Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[active-standby] Fix the oscillation logic #261

Merged
merged 2 commits into from
Jul 2, 2024

Conversation

lolyu
Copy link
Contributor

@lolyu lolyu commented Jul 1, 2024

Description of PR

Summary:
Fixes #253

Type of change

  • Bug fix
  • New feature
  • Doc/Design
  • Unit test

Approach

What is the motivation for this PR?

Fix the unexpected toggle introduced by the oscillation logic.

Signed-off-by: Longxiang Lyu lolv@microsoft.com

Work item tracking
  • Microsoft ADO (number only): 28397786

How did you do it?

The oscillation is introduced to allow the active side to toggle to standby if no heartbeat is received.
The workflow is described as the following:

(wait, active, up) ---> set oscillation timer [1]
...
(wait, active, up) ---> (wait, wait, up) [2]
(wait, wait, up) ---> (wait, active, up)
...
(wait, active, up) <--- oscillation timer expires, toggle to standby [3]

[1]: the ToR enters (wait, active, up), no heartbeats received, oscillation timer is set.
[2]: the ToR consistently probes the mux status, and transits between (wait, active, up) and (wait, wait, up).
[3]: when the oscillation timer expires and the ToR is (wait, active, up), make the toggle to standby request.

We need to ensure that, the ToR is only allowed to transits between (wait, active, up) and (wait, wait, up) during [2]. So any link prober active/standby, mux standby, or link down events should cancel the oscillation.

How did you verify/test it?

UT

Any platform specific information?

Documentation

Signed-off-by: Longxiang Lyu <lolv@microsoft.com>
yxieca
yxieca previously approved these changes Jul 1, 2024
Signed-off-by: Longxiang Lyu <lolv@microsoft.com>
@lolyu lolyu merged commit 2b7e4f9 into sonic-net:master Jul 2, 2024
9 checks passed
@yxieca
Copy link
Contributor

yxieca commented Jul 3, 2024

@lolyu do we not need this change in 202311?

@zjswhhh
Copy link
Contributor

zjswhhh commented Jul 3, 2024

@lolyu do we not need this change in 202311?

Was the issue found in production or nightly?

mssonicbld pushed a commit to mssonicbld/sonic-linkmgrd that referenced this pull request Aug 23, 2024
Approach
What is the motivation for this PR?
Fix the unexpected toggle introduced by the oscillation logic.

Signed-off-by: Longxiang Lyu lolv@microsoft.com

Work item tracking
Microsoft ADO (number only): 28397786
How did you do it?
The oscillation is introduced to allow the active side to toggle to standby if no heartbeat is received.
The workflow is described as the following:

(wait, active, up) ---> set oscillation timer [1]
...
(wait, active, up) ---> (wait, wait, up) [2]
(wait, wait, up) ---> (wait, active, up)
...
(wait, active, up) <--- oscillation timer expires, toggle to standby [3]
[1]: the ToR enters (wait, active, up), no heartbeats received, oscillation timer is set.
[2]: the ToR consistently probes the mux status, and transits between (wait, active, up) and (wait, wait, up).
[3]: when the oscillation timer expires and the ToR is (wait, active, up), make the toggle to standby request.

We need to ensure that, the ToR is only allowed to transits between (wait, active, up) and (wait, wait, up) during [2]. So any link prober active/standby, mux standby, or link down events should cancel the oscillation.

How did you verify/test it?
UT

Any platform specific information?
Documentation
mssonicbld pushed a commit to mssonicbld/sonic-linkmgrd that referenced this pull request Aug 23, 2024
Approach
What is the motivation for this PR?
Fix the unexpected toggle introduced by the oscillation logic.

Signed-off-by: Longxiang Lyu lolv@microsoft.com

Work item tracking
Microsoft ADO (number only): 28397786
How did you do it?
The oscillation is introduced to allow the active side to toggle to standby if no heartbeat is received.
The workflow is described as the following:

(wait, active, up) ---> set oscillation timer [1]
...
(wait, active, up) ---> (wait, wait, up) [2]
(wait, wait, up) ---> (wait, active, up)
...
(wait, active, up) <--- oscillation timer expires, toggle to standby [3]
[1]: the ToR enters (wait, active, up), no heartbeats received, oscillation timer is set.
[2]: the ToR consistently probes the mux status, and transits between (wait, active, up) and (wait, wait, up).
[3]: when the oscillation timer expires and the ToR is (wait, active, up), make the toggle to standby request.

We need to ensure that, the ToR is only allowed to transits between (wait, active, up) and (wait, wait, up) during [2]. So any link prober active/standby, mux standby, or link down events should cancel the oscillation.

How did you verify/test it?
UT

Any platform specific information?
Documentation
@mssonicbld
Copy link
Collaborator

Cherry-pick PR to 202405: #269

@mssonicbld
Copy link
Collaborator

Cherry-pick PR to 202311: #270

mssonicbld pushed a commit that referenced this pull request Aug 23, 2024
Approach
What is the motivation for this PR?
Fix the unexpected toggle introduced by the oscillation logic.

Signed-off-by: Longxiang Lyu lolv@microsoft.com

Work item tracking
Microsoft ADO (number only): 28397786
How did you do it?
The oscillation is introduced to allow the active side to toggle to standby if no heartbeat is received.
The workflow is described as the following:

(wait, active, up) ---> set oscillation timer [1]
...
(wait, active, up) ---> (wait, wait, up) [2]
(wait, wait, up) ---> (wait, active, up)
...
(wait, active, up) <--- oscillation timer expires, toggle to standby [3]
[1]: the ToR enters (wait, active, up), no heartbeats received, oscillation timer is set.
[2]: the ToR consistently probes the mux status, and transits between (wait, active, up) and (wait, wait, up).
[3]: when the oscillation timer expires and the ToR is (wait, active, up), make the toggle to standby request.

We need to ensure that, the ToR is only allowed to transits between (wait, active, up) and (wait, wait, up) during [2]. So any link prober active/standby, mux standby, or link down events should cancel the oscillation.

How did you verify/test it?
UT

Any platform specific information?
Documentation
mssonicbld pushed a commit that referenced this pull request Aug 23, 2024
Approach
What is the motivation for this PR?
Fix the unexpected toggle introduced by the oscillation logic.

Signed-off-by: Longxiang Lyu lolv@microsoft.com

Work item tracking
Microsoft ADO (number only): 28397786
How did you do it?
The oscillation is introduced to allow the active side to toggle to standby if no heartbeat is received.
The workflow is described as the following:

(wait, active, up) ---> set oscillation timer [1]
...
(wait, active, up) ---> (wait, wait, up) [2]
(wait, wait, up) ---> (wait, active, up)
...
(wait, active, up) <--- oscillation timer expires, toggle to standby [3]
[1]: the ToR enters (wait, active, up), no heartbeats received, oscillation timer is set.
[2]: the ToR consistently probes the mux status, and transits between (wait, active, up) and (wait, wait, up).
[3]: when the oscillation timer expires and the ToR is (wait, active, up), make the toggle to standby request.

We need to ensure that, the ToR is only allowed to transits between (wait, active, up) and (wait, wait, up) during [2]. So any link prober active/standby, mux standby, or link down events should cancel the oscillation.

How did you verify/test it?
UT

Any platform specific information?
Documentation
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

[active-standby] Unexpected toggle due to the oscillation logic
4 participants