Re: [ADSM-L] Protect Storage Pool and Replicate Node Process Issuance

Hi David. I have a customer where we have implemented a similar approach.
The environment is significantly smaller than yours. Our daily maintenance
script, which was originally based on the Admincenter is shown below.  I
still prefer to have a scheduled daily maintenance script that performs all
daily housekeeping functions as opposed to a myriad of schedules performing
one function at a time and manually managing timings for the completion of
the individual administrative command schedules.



Name           Line       Command

               Number

----------     ------
------------------------------------------------------------

MAINTENAN-     5          backup db devclass=VV  type=full wait=yes

CE_PLAN_-

CUSTOM

               10         prepare  source=dbbackup devcl=VV wait=YES

               15         protect stg backupcontainer w=y

               20         replicate node * maxsess=10 wait=yes

               25         EXPIRE INVENTORY SKIPDIRS=NO WAIT=YES RESOURCE=40

               30         reclaim stgpool ARCHIVETAPE threshold=50

               35         reclaim stgpool VVTAPE threshold=50

               40         mig stg archivepool lo=0

               45         mig stg vvpool lo=0

               50         del volhist type=all todate=-30



We still use virtual volumes for the DBB even though you can make a good
argument that the DBB is less important in a fully synchronized container
storage pool environment and can be put on a local file system (as in the
8.1 Blueprint). The nice thing about virtual volumes is that DRM manages the
DBB and RPF expiration. This server pair is at 7.1.6 and we have not had any
problems with directory container pools, nor moving containers to different
file systems (analogous to moving file device class volumes).



Best regards,

Joerg Pohlmann



-----Original Message-----
From: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU] On Behalf Of
David Beardsley
Sent: Tuesday, May 09, 2017 05:45
To: ADSM-L AT VM.MARIST DOT EDU
Subject: [ADSM-L] Protect Storage Pool and Replicate Node Process Issuance



Hello All,

  We are currently in the process of implementing node replication in our
environment.  We have an onsite server onsite, and one about 200 miles away
as our replication target.



Our implementation plan has been to start a "protect stg" for our container
pools in the morning when the majority of our backups are done, allow that
to complete, then later in the day run a "replicate node" command.



This seems like it should work to me, and I started by creating scripts to
run the protect storage pool and been running the replicate node command for
a few days by hand for a few reasons.  But since we have our processes
hammered out, the protect storage pool continued to run without issue, but
when the script issued the "replicate node * " command it crashed the
server.



I am curious to see what implementation specs others have or are looking at
in regards to implementing replication.



Now for some details on the instance that experienced the issue, note our
replication server is running the same OS and TSM server version:



Operating systems: Redhat 6.1

TSM server version: 7.1.7.100

Number of nodes: just over 1600

Nodes currently enabled for replication: about 5





Thanks

David Beardsley

Cornell University