Thomas Denier wrote:

>As far as I can tell, there were no node sessions in progress when the failure
>occurred, and no administrative processes other than process 227 itself. I
>find it difficult to imagine why ADSM would have any trouble with database
>contention under those circumstances. We are running a 3.1.2.20 server under
>MVS. Does anyone recognize this as a know problem?

Yes. We had the same 'deadlock' situations sometimes during 'heavy'
reclaim's which were initiated by a running 'expiration' (which took
at least 24 hrs. and was even more heavy).
IBM told me that's nothing to worry about.
Some processes were stopped just to prevent database-corruption.
Sometime reclaim's, which automatically would be restarted, sometime
the 'expiration' itself, which not automatically would be restarted
(because that was administrative scheduled).
Since we running 3.1.2.40, the 'problem' seems to be history.
(We're running AIX btw, but that doesn't make any difference, I presume)

Cheers,
Henk ten Have