8c44c7a960
The Hyper-V metric collection API has the following limitations: * some metrics get disabled when the instance is rebooted (network, cpu and memory) * network metrics cannot be enabled while the vm is stopped * metrics cannot be collected for passthrough disks * some times, the metric counters are reset when the vms are rebooted. We can't do much about this Hyper-V bug, which is unfortunately affecting the reliability of the collected data. At the same time, OVS ports and Cinder volumes aren't handled at all. To overcome those issues, this patch brings the following changes: * re-enable vm metrics each time the vms boot * enable vm metrics after live migrations * enable metrics metrics when binding OVS ports (networking-hyperv will handle 'hyperv' ports) * enable disk metrics when attaching SMB Cinder volumes, logging a warning for passthrough disks Closes-Bug: #1794456 Depends-On: Ibe7409b392b91204f0cea2bdd32c86ec03d06b67 Depends-On: Iaba32ca94b2c09661f150fa4b12b2c7eb83ff6d9 Change-Id: I94f2be20b02531d3ceabdef4272207e37ef34930 |
||
|---|---|---|
| .. | ||
| hacking | ||
| nova | ||
| tests | ||
| __init__.py | ||
| i18n.py | ||