Github user jihoonson commented on a diff in the pull request:
https://github.com/apache/tajo/pull/77#discussion_r16218363
--- Diff: tajo-client/src/main/java/org/apache/tajo/cli/TajoCli.java ---
@@ -628,6 +633,31 @@ public void close() {
}
}
+ // In TajoMaster HA mode, if TajoCli can't connect existing active
master,
+ // this should try to connect new active master.
+ private void checkMasterStatus() {
+ try {
+ if (conf.getBoolVar(TajoConf.ConfVars.TAJO_MASTER_HA_ENABLE)) {
+ if
(!HAServiceUtil.isMasterAlive(conf.get(TajoConf.ConfVars.TAJO_MASTER_CLIENT_RPC_ADDRESS
+ .varname), conf)) {
+ String baseDatabase = client.getBaseDatabase();
+ conf.set(ConfVars.TAJO_MASTER_CLIENT_RPC_ADDRESS.varname,
+ HAServiceUtil.getMasterClientName(conf));
+
+ client.close();
+ commands.clear();
+
+ client = new TajoClient(conf, baseDatabase);
+ context.setCurrentDatabase(client.getCurrentDatabase());
+
+ initHistory();
+ initCommands();
+ }
+ }
+ } catch (Exception e) {
--- End diff --
I think that exceptions should be handled properly. It seems that the
IOException can occur when a new tajo client is created. If an IOException
occurs, the 'client' value will be not updated, and TajoCli might try to
connect to the old master.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
