Thanks, a lot everyone!

Performance looks good now.  No deadlocks and nothing being purged from the cache.  I have started pointing more sssd clients at the cluster.

Thanks,

Douglas Duckworth, MSc, LFCS
HPC System Administrator
Scientific Computing Unit
Physiology and Biophysics
Weill Cornell Medicine
E: doug@med.cornell.edu
O: 212-746-6305
F: 212-746-8690

On Thu, Aug 31, 2017 at 2:51 AM, Ulrich Windl <Ulrich.Windl@rz.uni-regensburg.de> wrote:
>>> Douglas Duckworth <dod2014@med.cornell.edu> schrieb am 30.08.2017 um 16:06 in
Nachricht
<CAAKHBKm0FKBR4BLNpy6jrszFp35Eid_J2F19MKmfS+G47EKCJA@mail.gmail.com>:
[...]
> Only four clients are currently using this cluster so perhaps I should
> actually use DB_CONFIG before putting it into production.

Maybe you could start considering this from a moderately loaded and sized server:
set_cachesize 0 15000000 1
set_lg_regionmax 262144
set_lg_bsize 2097152
set_flags DB_LOG_AUTOREMOVE
set_lk_max_locks 30000
set_lk_max_objects 30000
---

Also note "olcDbConfig: <DB_CONFIG setting>" is probably the modern way to do it.

Regards,
Ulrich