Description of problem: ------------------------ The metric_rollups table shows that disk I/O for RHEV hosts is 0 even though there is disk activity on the RHEV host. vmdb_production=# select disk_usage_rate_average from metric_rollups where resource_type = 'Host' and capture_interval_name = 'hourly'\x\g\x Expanded display is on. -[ RECORD 1 ]-----------+-- disk_usage_rate_average | 0 -[ RECORD 2 ]-----------+-- disk_usage_rate_average | 0 Also, see https://bugzilla.redhat.com/show_bug.cgi?id=1404391 that addressed the collection of disk I/O metric for RHEV VMs. Version-Release number of selected component (if applicable): ------------------------------------------------------------- 5.7.1.1 How reproducible: ---------------- Always Steps to Reproduce: ------------------- 1.Manage RHEV provider 2.Enable C&U collection for the provider Actual results: --------------- The metric_rollups table shows that disk I/O for RHEV hosts is 0. Expected results: ----------------- The metric_rollups table should show accurate disk I/O activity for RHEV hosts. Additional info: ----------------
My appliance is managing a RHEV 4.0 provider.
Boris, How can we tell if the fix for this issue is fixed it the setup Nandini is using?
Hi Oved, Yes, this was tested on 5.7.1.1 and the issue occurs in 5.7.1.1.
Based on Shirly's investigation it seems host disk I/O data is not collected on RHV side, therefore turning this bug into a RFE. In order to have it implemented, changes are required both on RHV side (collect those metrics), and on CFME side. The bug for RHV side exists, and will be added here later on.