvious
> pattern.
>
>
>
> O - TSM Server: 5.5.1.1 (Solaris 9)
>
> O - ACSLS 7.1.0
>
> O - Library STK L700 populated with 18 x IBM-LTO3 FC-attached drives
>
>
>
> The error manifests itself in the TSM Server activity log as:
>
> O - ANR8855E ACSAPI(acs_mount) response with unsuccessful status,
> status=STATUS_LIBRARY_FAILURE.
>
> O - ANR1401W Mount request denied for volume L30161 - mount failed.
>
>
>
> On the ACSLS console I see:
>
> O - 2010-03-04 16:58:26     0    LSM   0, 0: Library error, Path
> unavailable
>
>
>
> Obviously, the causes the TSM process or session requesting the mount to
> fail. However, retrying the same TSM operation may well result in a
> successful mount of the very same cartridge; and the same tape drive may
> also successfully mount another tape shortly afterwards. I've only come to
> this TSM Server today and I can see that it's been experiencing this as far
> back as TSM's activity log goes.
>
>
>
> The only hit online that I've found so far that seems to fit is a posting
> on
> this list from back in 2000 where someone seemed to be witnessing the same:
> http://www.adsm.org/lists/html/ADSM-L/2000-08/msg00007.html
>
>
>
> Does anyone have any pointers as to where I might look next or any other
> info that might be handy to dig out?
>
>
>
> TIA,
>
>
>
> /David Mc
>
> London, UK
>



--
Wanda Prather  |  Senior Systems Integration Specialist  |  410.539.1135  |
wprather AT icfi DOT com  |  www.jasi.com
ICF Jacob & Sundstrom  | 401 E. Pratt St, Suite 2214, Baltimore, MD 21202