If ceph-ansible doesn't work (e.g. because of a bad config parameter), launch.sh keeps retrying it and this means you're never really sure if the deploy failed because it just keeps running. I get that you did this to work around first-time transient errors, but 2 times should be enough, right?
batrick
commented
6 years ago
If ceph-ansible doesn't work (e.g. because of a bad config parameter), launch.sh keeps retrying it and this means you're never really sure if the deploy failed because it just keeps running. I get that you did this to work around first-time transient errors, but 2 times should be enough, right?