Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 1684061

Summary:	ovs-vsctl emits "ovs\|00001\|db_ctl_base\|ERR\|no key "tx_errors" in Interface record" messages in compute
Product:	Red Hat OpenStack	Reporter:	Jijesh Kalliyat <jkalliya>
Component:	openvswitch	Assignee:	Aaron Conole <aconole>
Status:	CLOSED NOTABUG	QA Contact:	Roee Agiman <ragiman>
Severity:	medium	Docs Contact:
Priority:	medium
Version:	10.0 (Newton)	CC:	aconole, aguetta, apevec, chrisw, rhos-maint
Target Milestone:	---
Target Release:	---
Hardware:	Unspecified
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2019-03-26 17:33:06 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:

Description Jijesh Kalliyat 2019-02-28 10:41:24 UTC

Description of problem:

This bugzilla is forked from https://bugzilla.redhat.com/show_bug.cgi?id=1683175 , to address the following messages in OpenStack Compute nodes. As suggested by Daniel Berrange ( bz 1683175, comment #5) , ovs-vsctl might need a fix so it does not emit this log message at all, or at least only log it if "debug" type messages are requested.

Feb 24 22:46:59 <hostname> ovs-vsctl: ovs|00001|db_ctl_base|ERR|no key "tx_errors" in Interface record "vhud1a9329f-78" column statistics

virsh domifstat instance-00000207 vhud1a9329f-78
vhud1a9329f-78 rx_bytes 0
vhud1a9329f-78 rx_drop 599
vhud1a9329f-78 tx_bytes 0
vhud1a9329f-78 tx_errs 0
vhud1a9329f-78 tx_drop 0

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

Red Hat OpenStack Platform 10
Red Hat Enterprise Linux Server 7.4
kernel-3.10.0-693.17.1.el7.x86_64
libvirt-3.9.0-14.el7.x86_64

[root@compute1 ~]# libvirtd -V
libvirtd (libvirt) 3.9.0

[root@compute1 ~]# virsh version
Compiled against library: libvirt 3.9.0
Using library: libvirt 3.9.0
Using API: QEMU 3.9.0
Running hypervisor: QEMU 2.9.0

Comment 1 Aaron Conole 2019-03-08 15:16:53 UTC

Agreed with description, this isn't going to prevent anything in OvS from functioning.  Will present it upstream with a possible fix and see what happens, but I think it's really more like "informational error" rather than functional error.

Comment 4 Aaron Conole 2019-03-26 17:33:06 UTC

After looking at the code, ovs-vsctl already can be told not to emit this message.

From the referenced bug, the command issued is: 
  ovs-vsctl --timeout=5 get Interface vhu6111cccc-e9 statistics:tx_errors

Simply add the --if-exists flag.

  ovs-vsctl --timeout=5 --if-exists get Interface vhu6111cccc-e9 statistics:tx_errors

This will not log an error re: missing keys if the key isn't found.