server: export granular volume bytes and iops metrics#3259
Conversation
| } | ||
|
|
||
| protected Answer execute(GetVmDiskStatsCommand cmd) { | ||
| // TODO: implement for Vmware |
There was a problem hiding this comment.
Use/extend perfcounter, and see https://www.vmware.com/support/developer/converter-sdk/conv61_apireference/virtual_disk_counters.html for disk read/write and iops
|
@blueorangutan package |
|
@anuragaw a Jenkins job has been kicked to build packages. I'll keep you posted as I make progress. |
|
Packaging result: ✔centos6 ✔centos7 ✔debian. JID-2690 |
| final VirtualMachineMO vmMo = dcMo.findVm(vmName); | ||
| final List<VmDiskStatsEntry> diskStats = new ArrayList<>(); | ||
| for (VirtualDisk disk : vmMo.getAllDiskDevice()) { | ||
| // TODO Get UUID, PATH of VirtualDisk |
There was a problem hiding this comment.
WIP - need to figure this out before testing end to end. CC/ @rhtyd
…ics API This re-uses an existing volume stats dao/table to display the volume specific disk read/write bytes and iops statistics via the listVolumes and listVolumeMetrics API, as well as display granular metrics in the UI. TODO: implement support for VMware and possibly for XenServer? Signed-off-by: Rohit Yadav <rohit.yadav@shapeblue.com>
c3d7541 to
0b72f55
Compare
|
@blueorangutan package |
|
@rhtyd a Jenkins job has been kicked to build packages. I'll keep you posted as I make progress. |
|
Packaging result: ✔centos6 ✔centos7 ✔debian. JID-2699 |
Signed-off-by: Rohit Yadav <rohit.yadav@shapeblue.com>
Signed-off-by: Rohit Yadav <rohit.yadav@shapeblue.com>
Signed-off-by: Rohit Yadav <rohit.yadav@shapeblue.com>
Signed-off-by: Rohit Yadav <rohit.yadav@shapeblue.com>
2bbfcbe to
df25146
Compare
|
@blueorangutan package |
|
@borisstoyanov a Jenkins job has been kicked to build packages. I'll keep you posted as I make progress. |
|
Packaging result: ✔centos6 ✔centos7 ✔debian. JID-2705 |
|
@blueorangutan test |
|
@borisstoyanov a Trillian-Jenkins test job (centos7 mgmt + kvm-centos7) has been kicked to run smoke tests |
|
Trillian test result (tid-3502)
|
|
@blueorangutan test centos7 vmware-65u2 |
|
@rhtyd a Trillian-Jenkins test job (centos7 mgmt + vmware-65u2) has been kicked to run smoke tests |
borisstoyanov
left a comment
There was a problem hiding this comment.
LGTM test results attached
test_results.xlsx
|
@blueorangutan test centos7 vmware65u2 |
|
@rhtyd unsupported parameters provided. Supported mgmt server os are: |
|
@blueorangutan test centos7 vmware-65u2 |
|
@rhtyd a Trillian-Jenkins test job (centos7 mgmt + vmware-65u2) has been kicked to run smoke tests |
anuragaw
left a comment
There was a problem hiding this comment.
LGTM from code and local testing
|
Trillian test result (tid-3703)
|
Signed-off-by: Rohit Yadav <rohit.yadav@shapeblue.com>
|
Fixed conflicts |
|
@rhtyd a Jenkins job has been kicked to build packages. I'll keep you posted as I make progress. |
|
Packaging result: ✖centos6 ✔centos7 ✔debian. JID-2908 |
|
@blueorangutan package |
|
@rhtyd a Jenkins job has been kicked to build packages. I'll keep you posted as I make progress. |
|
Packaging result: ✔centos6 ✔centos7 ✔debian. JID-2913 |
|
@blueorangutan package |
|
@rhtyd a Jenkins job has been kicked to build packages. I'll keep you posted as I make progress. |
|
@blueorangutan package |
|
@rhtyd a Jenkins job has been kicked to build packages. I'll keep you posted as I make progress. |
|
Packaging result: ✔centos6 ✔centos7 ✔debian. JID-17 |
|
@blueorangutan test centos7 vmware-65u2 |
|
@rhtyd a Trillian-Jenkins test job (centos7 mgmt + vmware-65u2) has been kicked to run smoke tests |
|
Trillian test result (tid-15)
|
|
@blueorangutan test centos7 vmware-65u2 |
|
@rhtyd a Trillian-Jenkins test job (centos7 mgmt + vmware-65u2) has been kicked to run smoke tests |
|
Trillian test result (tid-27)
|
Problem: The VM metrics has aggregated volume bytes read/write and iops metrics but not on per volume basis.
Root Cause: The volume stats sub-system is not used to export the metrics, the support is not available for VMware.
Solution: Use the volume stats sub-system and DB table to export the metrics via the listVolumes and listVolumeMetrics API, and implement support for VMware and fix issue with network and disk metrics in the VM metrics view.
Types of changes
listVolume response will now have:
listVolumeMetrics response will have: