c-treeRTG for isCOBOL : Data Replication : Usage and configuration - Legacy method
Usage and configuration - Legacy method
In this chapter we’re going to setup an asynchronous data replication between two c-tree servers without using the Replication Manager utility.
Transaction logging required
The c-tree replication feature relies on a transaction log based facility. Your files must be under transaction control to take advantage of c-tree replication.
You can enable transaction logging in the configuration by setting <transaction> in ctree.conf or iscobol.file.index.transaction.logging (boolean) among isCOBOL properties. This will affect new files.
Transaction logging can be applied to existing files using the -tron option of the ctutil utility.
Shutdown the c-tree servers
It’s good practice to stop the involved c-tree servers before proceeding with the steps discussed below.
Prepare the Source
No particular operation is required on the Source c-tree server. Just be sure that the installed license allows replication features.
Prepare the Target
Be sure this second c-tree server resides in its own directory and has a SERVER_NAME different than the name of the Source c-tree server. For example, you can leave the default name "FAIRCOMS" in the Source c-tree server and use the name "FAIRCOMT" for the Target c-tree server.
1. Edit config/ctsrvr.cfg
a. provide a different SERVER_NAME, e.g. replace
SERVER_NAME FAIRCOMS
with
SERVER_NAME FAIRCOMT
b. activate the ctagent plugin by removing the leading semicolon that comments the line out:
Windows
PLUGIN ctagent;./agent/ctagent.dll
Linux/Unix
PLUGIN ctagent;./agent/libctagent.so
2. replace the content of config/ctagent.json with this code:
{
  "detached": true,
  "configurationFileList": [
     "../config/replication/ctreplagent1.cfg"
]}
3. edit config/replication/ctreplagent1.cfg
a. replace "SourceIP" with the source machine host name in the following line:
source_server FAIRCOMS@SourceIP
b. replace "FAIRCOMS" with "FAIRCOMT" in the following line:
target_server FAIRCOMS@localhost
4. edit config\replication\filter1.xml and
a. replace ".\ctreeSQL.dbs\admin_test.dat" with "*.dat" as follows:
<file status="include">*.dat</file>
b. add a purpose for replicating modifications on existing files as follows:
<purpose>create_file</purpose>
<purpose>open_file</purpose>
<purpose>read_log</purpose>
5. edit config\replication\source_auth.cfg changing the PASSWD value to "ADMIN":
USERID ADMIN
PASSWD ADMIN
6. edit config\replication\target_auth.cfg changing the PASSWD value to "ADMIN":
USERID ADMIN
PASSWD ADMIN
7. open a command prompt
a. change to the config\replication directory
b. run these commands to generate the new source_auth.set and the new target_auth.set:
../../tools/ctcmdset source_auth.cfg
../../tools/ctcmdset target_auth.cfg
8. copy existing c-tree files from the Source server to the Target server
(Optional) Switching from Asynchronous Replication to Synchronous Replication
Synchronous replication is mainly used for high-end transactional applications that need instant failover if the primary node fails and cannot tolerate data loss. As drawback, synchronous replication introduces latency that slows the primary application.
To make the replication synchronous
1. edit config/replication/ctreplagent1.cfg and activate the syncagent option by removing the leading semicolon that comments the line out:
syncagent yes
2. edit config\replication\filter1.xml and add
<purpose>sync_commit</purpose>
Testing the Replication Agent
After the above steps, start (or restart) the c-tree servers.
When both Source and Target c-tree servers are up and running, you can start editing the c-tree files on the Source server. Every modification will be replicated to the corresponding files on the Target server.