Fullstack test_securitygroup.TestSecurityGroupsSameNetwork fails often after SG rule delete

[Slawek Kaplonski]

Fullstack test_securitygroup.TestSecurityGroupsSameNetwork is failing quite often like for example at: http://logs.openstack.org/12/470912/38/check/neutron-fullstack/ba16b38/logs/testr_results.html.gz

It happens quite lot recently (48 times in last 7 days): http://logstash.openstack.org/#dashboard/file/logstash.json?query=build_name%3A%5C%22neutron-fullstack%5C%22%20AND%20message%3A%5C%22line%20264%2C%20in%20test_securitygroup%5C%22

[Slawek Kaplonski (slaweq)]

It's possible that this issue will be fixed by https://review.openstack.org/#/c/572295/

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/572295
Committed: https://git.openstack.org/cgit/openstack/neutron/commit/?id=11b41d73b48d46b7725bb97d4b85158d8bc2acfb
Submitter: Zuul
Branch: master

commit 11b41d73b48d46b7725bb97d4b85158d8bc2acfb
Author: Slawek Kaplonski <email address hidden>
Date: Tue Jun 5 10:05:58 2018 +0200

    [Fullstack] Wait for SG to be applied by L2 agent

    In fullstack test_securitygroup there were used simple
    net_helpers.assert_ping() and assert_not_ping() functions
    which tries to ping IP address 3 times with some short timeout
    and test fails if result of ping will not be as expected.

    Unfortunately sometimes in fullstack tests it might be not enough
    if test is creating new SG or SG rule or apply SG to port and
    just after that checks connection because L2 agent don't have
    enough time to apply all rules on "host".
    This patch changes it to use block_until_ping() and
    block_until_no_ping() methods from FakeMachine fixture.
    It will also check if ping is possible/not possible but
    will try to check it for 1 minute before fails.

    Similar change is also done for methods which checks TCP
    connectivity using netcat helper class. It now uses
    common_utils.wait_until_true() helper function instead of
    fail immediatelly.

    Change-Id: I9e523d803e3c49d5d090ae5b9d36d43ce7311535
    Closes-Bug: #1774006
    Closes-Bug: #1767829

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/queens
Review: https://review.openstack.org/573611

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/pike
Review: https://review.openstack.org/573612

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/573612
Committed: https://git.openstack.org/cgit/openstack/neutron/commit/?id=28e59962523b37aef7d9faf0bf626946fef2c944
Submitter: Zuul
Branch: stable/pike

commit 28e59962523b37aef7d9faf0bf626946fef2c944
Author: Slawek Kaplonski <email address hidden>
Date: Tue Jun 5 10:05:58 2018 +0200

    [Fullstack] Wait for SG to be applied by L2 agent

    In fullstack test_securitygroup there were used simple
    net_helpers.assert_ping() and assert_not_ping() functions
    which tries to ping IP address 3 times with some short timeout
    and test fails if result of ping will not be as expected.

    Unfortunately sometimes in fullstack tests it might be not enough
    if test is creating new SG or SG rule or apply SG to port and
    just after that checks connection because L2 agent don't have
    enough time to apply all rules on "host".
    This patch changes it to use block_until_ping() and
    block_until_no_ping() methods from FakeMachine fixture.
    It will also check if ping is possible/not possible but
    will try to check it for 1 minute before fails.

    Similar change is also done for methods which checks TCP
    connectivity using netcat helper class. It now uses
    common_utils.wait_until_true() helper function instead of
    fail immediatelly.

    Change-Id: I9e523d803e3c49d5d090ae5b9d36d43ce7311535
    Closes-Bug: #1774006
    Closes-Bug: #1767829
    (cherry picked from commit 11b41d73b48d46b7725bb97d4b85158d8bc2acfb)

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/573611
Committed: https://git.openstack.org/cgit/openstack/neutron/commit/?id=2ae4d6cd0044a62d0552c6e07592bf63fdbf7bba
Submitter: Zuul
Branch: stable/queens

commit 2ae4d6cd0044a62d0552c6e07592bf63fdbf7bba
Author: Slawek Kaplonski <email address hidden>
Date: Tue Jun 5 10:05:58 2018 +0200

    [Fullstack] Wait for SG to be applied by L2 agent

    In fullstack test_securitygroup there were used simple
    net_helpers.assert_ping() and assert_not_ping() functions
    which tries to ping IP address 3 times with some short timeout
    and test fails if result of ping will not be as expected.

    Unfortunately sometimes in fullstack tests it might be not enough
    if test is creating new SG or SG rule or apply SG to port and
    just after that checks connection because L2 agent don't have
    enough time to apply all rules on "host".
    This patch changes it to use block_until_ping() and
    block_until_no_ping() methods from FakeMachine fixture.
    It will also check if ping is possible/not possible but
    will try to check it for 1 minute before fails.

    Similar change is also done for methods which checks TCP
    connectivity using netcat helper class. It now uses
    common_utils.wait_until_true() helper function instead of
    fail immediatelly.

    Change-Id: I9e523d803e3c49d5d090ae5b9d36d43ce7311535
    Closes-Bug: #1774006
    Closes-Bug: #1767829
    (cherry picked from commit 11b41d73b48d46b7725bb97d4b85158d8bc2acfb)

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/ocata
Review: https://review.openstack.org/577197

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/577197
Committed: https://git.openstack.org/cgit/openstack/neutron/commit/?id=67de855c0d0f6d4834136d3452f5efde955e29dd
Submitter: Zuul
Branch: stable/ocata

commit 67de855c0d0f6d4834136d3452f5efde955e29dd
Author: Slawek Kaplonski <email address hidden>
Date: Tue Jun 5 10:05:58 2018 +0200

    [Fullstack] Wait for SG to be applied by L2 agent

    In fullstack test_securitygroup there were used simple
    net_helpers.assert_ping() and assert_not_ping() functions
    which tries to ping IP address 3 times with some short timeout
    and test fails if result of ping will not be as expected.

    Unfortunately sometimes in fullstack tests it might be not enough
    if test is creating new SG or SG rule or apply SG to port and
    just after that checks connection because L2 agent don't have
    enough time to apply all rules on "host".
    This patch changes it to use block_until_ping() and
    block_until_no_ping() methods from FakeMachine fixture.
    It will also check if ping is possible/not possible but
    will try to check it for 1 minute before fails.

    Similar change is also done for methods which checks TCP
    connectivity using netcat helper class. It now uses
    common_utils.wait_until_true() helper function instead of
    fail immediatelly.

    Change-Id: I9e523d803e3c49d5d090ae5b9d36d43ce7311535
    Closes-Bug: #1774006
    Closes-Bug: #1767829
    (cherry picked from commit 11b41d73b48d46b7725bb97d4b85158d8bc2acfb)

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/neutron 13.0.0.0b3 development milestone.

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/neutron 12.0.4 release.

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/neutron 11.0.6 release.

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/neutron ocata-eol release.