This should resolve a race condition where multiple cron jobs for multiple clusters run at the same time. Whichever job had most recently written to /tmp/config.json seemed to control credentials and which cluster was being run. This was causing 1 cluster's data to come in for 2 providers in cost management.
Summary
This should resolve a race condition where multiple cron jobs for multiple clusters run at the same time. Whichever job had most recently written to
/tmp/config.jsonseemed to control credentials and which cluster was being run. This was causing 1 cluster's data to come in for 2 providers in cost management.