CCM timeouts

[adespawn]

According to CI logs, the ccm fails with timeout on the ccm start --wait-for-binary-proto command.

Problematic list of commands:

ccm create arbitraryname -v 3.11.4
ccm populate -n 1:0
ccm start --wait-for-binary-proto

Steps to reproduce:

1. Modify the integration test task, by removing build and run integration tests step.
2. Add a script with offending commands as a CI step (see this file)
3. Alternatively you can add ssh step to the CI instead of the script, and run the offending commands manually after connecting with ssh.
4. Run the CI

Problem:

With some probability (around 20% observed before summer, around 80% now), the last command (ccm start --wait-for-binary-proto will run indefinitely. This happens only on the first launch of this command on a CI VM - meaning that once this command finishes successfully, you need to obtain a new VM to observe this issue.

Taken debugging steps

The ccm does not provide any logging.
When run with strace, I was able to determine, that a Cassandra java process is started, but I was unable to find any other useful information in that log

strace-logs.zip

Proposed Implemented workaround

With enabling of the integration tests on ScyllaDB we can for now skip testing on a Cassandra.