Fix the logic error for hostwd

[Tao Li]

Brief Description
-----------------
In each period, the pmon_grace_loops is reduced two. In fact, minus one is enough.

Severity
--------
major

Steps to Reproduce
------------------

Expected Behavior
------------------

Actual Behavior
----------------

Reproducibility
---------------

System Configuration
--------------------
STD system

Branch/Pull Time/Commit
-----------------------
2019-10-04_16

[Eric MacDonald (rocksolidmtce)]

You did not fill in the template.

What is the impact to you ?

Even though it decrements the number by 2 the start value is 5.
I recall the design being more than 2 back to back misses is a failure resulting in watchdog timeout reboot.

If there is a code change for this issue then be sure to maintain the current behavior in terms of misses before self reboot.

Guessing that would be removing one of the "ctrl->pmon_grace_loops--" and reducing the config value of "config->hostwd_failure_threshold" to 3.

Severity is minor.

I'll take the issue.

[Ghada Khalil (gkhalil)]

Low priority / does not gate any starlingx release given there is no end system impact.

[wanghao (wanghao749)]

Hi, Eric, there is a fixing patch: https://review.openstack.org/#/c/652982/. We can work on this together. Thanks.

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/652982
Committed: https://git.openstack.org/cgit/starlingx/metal/commit/?id=4150f91ff0687268e094a58ab177a12d56e97114
Submitter: Zuul
Branch: master

commit 4150f91ff0687268e094a58ab177a12d56e97114
Author: wanghao <email address hidden>
Date: Tue Apr 16 19:33:22 2019 +0800

    Fix the logic error in hostwd

    In each period, the pmon_grace_loops is reduced two. In fact,
    minus one is enough.

    Closes-Bug: #1824983
    Signed-off-by: wanghao <email address hidden>

    Change-Id: I38ad97be5554f320d3a6e7366e39afe518ae7a92

[Ghada Khalil (gkhalil)]

Adding the stx.2.0 release label as the fix was merged for that release

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/675452
Committed: https://git.openstack.org/cgit/starlingx/metal/commit/?id=8f9decd6e53c1712084399350aefb2b50b9f5236
Submitter: Zuul
Branch: master

commit 8f9decd6e53c1712084399350aefb2b50b9f5236
Author: Alex Kozyrev <email address hidden>
Date: Thu Aug 8 15:26:08 2019 -0400

    Fix number of allowed missed messages from pmon in hostwd

    The following commit attempted to fix failure logic in hostwd:
    4150f91ff0687268e094a58ab177a12d56e97114
    hostwd_failure_threshold has been reduced from 5 to 3 in it, but the
    logic around pmon_grace_loops hasn't been touched.
    hostwd_failure_threshold is still reduced by 2 every time we miss a
    message from pmon. So, we only allow 1 missed message before the reboot.
    Removing this double substraction and reverting to old beavior:
    allow 2 missed messages from PMON and reboot a system on a 3rd one.

    Closes-Bug: 1824983
    Change-Id: I57293f9383302d042faee78189409043e21b5801
    Signed-off-by: Alex Kozyrev <email address hidden>

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: r/stx.2.0
Review: https://review.opendev.org/676253

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/676253
Committed: https://git.openstack.org/cgit/starlingx/metal/commit/?id=7ffbe49712e8b12a69646d9ac98e07b156c4a64a
Submitter: Zuul
Branch: r/stx.2.0

commit 7ffbe49712e8b12a69646d9ac98e07b156c4a64a
Author: Alex Kozyrev <email address hidden>
Date: Thu Aug 8 15:26:08 2019 -0400

    Fix number of allowed missed messages from pmon in hostwd

    The following commit attempted to fix failure logic in hostwd:
    4150f91ff0687268e094a58ab177a12d56e97114
    hostwd_failure_threshold has been reduced from 5 to 3 in it, but the
    logic around pmon_grace_loops hasn't been touched.
    hostwd_failure_threshold is still reduced by 2 every time we miss a
    message from pmon. So, we only allow 1 missed message before the reboot.
    Removing this double substraction and reverting to old beavior:
    allow 2 missed messages from PMON and reboot a system on a 3rd one.

    Closes-Bug: 1824983
    Change-Id: I57293f9383302d042faee78189409043e21b5801
    Signed-off-by: Alex Kozyrev <email address hidden>
    (cherry picked from commit 8f9decd6e53c1712084399350aefb2b50b9f5236)

[Ghada Khalil (gkhalil)]

As per discussion with John Kung and Alex Kozyrev, the subsequent commit for this bug was cherrypicked to the stx.2.0 release branch as the original commit introduced an issue. See above for more details.