Red Hat Bugzilla – Bug 1321102
Warn if there are not enough IPs in the pool (otherwise deployment hangs)
Last modified: 2018-06-27 09:26:26 EDT
Description of problem: When using the static IP allocation feature, no validations are run to see if there are enough IPs in the pool for the requested number of nodes. The deployment simply hangs until it times out. Version-Release number of selected component (if applicable): python-tripleoclient-0.3.1-1.el7ost.noarch openstack-tripleo-heat-templates-0.8.12-1.el7ost.noarch How reproducible: 100 % Steps to Reproduce: 1. In your environment file, create an IP pool for the compute nodes, but make sure you won't have enough for them all. In this example there is only 1 IP: parameter_defaults: NovaComputeIPs: internal_api: - 10.35.191.100 2. Deploy with more nodes than the pool size Actual results: Deployment hangs Expected results: User should get a clear error right on the beginning of the deployment, that more IPs are needed in the pool. Deployment should be aborted and not even attempted.
This bug did not make the OSP 8.0 release. It is being deferred to OSP 10.
While I think this BZ would be good to address, I think the fact that you need to have sufficient IP space in your allocation pools is well documented. I changed the topic of this BZ to indicate that is a feature request. I think this is best addressed with an addition to the tripleo-validations. I have added this BZ to the list of future enhancements to networking validations in tripleo-validations. I am setting the target to OSP 11 for now, although it isn't yet clear if this particular enhancement will be able to land in Ocata.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHEA-2018:2086