bug & workaround: SCM agent fails on very large disks

[#1]

(attempting to follow http://www.cloudera.com/documentation/manager/5-1-x/Cloudera-Manager-Introduction/cm5i_getting_help_... and this seems to be the best place - in JIRA I can't see a sensible project to file the ticket - have searched).

Recently I mounted an Amazon EFS volume on our data nodes, this broke SCM as "total_space_bytes" is 9223372036854775808 which is reported as a long int and breaks the Avro schema. If it was only one byte less...

I've done a bodge job on this locally:
/usr/lib/cmf/agent/build/env# diff ./lib/python2.7/site-packages/cmf-5.8.1-py2.7.egg/cmf/monitor/host/filesystem_map.py{.orig,}
188c188
< total_space_bytes=usage['total'],
---
> total_space_bytes=int(min(usage['total'],2**63-1)),

[#2 · Accepted solution]

This issue has been fixed in an upcoming release.