Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 1692123

Summary: [Test-Only]Test metrics store scale and provide sizing estimation for metrics store node
Product: [oVirt] ovirt-engine-metrics Reporter: Shirly Radco <sradco>
Component: GenericAssignee: Shirly Radco <sradco>
Status: CLOSED CURRENTRELEASE QA Contact: David Vaanunu <dvaanunu>
Severity: high Docs Contact:
Priority: high    
Version: 1.2.1.1CC: bugs, dagur, dvaanunu, isaranov, jzmeskal, lleistne, mlehrer
Target Milestone: ovirt-4.3.6Keywords: Performance
Target Release: ---Flags: sradco: ovirt-4.3?
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2019-09-26 19:43:33 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: Metrics RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 1631193    

Description Shirly Radco 2019-03-24 13:16:00 UTC
Description of problem:
We need to provide the users a sizing estimation for the amount of host/vms, hosts/vms networks, hosts/vms disks per metrics store node.
In 3.9 the metrics store all-in-one supported around 50 hosts and 1000 vms, with 1 nic and 1 disk. In 3.11 this was not tested yet.

We required a more detailed sizing guide

Version-Release number of selected component (if applicable):


How reproducible:


Steps to Reproduce:
1. Install ovirt metrics store based on OpenShift 3.11 
2. Configure the environment to send the metrics and logs to the metrics store
3. Test the scale limit of a all-in-one installation

Actual results:


Expected results:


Additional info:

Comment 1 Lucie Leistnerova 2019-03-26 14:14:58 UTC
Daniel, please check with Shirly current status of metrics installation process.

Comment 2 Daniel Gur 2019-03-26 14:35:48 UTC
1. It was not in scale team plan for 4.3 and we don't have the capacity for it, especially as our scale lab is being shut down for migration to a different location.
2. Shirli, Jan is it functionally tested and working with no problems? As I understood it is not part of 4.3 official support.

Comment 3 Jan Zmeskal 2019-03-27 08:33:19 UTC
(In reply to Daniel Gur from comment #2)
> 1. It was not in scale team plan for 4.3 and we don't have the capacity for
> it, especially as our scale lab is being shut down for migration to a
> different location.
> 2. Shirli, Jan is it functionally tested and working with no problems? As I
> understood it is not part of 4.3 official support.

Hi Daniel!

Regarding point #2, I'll actually redirect your question to Ivana Saranova who is main QE for this effort. As far as I know, she has not managed to follow the installation/configuration flow completely. However, I might be mistaken since I've been focusing on OCP on RHV effort lately.

Comment 4 Ivana Saranova 2019-04-04 14:36:30 UTC
Hi,

sorry for late response, I've read this message in middle of midterm exams and lost in the pile of bugs I have ON_QA.
The current status is that the metrics store is correctly installed and I can connect engines with hosts to it, see Kibana dashboards. I still have to test the 4.2 engine with 4.2 hosts compatibility, host-deploy and other functionalities of the metrics store. Trying to focus on 4.3 now, however I'm testing 4.2 as well.

If you have questions, please feel free to ask me or Shirly.

Comment 5 Daniel Gur 2019-05-26 10:12:39 UTC
Retargeting to 4.3.5 as our scale lab where we need to test this - was not brought up yet in the new location.

Comment 9 David Vaanunu 2019-06-27 07:43:05 UTC
Bug 1724483 - master0 server installation takes a long time

Comment 10 Sandro Bonazzola 2019-07-11 10:53:20 UTC
Moved out to 4.3.6 for capacity, would this work for you?

Comment 11 Daniel Gur 2019-07-11 11:51:39 UTC
yes.

Comment 12 David Vaanunu 2019-09-16 08:24:42 UTC
Bug for improving was opened:
Bug 1748964 - Metrics needs 16CPU cores to support monitoring of 1000VMs

Comment 14 Sandro Bonazzola 2019-09-26 19:43:33 UTC
This bugzilla is included in oVirt 4.3.6 release, published on September 26th 2019.

Since the problem described in this bug report should be
resolved in oVirt 4.3.6 release, it has been closed with a resolution of CURRENT RELEASE.

If the solution does not work for you, please open a new bug report.