Skip to content

Conversation

@github-actions
Copy link
Contributor

PR Body: Backend persist lastUpdateMs, it will be modified by heartbeat editlog. But heartbeat editlog not persist hbTime, and hbTime always equal 0, it will make backend's lastUpdateMs = 0 in bdb image.

fix details:

  1. heartbeat response persist hbTime;
  2. only be state change will write an editlog. but we make a change: even a backend is healthy, still write a healthy response editlog every 5 min. Inorder to make backend's lastUpdateMs periodly updated in bdb image. But notice that this change wouldn't increase real editlog num. Because heartbeat mgr will patch all fe/be's heartbeat into one editlog. Even no fe/be state change, it still write an editlog which not contains any node's response.
  3. for a dead heartbeat response, set hbTime to last succ hbTime, then replayer can set correct lastUpdateMs;
    Cherry-picked from [fix](heartbeat) fix heartbeat editlog no persist hbTime #42653

Backend persist lastUpdateMs, it will be modified by heartbeat editlog.
But heartbeat editlog not persist hbTime, and hbTime always equal 0, it
will make backend's lastUpdateMs = 0 in bdb image.

fix details:
1. heartbeat response persist hbTime;
2. only be state change will write an editlog. but we make a change:
even a backend is healthy, still write a healthy response editlog every
5 min. Inorder to make backend's lastUpdateMs periodly updated in bdb
image. But notice that this change wouldn't increase real editlog num.
Because heartbeat mgr will patch all fe/be's heartbeat into one editlog.
Even no fe/be state change, it still write an editlog which not contains
any node's response.
3. for a dead heartbeat response, set hbTime to last succ hbTime, then
replayer can set correct lastUpdateMs;
@github-actions
Copy link
Contributor Author

run buildall

@doris-robot
Copy link

Thank you for your contribution to Apache Doris.
Don't know what should be done next? See How to process your PR

Since 2024-03-18, the Document has been moved to doris-website.
See Doris Document.

@yujun777 yujun777 closed this Nov 7, 2024
@yujun777 yujun777 reopened this Nov 7, 2024
@dataroaring
Copy link
Contributor

run buildall

@dataroaring dataroaring merged commit 7abf3c5 into branch-3.0 Nov 7, 2024
@dataroaring dataroaring deleted the auto-pick-42653-branch-3.0 branch December 27, 2024 07:15
gavinchou pushed a commit that referenced this pull request Nov 21, 2025
…er versions (#58135)

Prior to this #42986 PR, the
`getLastUpdateMs` field of the backend was not written to the fe image.
This caused the tablet mapping on the cloud to lose the PrimaryBackend
information after the upgrade, resulting in the regeneration of an
incorrect PrimaryBackend and uneven tablet distribution.

detail:
<img width="930" height="859" alt="image"
src="https://github.com/user-attachments/assets/23d62adc-0bbb-49cb-b5b0-d1768f6422c7"
/>
github-actions bot pushed a commit that referenced this pull request Nov 21, 2025
…er versions (#58135)

Prior to this #42986 PR, the
`getLastUpdateMs` field of the backend was not written to the fe image.
This caused the tablet mapping on the cloud to lose the PrimaryBackend
information after the upgrade, resulting in the regeneration of an
incorrect PrimaryBackend and uneven tablet distribution.

detail:
<img width="930" height="859" alt="image"
src="https://github.com/user-attachments/assets/23d62adc-0bbb-49cb-b5b0-d1768f6422c7"
/>
github-actions bot pushed a commit that referenced this pull request Nov 21, 2025
…er versions (#58135)

Prior to this #42986 PR, the
`getLastUpdateMs` field of the backend was not written to the fe image.
This caused the tablet mapping on the cloud to lose the PrimaryBackend
information after the upgrade, resulting in the regeneration of an
incorrect PrimaryBackend and uneven tablet distribution.

detail:
<img width="930" height="859" alt="image"
src="https://github.com/user-attachments/assets/23d62adc-0bbb-49cb-b5b0-d1768f6422c7"
/>
yiguolei pushed a commit that referenced this pull request Nov 26, 2025
…er versions (#58135)

Prior to this #42986 PR, the
`getLastUpdateMs` field of the backend was not written to the fe image.
This caused the tablet mapping on the cloud to lose the PrimaryBackend
information after the upgrade, resulting in the regeneration of an
incorrect PrimaryBackend and uneven tablet distribution.

detail:
<img width="930" height="859" alt="image"
src="https://github.com/user-attachments/assets/23d62adc-0bbb-49cb-b5b0-d1768f6422c7"
/>
nagisa-kunhah pushed a commit to nagisa-kunhah/doris that referenced this pull request Dec 14, 2025
…er versions (apache#58135)

Prior to this apache#42986 PR, the
`getLastUpdateMs` field of the backend was not written to the fe image.
This caused the tablet mapping on the cloud to lose the PrimaryBackend
information after the upgrade, resulting in the regeneration of an
incorrect PrimaryBackend and uneven tablet distribution.

detail:
<img width="930" height="859" alt="image"
src="https://github.com/user-attachments/assets/23d62adc-0bbb-49cb-b5b0-d1768f6422c7"
/>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants