Bug 1119898

Summary:	[RFE] Indicate which network is utilized when engine generates Network Usage alert
Product:	Red Hat Enterprise Virtualization Manager	Reporter:	Jake Hunsaker <jhunsaker>
Component:	RFEs	Assignee:	Rob Young <royoung>
Status:	CLOSED WONTFIX	QA Contact:	Shai Revivo <srevivo>
Severity:	medium	Docs Contact:
Priority:	medium
Version:	3.4.0	CC:	dholler, jhunsaker, lpeer, mkalinin, srevivo
Target Milestone:	---	Keywords:	FutureFeature
Target Release:	---	Flags:	nyechiel: Triaged+
Hardware:	All
OS:	All
Whiteboard:
Fixed In Version:		Doc Type:	Enhancement
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2020-03-23 14:39:40 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	Network	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Jake Hunsaker 2014-07-15 19:10:00 UTC

Currently, the engine generates the following alert when it detects that a hypervisor's network usage is high:


Used Network resources of host $hypervisor [100%] exceeded defined threshold [95%].


This does not help us in situations where we are investigating a sporadic or intermittent network usage spike and looking back through the audit_log or similar. This RFE is to adjust this message to include which logical network(s) are using the highest bandwidth, e.g.:


Networks $network1, $network2 on hypervisor $hypervisor are using a total of $usage_perc of network resources.


Or anything similar that will identify the networks triggering the alert.

Comment 1 Nir Yechiel 2014-07-30 12:04:23 UTC

Hi Jake,

Starting with 3.4 the individual NIC/NICs should be specified and not just the hypervisor name - see BZ 1070667. This should provide a significant improvement already. Can you take a look on that and see if this can help?

For this customer request I guess that several networks (VLANs) are used in the same NIC, so he prefer to get the information on a per network level? Also, keep in mind that the congestion/network load can be caused due to outside speakers in the network, so it's not always trivial to provide this kind of info on the hypervisor level.

Comment 2 Nir Yechiel 2014-08-27 11:27:52 UTC

Jake, can you please comment on this?

Thanks,
Nir

Comment 3 Jake Hunsaker 2014-08-27 14:13:38 UTC

Nir,

For our purposes right now the individual NIC report should be sufficient after going over the situation with the customer.

However, we have numerous customers that have a decent amount of VLANs on the same physical NIC so I could see in the future where we'd still want to pin the alert to a specific logical network.

Comment 5 Yaniv Lavi 2016-11-23 10:24:47 UTC

We will be resolving this as part of alerting on the metric store.

Comment 8 Sandro Bonazzola 2019-09-26 13:33:52 UTC

Seems to be network related logging on the engine, moved to network

Comment 9 Michal Skrivanek 2020-03-19 15:40:53 UTC

We didn't get to this bug for more than 2 years, and it's not being considered for the upcoming 4.4. It's unlikely that it will ever be addressed so I'm suggesting to close it.
If you feel this needs to be addressed and want to work on it please remove cond nack and target accordingly.

Comment 10 Dominik Holler 2020-03-23 14:39:40 UTC

Using a dedicated network monitoring tool would be more beneficial in this scenario.