The CNS TSM infrastructure comprises several server instances. These are started manually. Herein are recorded instructions for the startup procedures.
Each of the TSM servers has a CNAME which points at the A record for its hosting box. You can simply ssh to that cname. So for the EXT server, you can ssh to tsm-ext.cns.ufl.edu.
When started normally, each TSM instance should be present in the process table with a notation indicating which server it is.
tsm2en1 / 260 ps -fade | grep dsmserv
root 389324 1 0 Oct 01 - 475:41 dsmserv quiet -- CTRL
root 413934 1 47 Oct 01 - 1741:34 dsmserv quiet -- ERP
root 417824 1 0 Oct 01 - 98:49 dsmserv quiet -- WEBCT
root 426038 1 5 Oct 01 - 1860:53 dsmserv quiet -- EXT
root 450596 1 6 Oct 01 - 502:12 dsmserv quiet -- GLMAIL02
root 454884 1 5 Oct 01 - 825:38 dsmserv quiet -- INT
root 458988 1 0 Oct 01 - 17:03 dsmserv quiet -- VI
root 466958 1 0 Oct 01 - 237:45 dsmserv quiet -- COPIES
root 471048 1 6 Oct 01 - 487:08 dsmserv quiet -- GLMAIL01
root 516226 1 0 Oct 02 - 0:55 dsmserv quiet -- TEST
If you find that the server you want to start is present in the process table, then STOP.
The servers are started with a script, /var/tsm/base/bin/start-servers It looks up the servers in the global configuration, and starts the specified ones in order. Usually, you'll be starting just one
tsm2en1 / 262 /var/tsm/base/bin/start-servers -server test
Starting test...
( cd /var/tsm/test; . /var/tsm/servers/test/dsmserv.env; dsmserv quiet -- TEST < /dev/null > /dev/null 2>&1 & );
Pause for 2...
sleep 2
If you omit the server specification, then the script will attempt to start all of them. It should not be a problem to attempt a second start of an already-running server; the second attempt will fail with some diagnostics. But to attempt this will induce cardiac palpitations in your TSM admin; please avoid it.