Edge-UI error when Cassandra is down

Not applicable

We found the below issue while working on backup and restore.

1. In our test environment,we had Cassandra in 2 servers in a ring. If one of Cassandra is down then I could see the below error while opening API proxies in edgeUI. Please see the attachment here ds-accessing-error-in-edgeui.png .

2. As part of backup & recovery guidelines this error should come when morethan 50% of datastore components are down. In our case we have 1 DS is running fine and another one is down( Hence 50% is good), still we are facing the issue.

0 1 165
1 REPLY 1

2 node cassandra clusters are not recommended apigee setup. What is the replication strategy used in configuration file? The replication factor indicates how many nodes have copies of the data. quoram means that majority nodes need to reply to the coordinator. For consistency quarom it is often recommended to use odd number cassandra nodes.

I would suggest you to test this high availability/resilience of Apigee with at least 3 node DS setup as documented in Apigee docs.