A few more things:

 Password for SE mode is the serial number.

 You need to stop access to the device (unmount or shutdown of Networker)

 Then at the DD CLI

  priv set SE
  filesystem disable
  se sysparam set OST_ABANDON_TIMEOUT=43200
  filesystem enable

time is in seconds

On 10/17/2012 03:16 PM, Yaron Zabary wrote:

Answering myself:

Go into SE mode:

priv set SE

Checking the parameter:

SE@datadomain## se sysparam show OST_ABANDON_TIMEOUT
Name Description Current Default Override
------------------- --------------------------------- ------- -------
--------
OST_ABANDON_TIMEOUT DDCP abandon context timeout(sec) 10800 10800 none
------------------- --------------------------------- ------- -------
--------

Changing (probably, didn't test it yet):

se sysparam show OST_ABANDON_TIMEOUT=21600

On 10/16/2012 04:23 PM, Yaron Zabary wrote:

Hello all,

Recently, we deployed a new DD670. Ever since then, the incremental
backups of a Novell volume are failing (fulls are OK). Before that,
backups were going to an AFTD on a Solaris X86 storage node. After some
investigation, I suspect that this is the result of an inactivity
timeout withing the DD (or ddboost library). I was able to find some
documentation about OST_ABANDON_TIMEOUT, for example esg125254. This
parameter is to be changed by EMC only. I found some reference to an SE
(system engineer ?) mode and was able to get into it (priv set SE in the
CLI, http://blog.colovirt.com/2008/12/29/data-domain-hung-alert/) .
However, the commands are not documented, so I cannot tell how to change
this. I opened a case with my vendor, but they are still tryning to
figure this out. Can anyone tell me where and how to set this variable ?

TIA.

--

-- Yaron.