Description of problem: Metric details are missing for any hosts of Openstack Platform Director system. Version-Release number of selected component (if applicable): 5.6.0 -> CFME Installed Version of RHOS is 9. How reproducible: Always at Customer end. Steps to Reproduce: 1. Select any host from OSP-Director Provider 2. Click on Monitoring button and selecting "Utilization" returns an missing details of Metrics. 3. Exception observed in evm.log is related to read time out when there is break in connectivity between cf and OSP. [----] E, [2017-02-10T10:47:49.644386 #20116:b3b988] ERROR -- : MIQ(ManageIQ::Providers::Openstack::InfraManager::MetricsCapture#perf_collect_metrics) [realtime] for: [ManageIQ::Providers::Openstack::InfraManager::Host], [1], [355b8af5-818f-40e3-b0a6-56d7e8ea3f41 (NovaCompute)] Timings at time of error: {:capture_state=>0.03525042533874512, :connect=>62.28718280792236} Actual results: Metric details are missing with read time out exception. Expected results: Metric details should be visible. Additional info: During troubleshooting process, we checked the following stuff: 1. Checking the ceilometer service is up or not? Status: checked the #openstack service-status and #ceilometer meter-list providing all the metering details. 2. Checked the Network connectivity between cfme and OSP Status: Deployed an appliance on the provider and have it act as proxy. For proxy or worker appliance, one worker appliance created in the same network and connected it to the master database. On this new appliance, Enabled only the c&u roles. 3. Checking the connectivity to metering service from the cfproxy.local appliance using curl command. Status: Checked, ceilometer api is providing the metering details.
Greg: It would be really great if you could provide an inputs on this as it does not seems to be a networking related issue and all the services are up and running.
*** Bug 1421102 has been marked as a duplicate of this bug. ***
Hi Neha! I believe this is a duplicate of https://bugzilla.redhat.com/show_bug.cgi?id=1421729. To summarize: Ceilometer is being phased out for metrics collection, replaced by Gnocchi. What CF does is check for the Gnocchi service in Keystone; if it exists, it uses that; if not, it falls back to Ceilometer. Unfortunately OSP9 goes 'in between' in that it installs Gnocchi and registers it in Keystone for use with Aodh alarms, but uses Ceilometer for metrics. A simple quick fix would be to simply remove Gnocchi from the Keystone catalog; this is fine as long as you're not using Aodh alarms. It's also a good way to verify this is the problem you're running into. As a longer-term solution, we can also backport https://github.com/ManageIQ/manageiq/pull/13918; this PR adds a setting that allows you to specify a metering service for CF to use.
Please assess the importance of this issue and update the priority accordingly. Somewhere it was missed in the bug triage process. Please refer to https://bugzilla.redhat.com/page.cgi?id=fields.html#priority for a reminder on each priority's definition. If it's something like a tracker bug where it doesn't matter, please set it to Low/Low.
Neha, does my comment in https://bugzilla.redhat.com/show_bug.cgi?id=1421103#c6 make sense? You can try using the setting specified in the PR linked there to work around this problem.
Hello Chen, Thanks for the inputs and Pull Requests. Will try implementing the same and see if it helps in providing the workaround the same. Regards, Neha Chugh
Bug Closure Dear customer, The CloudForms team is reviewing the current CloudForms Bug(defect) backlog in order to target engineering efforts. We are closing any bugs for versions that no longer have an active errata stream or that have hit their age limit. We are committing to better management of the backlog as we move forward. If you have an bug that you are still able to reproduce on a current version of CloudForms please open a new bug. If you have any concerns about this, please let us know. Thanks and regards!