TC shouldn't raise an exception when deleting qdisc if device doesn't exist

[Rodolfo Alonso]

TC shouldn't raise an exception when deleting qdisc if device doesn't exist.

When Linux Bridge agent deletes a port or detect a port was deleted, QoS extension is informed to clean any traffic shapping in this port by deleting the qdisc. If the port was already deleted, TC will raise an exception:

ProcessExecutionError: Exit code: 1; Stdin: ; Stdout: ; Stderr: Cannot find device "tape0f6e79a-0b"

Logs: http://logs.openstack.org/85/473685/3/check/gate-tempest-dsvm-neutron-scenario-linuxbridge-ubuntu-xenial-nv/427484d/logs/screen-q-agt.txt.gz?level=WARNING#_Jun_13_07_36_14_943304

In this case, TC should catch silently this exception.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/474244

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/474244
Committed: https://git.openstack.org/cgit/openstack/neutron/commit/?id=0a185354a16d2bbd87bea4371933032d0f349e60
Submitter: Jenkins
Branch: master

commit 0a185354a16d2bbd87bea4371933032d0f349e60
Author: Rodolfo Alonso Hernandez <email address hidden>
Date: Wed Jun 14 15:46:28 2017 +0100

    TC doesn't rise exception if device doesn't exist

    If device doesn't exit, TC doesn't need to raise an exception
    during the deletion of the qdisc.

    In iproute2 source code [1], the error code used is "1".

    [1] https://kernel.googlesource.com/pub/scm/linux/kernel/git/shemminger/iproute2/+/master/tc/tc_qdisc.c#188

    Change-Id: I9abecd6ef9d74abff27fb0ed0f5f4535c0622fd6
    Closes-Bug: #1697937

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/474166
Committed: https://git.openstack.org/cgit/openstack/neutron/commit/?id=5aa7192ffa962b3063f7e8137d77dfb6a1e33a82
Submitter: Jenkins
Branch: master

commit 5aa7192ffa962b3063f7e8137d77dfb6a1e33a82
Author: Kevin Benton <email address hidden>
Date: Wed Jun 14 04:29:07 2017 -0700

    Common Agent loop: Catch delete_port extension failures

    Catch delete port extension failures like we do with normal port
    removal to avoid interrupting the deletion processing for other
    ports.

    Related-Bug: #1697937
    Change-Id: I42d86a86965d30e14c85ce975681a6b82be41ecc

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/neutron 11.0.0.0b3 development milestone.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/ocata
Review: https://review.openstack.org/494112

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/494112
Committed: https://git.openstack.org/cgit/openstack/neutron/commit/?id=8691deb707a1cc499688e1962036c10f02cd872a
Submitter: Jenkins
Branch: stable/ocata

commit 8691deb707a1cc499688e1962036c10f02cd872a
Author: Rodolfo Alonso Hernandez <email address hidden>
Date: Wed Jun 14 15:46:28 2017 +0100

    Don't raise exception if TC device doesn't exist

    If device doesn't exit, TC doesn't need to raise an exception
    during the deletion of the qdisc.

    In iproute2 source code [1], the error code used is "1".

    [1] https://kernel.googlesource.com/pub/scm/linux/kernel/git/shemminger/iproute2/+/master/tc/tc_qdisc.c#188

    Change-Id: I9abecd6ef9d74abff27fb0ed0f5f4535c0622fd6
    Closes-Bug: #1697937
    (cherry picked from commit 0a185354a16d2bbd87bea4371933032d0f349e60)

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/neutron 10.0.4 release.