iSCSI cleanup issue if using sedtargets

[Gorka Eguileor]

With the latest iSCSI disconnect refactoring [1] we have introduced a regression where not all devices are properly cleaned up on a multipath connection if we are doing target discovery.

On connection we do the discovery so we get the targets, but on disconnection the code is expecting to have all the ips, iqns, and luns information on the connection properties so it will only flush the multipath and disconnect volumes that were connected using the information from the connection property.

[1] https://review.openstack.org/455392

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/475760

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/475760
Committed: https://git.openstack.org/cgit/openstack/os-brick/commit/?id=1905398f617ad02fb8546ac621eb1b292003a59a
Submitter: Jenkins
Branch: master

commit 1905398f617ad02fb8546ac621eb1b292003a59a
Author: Gorka Eguileor <email address hidden>
Date: Tue Jun 20 14:15:26 2017 +0200

    Fix iSCSI cleanup issue when using discovery

    With the latest iSCSI disconnect refactoring [1] we have introduced a
    regression where not all devices are properly cleaned up on a multipath
    connection if we are using target discovery.

    On connection we do the discovery so we get the targets, but on
    disconnection the code is expecting to have all the ips, iqns, and luns
    information on the connection properties, so it will only flush the
    multipath and disconnect volumes that were connected using the
    information from the connection property.

    This patch fixes this retrieving the targets information from the
    discoverydb that was gathered during the connection phase.

    Output from discoverydb query looks like this:

     SENDTARGETS:
     DiscoveryAddress: 192.168.1.33,3260
     DiscoveryAddress: 192.168.1.2,3260
     Target: iqn.2004-04.com.qnap:ts-831x:iscsi.cinder-20170531114245.9eff88
         Portal: 192.168.1.3:3260,1
             Iface Name: default
         Portal: 192.168.1.2:3260,1
             Iface Name: default
     Target: iqn.2004-04.com.qnap:ts-831x:iscsi.cinder-20170531114447.9eff88
         Portal: 192.168.1.3:3260,1
             Iface Name: default
         Portal: 192.168.1.2:3260,1
             Iface Name: default
     DiscoveryAddress: 192.168.1.38,3260
     iSNS:
     No targets found.
     STATIC:
     No targets found.
     FIRMWARE:
     No targets found.

    [1] https://review.openstack.org/455392

    Change-Id: Iada5d4fbeb07aeaf3afb953a289b6b89778c382c
    Closes-Bug: #1699061

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/os-brick 1.14.0 release.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/478893

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/478893
Committed: https://git.openstack.org/cgit/openstack/os-brick/commit/?id=be37c2e040e2d77c443dbd709f253ab9a35f721c
Submitter: Jenkins
Branch: master

commit be37c2e040e2d77c443dbd709f253ab9a35f721c
Author: Gorka Eguileor <email address hidden>
Date: Thu Jun 29 13:59:04 2017 +0200

    Fix iSCSI cleanup fix on discovery backends

    On Change-Id Iada5d4fbeb07aeaf3afb953a289b6b89778c382c we tried to fix
    an issue with the multipath detach of backends that used discovery on
    attach, but contrary to the commit message and the docstrings it didn't
    look for ip,port in the discoverydb but ip:port instead, which meant
    that it would never find what it was looking for.

    This patch fixes that fix to make it search for the right regex.

    TrivialFix
    Closes-Bug: #1699061

    Change-Id: Ibfa1a78a555e984c662f668677451f5a3ed55602

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/os-brick 1.15.0 release.