Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

The FDP team is no longer accepting new bugs in Bugzilla. Please report your issues under FDP project in Jira. Thanks.

Bug 1747531

Summary:	[RFE] Add visibility and config on vhost tx-retries
Product:	Red Hat Enterprise Linux Fast Datapath	Reporter:	Kevin Traynor <ktraynor>
Component:	openvswitch2.11	Assignee:	Kevin Traynor <ktraynor>
Status:	CLOSED ERRATA	QA Contact:	qding
Severity:	low	Docs Contact:
Priority:	unspecified
Version:	FDP 19.G	CC:	cfontain, ctrautma, fbaudin, fhallal, fleitner, gmuthukr, jhsiao, ralongi, vchundur
Target Milestone:	---
Target Release:	---
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:	openvswitch2.11-2.11.0-23.el7fdn.x86_64.rpm	Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:
Clones:	1759336 (view as bug list)		Environment:
Last Closed:	2019-11-06 05:21:19 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:
Bug Depends On:
Bug Blocks:	1759336

Description Kevin Traynor 2019-08-30 17:43:43 UTC

vhost tx retries can provide some mitigation against tx dropped packets due to a temporarily slow guest/limited tx queue size for a vhost interface, by quickly retrying to send the packets that had failed to send.

However, there is currently no way to know if these retries are occurring as they are not reported.

That information is useful as it indicates that the guest is not optimally configured/needs more resources to consume from the tx queue faster, or the tx queue size needs to be increased.

Another issue could be if a system was fully stressed to the max, the extra cycles taken for tx retries could in theory mean that rx processing of an interface by the same PMD did not get enough cycles and there were rx drops.

Additionally, vhost tx retries cannot be disabled, so there is no way to debug what the behaviour is without those retries which may or may not be happening.

This RFE is to add visibility of how many tx retries are occurring and add a config so they can be disabled.

Comment 2 Kevin Traynor 2019-09-27 13:23:54 UTC

These commits are in upstream OVS 2.12. For this Bz they are also backported to OVS 2.11 pkg, see Fixed in Version.

Comment 4 qding 2019-10-15 05:10:06 UTC

Verified with openvswitch2.11-2.11.0-24.el7fdp.x86_64

[root@dell-per730-04 ~]# 

[root@dell-per730-04 ~]# ovs-vsctl get Interface vhost0 statistics:tx_retries
3552306
[root@dell-per730-04 ~]# ovs-vsctl get Interface vhost0 statistics:tx_retries
4603019
[root@dell-per730-04 ~]# ovs-vsctl get Interface vhost0 statistics:tx_retries
4603019
[root@dell-per730-04 ~]# ovs-vsctl get Interface vhost0 statistics:tx_retries
5976584
[root@dell-per730-04 ~]# ovs-vsctl get Interface vhost0 statistics:tx_retries
5976584
[root@dell-per730-04 ~]# ovs-vsctl get Interface vhost0 statistics:tx_retries
7350019
[root@dell-per730-04 ~]# ovs-vsctl set Interface vhost0 options:tx-retries-max=0
[root@dell-per730-04 ~]# ovs-vsctl set Interface vhost0 options:tx-retries-max=0
[root@dell-per730-04 ~]# ovs-vsctl get Interface vhost0 statistics:tx_retries
11470313
[root@dell-per730-04 ~]# ovs-vsctl get Interface vhost0 statistics:tx_retries
11525365
[root@dell-per730-04 ~]# ovs-vsctl get Interface vhost0 statistics:tx_retries
11525365
[root@dell-per730-04 ~]# ovs-vsctl get Interface vhost0 statistics:tx_retries
11525365
[root@dell-per730-04 ~]# ovs-vsctl get Interface vhost0 statistics:tx_retries
11525365
[root@dell-per730-04 ~]# ovs-vsctl get Interface vhost0 statistics:tx_retries
11525365
[root@dell-per730-04 ~]# ovs-vsctl get Interface vhost0 statistics:tx_retries
11525365
[root@dell-per730-04 ~]# ovs-vsctl get Interface vhost0 statistics:tx_retries
11525365
[root@dell-per730-04 ~]#

Comment 6 errata-xmlrpc 2019-11-06 05:21:19 UTC

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

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

https://access.redhat.com/errata/RHBA-2019:3719