[OVSDB-264] No Flow Entries are installed to br-int while trying the NetVirt Cluster (clusteraware) Created: 20/Jan/16  Updated: 09/Feb/16  Resolved: 09/Feb/16

Status: Resolved
Project: ovsdb
Component/s: openstack.net-virt
Affects Version/s: unspecified
Fix Version/s: None

Type: Bug
Reporter: Venkatrangan Govindarajan Assignee: Anil Vishnoi
Resolution: Done Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified
Environment:

Operating System: All
Platform: All

Attachments: File odl1_log.tgz     File odl2_log.tgz     File odl3_log.tgz    
External issue ID: 5018
Priority: Low

 Description   

Created a NetVirt cluster (ODL1,ODL2,ODL3) Configured the Openstack Control and Compute nodes to send neutron requests to ODL1, The ODL2 and ODL3 set
as OVS Managers

local.conf (control node): https://gist.github.com/gvranganvtn/4c498ff24d3aa2a95693

local.conf (compute node):
https://gist.github.com/gvranganvtn/9be5ae6c385aeac5abef

ubuntu@odl-os-control-node:~/devstack$ sudo ovs-vsctl show
7379aa64-0416-45e2-9531-351397cc742a
Manager "tcp:11.12.13.5:6640"
Manager "tcp:11.12.13.4:6640"
is_connected: true
Bridge br-ex
Controller "tcp:11.12.13.4:6653"
is_connected: true
Controller "tcp:11.12.13.5:6653"
fail_mode: secure
Port br-ex
Interface br-ex
type: internal
Port "eth2"
Interface "eth2"
Bridge br-int
Controller "tcp:11.12.13.4:6653"
is_connected: true
Controller "tcp:11.12.13.5:6653"
fail_mode: secure
Port "tap76ec97f0-cf"
Interface "tap76ec97f0-cf"
type: internal
Port "tap6205c62f-88"
Interface "tap6205c62f-88"
Port br-int
Interface br-int
type: internal
ovs_version: "2.3.2"

Compute node:

ubuntu@odl-os-compute-node:~$ sudo ovs-vsctl show
bf460c25-5849-4874-9758-4fcbafbf9f5e
Manager "tcp:11.12.13.4:6640"
is_connected: true
Manager "tcp:11.12.13.5:6640"
is_connected: true
Bridge br-ex
Controller "tcp:11.12.13.4:6653"
is_connected: true
Controller "tcp:11.12.13.5:6653"
is_connected: true
Port "eth2"
Interface "eth2"
Port br-ex
Interface br-ex
type: internal
Bridge br-int
Controller "tcp:11.12.13.4:6653"
is_connected: true
Controller "tcp:11.12.13.5:6653"
is_connected: true
Port "tapc3170fef-e6"
Interface "tapc3170fef-e6"
Port br-int
Interface br-int
type: internal
ovs_version: "2.3.2"

Created a Network and two instances.
No flows are installed.

Erros in Karaf log in ODL2 and ODL3

ODL2 Karaf Exception: https://gist.github.com/gvranganvtn/d45d7f538ae66f6b8615

ODL3 Karaf exception:
https://gist.github.com/gvranganvtn/9b439346950307546b5b

ODL1 did not display any exception .

 Comments   
Comment by Venkatrangan Govindarajan [ 20/Jan/16 ]

Used stable/Beryllium branch or testing

Comment by Venkatrangan Govindarajan [ 20/Jan/16 ]

Tar and gzipped, please tar -zxvf

Comment by Venkatrangan Govindarajan [ 20/Jan/16 ]

Attachment odl1_log.tgz has been added with description: ODL1 Logs

Comment by Venkatrangan Govindarajan [ 20/Jan/16 ]

Attachment odl2_log.tgz has been added with description: ODL 2 Logs

Comment by Venkatrangan Govindarajan [ 20/Jan/16 ]

Attachment odl3_log.tgz has been added with description: ODL 3 Logs

Comment by Venkatrangan Govindarajan [ 20/Jan/16 ]

Retested today, similar observation ,But some flows were installed

Compute Node
ubuntu@odl-os-compute-node:~/devstack$ sudo ovs-ofctl dump-flows br-int -OOpenFlow13
OFPST_FLOW reply (OF1.3) (xid=0x2):
cookie=0x0, duration=12.057s, table=90, n_packets=0, n_bytes=0, priority=61007,ip,dl_dst=fa:16:3e:bc:26:35 actions=goto_table:100

Control Node

ubuntu@odl-os-control-node:~/devstack$ sudo ovs-ofctl dump-flows br-int -OOpenFlow13
OFPST_FLOW reply (OF1.3) (xid=0x2):
cookie=0x0, duration=103.301s, table=0, n_packets=5, n_bytes=390, in_port=1,dl_src=fa:16:3e:b3:00:ad actions=set_field:0x40b->tun_id,load:0x1->NXM_NX_REG0[],goto_table:20
cookie=0x0, duration=23.423s, table=40, n_packets=0, n_bytes=0, priority=61007,ip,dl_src=fa:16:3e:68:90:fb actions=goto_table:50

Comment by Venkatrangan Govindarajan [ 20/Jan/16 ]

Also, throughout the time...some logs wer in the karaf console...
opendaylight-user@root>ConfigProperties missing a value for ovsdb.userspace.enabled
ConfigProperties missing a value for ovsdb.controller.address
ConfigProperties missing a value for of.address
ConfigProperties missing a value for of.listenPort
ConfigProperties missing a value for of.listenPort
ConfigProperties missing a value for ovsdb.userspace.enabled
ConfigProperties missing a value for ovsdb.controller.address
ConfigProperties missing a value for of.address
ConfigProperties missing a value for of.listenPort
ConfigProperties missing a value for of.listenPort

Comment by Sam Hague [ 20/Jan/16 ]

(In reply to Venkatrangan Govindarajan from comment #6)
> Also, throughout the time...some logs wer in the karaf console...
> opendaylight-user@root>ConfigProperties missing a value for
> ovsdb.userspace.enabled
> ConfigProperties missing a value for ovsdb.controller.address
> ConfigProperties missing a value for of.address
> ConfigProperties missing a value for of.listenPort
> ConfigProperties missing a value for of.listenPort
> ConfigProperties missing a value for ovsdb.userspace.enabled
> ConfigProperties missing a value for ovsdb.controller.address
> ConfigProperties missing a value for of.address
> ConfigProperties missing a value for of.listenPort
> ConfigProperties missing a value for of.listenPort

Those logs are OK - it just means the values were not used. They are currently logged as errors using System.err.println. We shoudl convert those to logger and fix the check to ignore if a default null value is used. The log comes out because a null default is used, the config is not found and null is returned but the method expects a value, sees null and logs the error. The check should ignore if a default of null was used.

Comment by Sam Hague [ 02/Feb/16 ]

Venkat, is this bug fixed now? I think these failures were from the rc0 build that had issues which have been fixed.

Comment by Venkatrangan Govindarajan [ 03/Feb/16 ]

Reduce priority as this test uused odl-ovsdb-openstack-clusteraware feature, which is not targetted for Be release

Comment by Anil Vishnoi [ 09/Feb/16 ]

This issue should be fixed now. This was happening because of the routed RPC timing issue discussed in below bug.

https://bugs.opendaylight.org/show_bug.cgi?id=4866

and fixed through

https://git.opendaylight.org/gerrit/34207
https://git.opendaylight.org/gerrit/34115

Generated at Wed Feb 07 20:35:55 UTC 2024 using Jira 8.20.10#820010-sha1:ace47f9899e9ee25d7157d59aa17ab06aee30d3d.