Abhishek Verma

Incorrect order in which CassandraDaemon nodes are started when cluster size is > 10

1

I started a 20-node Cassandra cluster and the nodes start in the order: node-0, node-1, node-10, node-11,.., node19, node-2, node-3, ..., node-9 This is because we internally sort them by...

This can happen for example, if the Mesos master sends an authentication timeout if it is overloaded.

…-Dcassandra.replace_address if a task fails and restarts. Summary: Reverts https://github.com/mesosphere/dcos-cassandra-service/pull/362. After testing with some deployments, we have found out that this causes problems when multiple Cassandra tasks fail at the...

Bug: "Unable to find a volume with persistence id" while creating a new cluster

2

I am getting this error almost every time when I try to start a new cluster: ``` INFO [2016-05-04 03:55:56,117] org.apache.mesos.offer.OfferEvaluator: Found Offer meeting placement constraints: id { value: "448075ee-ee69-4fba-9d9d-9157746b3841-O644"...