Issues with the HA Tables - Fault-tolerant front end configuration sample.
Hello Tibco team, I am running into some issues using the High Availability Tables on Streambase and would like to have your feedback.
I got Kafka and Zookeeper running, tried to keep the configurations as simple as possible.
They both seem to run fine, some warnings but no errors, printscreen in attachment.
I imported the sample and built them with: mvn install -DskipTests=true
And using the streambase command prompt, to install the first node:
epadmin install node --nodename A.publisher --nodedirectory nodes --application "deploy_apps/publish_deploy/target/lv_sample_ha_kafka_publish_deploy-10.6.1-ep-application.zip
And to run it: epadmin --servicename A.publisher start node
Now for the Peer deploy:
epadmin install node --nodename Peer1.ldm_cluster --nodedirectory nodes --application "deploy_apps/peer_deploy/target/lv_sample_ha_peers_deploy-10.6.1-ep-application.zip"
And to run it: epadmin --servicename Peer1.ldm_cluster start node
Last but not the least, the front-end:
epadmin install node --nodename Front-ag_fe.ldm_cluster --nodedirectory nodes --application "deploy_apps/frontend_deploy/target/lv_sample_ha_frontend_deploy-10.6.1-ep-application.zip"
And: epadmin --servicename Front-ag_fe.ldm_cluster start node
However when I connect via Spotfire or LiveView Web the HA tables does not show.
Am I missing some step ? What can possibly be wrong ?
I am using Windows 10 and Java:
java version "11.0.9" 2020-10-20 LTS
Java(TM) SE Runtime Environment 18.9 (build 11.0.9+7-LTS)
Java HotSpot(TM) 64-Bit Server VM 18.9 (build 11.0.9+7-LTS, mixed mode)
Thank you in advance,
Best regards, Dessin Saraiva.
Attachments
| Attachment | Size |
|---|---|
 screenshot_1.png | 423.89 KB |
| screenshot_2.png | 122.18 KB |
| screenshot_3.png | 96.95 KB |
| screenshot_4.png | 82.07 KB |
(1) Answer
Did you create the "Orders" topic on the kafka server as described in the README? By default the peer node will not complete startup if it can't subscribe to the Orders topic.
The sample should just startup and present the HA table to the Services Layer (SL), so obviously something went wrong. If you get snapshots from the SL and DL, that would likely provide enough info to discover what went wrong.
To get a snapshot:
epadmin create snapshot --installpath yourNode.YourCluster
If you can, attach the snapshots to this ticket and I'll take a look.
If you want to explore yourself, you will find the LiveView server log in:
YourNode.YourCluster/logs/default-engine-for*.log
The peer server (the DL) is likely the most interesting. Look for WARNINGs and ERRORs.
You can also learn a good deal by looking at the LVNodeInfo table in the SL. That table has lots of information - for diagnosing this issue it will show what nodes are part of the cluster, what their URL's are, what table groups exist and what nodes are particpating in those groups.
