Hello.
We're planning to upgrade our Terracotta 10.3 clustering (iBM Memory Max) to the latest released version ( fix #37 ).
I could read the following instructions in the readme file:
8.1 Shut down the server array. A safe shutdown procedure is as follows.
a. Shut down the mirror servers using the stop-tc-server script.
If you are using a wrapper solution to manage the mirror servers, execute the wrapper shut
down command to shut down the mirror servers instead of using the stop-tc-server script.
b. Shut down the clients. A terracotta client will shut down when you shut down your application.
c. Shut down the active servers using the stop-tc-server script.
If you are using a wrapper solution to manage the servers, execute the wrapper shut down
command to shut down the servers instead of using the stop-tc-server script.
8.2 This fix overwrites server scripts,wrapper configuration files and default tc-config.xml. If
you have any custom settings defined on those files (ex:MaxDirectMemorySize), then you need to
restore those settings back after applying the fix.
With the above, please, could you help me with the following questions?
- So, before upgrading it, we'll need to stop all of our I.S and then stop all Terracotta running instances, right?
- We're going to need to backup our server scripts, wrapper configuration files and the default tc-config.xml file, right?
We start-stop our Terracotta server during some I.S server maintenances/o.s upgrades with the following script long ago deployed in our environment:
IS_HOME=/WebM/IS01
NODE=IS01_1
JAVA_HOME=${IS_HOME}/jvm/jvm/
export JAVA_HOME
${IS_HOME}/Terracotta/server/bin/start-tc-server.sh -f ${IS_HOME}/Terracotta/server/is01_tc-config.xml -n ${NODE} > ${IS_HOME}/Terracotta/server/is01_nohup &
With the above, should we only restore the backed up start-tc-server.sh + is01_tc-config.xml files ?
Or are going to need to setup the entire Terracotta environment again?
Thank you in advance!