Bug 1242497

Summary:

RFE: add support to highlight DB/DP inconsistencies

Product:

Red Hat OpenStack

Reporter:

Flavio Leitner <fleitner>

Component:

plotnetcfg

Assignee:

Jiri Benc <jbenc>

Status:

CLOSED ERRATA

QA Contact:

Toni Freger <tfreger>

Severity:

medium

Docs Contact:

Priority:

medium

Version:

6.0 (Juno)

CC:

jbenc, jschluet, kzhang, lhh, nyechiel, rkhan, sclewis, tfreger, yeylon

Target Milestone:

beta

Keywords:

FutureFeature

Target Release:

8.0 (Liberty)

Hardware:

All

OS:

Linux

Whiteboard:

Fixed In Version:

plotnetcfg-0.4.1-2.el7ost

Doc Type:

Enhancement

Doc Text:

Story Points:

---

Clone Of:

Environment:

Last Closed:

2016-04-07 21:02:19 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:

Attachments:

Description	Flags
VLAN and VXLAN	none
AIO with reproduced steps	none

Description Flavio Leitner 2015-07-13 12:40:38 UTC

Description of problem:

OVS tracks its ports in a database in userspace, but those interfaces may not be part of the datapath causing troubles and confusion.

Since plotnetcfg builds a diagram the network on complex environment including OVS, it would be great if it could highlight such cases so that one could spot the problem easily.

Comment 2 Jiri Benc 2015-07-13 13:03:09 UTC

Support for this is already in the plotnetcfg git, will be part of the next version.

Comment 4 Jiri Benc 2015-09-10 12:21:06 UTC

Fixed in plotnetcfg 0.4.1, available in Fedora and EPEL7. Not sure how this works with RHOS.

Comment 5 Flavio Leitner 2015-09-23 13:49:27 UTC

Hi Lon,

Could you explain the steps to update the package on RHOS?
Thanks,
fbl

Comment 9 Toni Freger 2015-10-21 10:53:29 UTC

Hi Jiri, 

We'd like to understand this RFE from testing point of view in order to prioritize it properly.

What kind of action required from QE, in order to test it?

Comment 10 Jiri Benc 2015-10-21 11:13:11 UTC

ip l a type veth
ovs-vsctl add-br ovs0
ovs-vsctl add-port ovs0 veth0
ip l d veth0
ip l a type veth
plotnetcfg | dot -Tpdf

The real "veth0" interface (the one in the red/green/gray box) should not be shown as connected to the "ovs0" interface. Instead, it should be unconnected and there should be a separate "veth0" interface in a dotted box connected to the "ovs0" bridge.

Please note that future ovs versions (>=2.5) will reconnect the interfaces automatically, thus make this harder to reproduce.

Comment 11 Toni Freger 2015-11-01 10:52:09 UTC

Jiri, thanks for the explanation, we would like to test plotencfg, as it seems to be also very helpful as a debugging tool in complex ovs environment.

Comment 12 Jon Schlueter 2015-11-24 04:27:18 UTC

from quick read through it looks like later version of plotnetcfg addresses this RFE updating to Modified but would like input if there is more than a simple package update for OSP 8.

Comment 14 Toni Freger 2016-02-14 13:41:06 UTC

Have tested on latest osp8 AIO from /OpenStack/8.0-RHEL-7/2016-02-11.2
VLANs and VXLAN environments - see attached files.
The scenario you've written in comment #10 passed successfully as well.

Comment 15 Toni Freger 2016-02-14 13:42:01 UTC

Created attachment 1127038 [details]
VLAN and VXLAN

Comment 16 Toni Freger 2016-02-14 13:43:14 UTC

Created attachment 1127039 [details]
AIO with reproduced steps

Comment 19 errata-xmlrpc 2016-04-07 21:02:19 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://rhn.redhat.com/errata/RHEA-2016-0603.html