This is an automated email from the ASF dual-hosted git repository.

bobhan1 pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/doris.git


The following commit(s) were added to refs/heads/master by this push:
     new 16994af355c [fix](regression) Fix tablet state change publish test 
(#68434)
16994af355c is described below

commit 16994af355ce36c1c20436b933b687ef7cbf5211
Author: bobhan1 <[email protected]>
AuthorDate: Thu Sep 24 14:52:09 2026 +0800

    [fix](regression) Fix tablet state change publish test (#68434)
    
    ### What problem does this PR solve?
    
    Issue Number: N/A
    
    Related PR: #49332, #48400
    
    Problem Summary:
    
    The tablet state change publish regression still expects NULL in two
    shadow-column values after an INT-to-VARCHAR schema change. Since
    #49332, concurrent INSERTs correctly populate these values as `99` and
    `88`. Regenerate the expected output with the regression framework.
    
    A local run also hit `EHOSTDOWN` in the first concurrent INSERT after
    restarting a BE. `startBackends()` waits for heartbeat and HTTP
    readiness, while the peer BE's cached brpc connection can still be
    recovering. Add a five-second wait after restart, while
    `ReportHandler.block` remains enabled, before resuming the existing
    fault-injection sequence. The case already uses a one-second FE
    heartbeat interval.
    
    ### Release note
    
    None
    
    ### Check List (For Author)
    
    - Test
        - [x] Regression test
    - `./build.sh --be --fe --cloud -j100` passed; FE, BE and MS runtime
    logs confirmed the rebuilt base revision.
    - The scoped Docker Cloud suite passed with `-forceGenOut` before adding
    the restart wait. The generated output changes only the two expected
    cells; no uncaught worker-thread exception occurred in that run.
    - Final normal-mode validation was blocked before suite execution by a
    missing local runtime image (`ImageNotFound`). The added restart wait is
    not yet runtime-validated; CI validation is pending.
          - `git diff --check` passed.
        - [ ] Unit Test
        - [ ] Manual test
        - [ ] No need to test or manual test
    
    - Behavior changed:
        - [x] No.
        - [ ] Yes.
    
    - Does this need documentation?
        - [x] No.
        - [ ] Yes.
    
    ### Check List (For Reviewer who merge this PR)
    
    - [ ] Confirm the release note
    - [ ] Confirm test cases
    - [ ] Confirm document
    - [ ] Add branch pick label
---
 .../cloud/test_tablet_state_change_in_publish_phase.out               | 4 ++--
 .../cloud/test_tablet_state_change_in_publish_phase.groovy            | 3 +++
 2 files changed, 5 insertions(+), 2 deletions(-)

diff --git 
a/regression-test/data/fault_injection_p0/cloud/test_tablet_state_change_in_publish_phase.out
 
b/regression-test/data/fault_injection_p0/cloud/test_tablet_state_change_in_publish_phase.out
index 0ece86d0fb4..fc3adc65465 100644
--- 
a/regression-test/data/fault_injection_p0/cloud/test_tablet_state_change_in_publish_phase.out
+++ 
b/regression-test/data/fault_injection_p0/cloud/test_tablet_state_change_in_publish_phase.out
@@ -13,8 +13,8 @@
 -- !dup_key_count --
 
 -- !sql --
-1      \N      99
+1      99      99
 2      2       2
 3      3       3
-10     \N      88
+10     88      88
 
diff --git 
a/regression-test/suites/fault_injection_p0/cloud/test_tablet_state_change_in_publish_phase.groovy
 
b/regression-test/suites/fault_injection_p0/cloud/test_tablet_state_change_in_publish_phase.groovy
index c3c245ce37e..f949101f62d 100644
--- 
a/regression-test/suites/fault_injection_p0/cloud/test_tablet_state_change_in_publish_phase.groovy
+++ 
b/regression-test/suites/fault_injection_p0/cloud/test_tablet_state_change_in_publish_phase.groovy
@@ -90,6 +90,9 @@ suite("test_tablet_state_change_in_publish_phase", "docker") {
             // block FE's task report handler to avoid alter task re-sended to 
BE before we enable debug points for SC
             GetDebugPoint().enableDebugPointForAllFEs("ReportHandler.block")
             cluster.startBackends(beIndex)
+            // Heartbeat/HTTP readiness does not imply the peer BE's brpc 
connection has recovered.
+            // Wait longer than brpc's default 3s health-check interval before 
starting load 1.
+            Thread.sleep(5000)
             
GetDebugPoint().enableDebugPointForAllBEs("CloudSchemaChangeJob::_convert_historical_rowsets.block")
             
GetDebugPoint().enableDebugPointForAllBEs("CloudSchemaChangeJob.process_alter_tablet.sleep")
             GetDebugPoint().disableDebugPointForAllFEs("ReportHandler.block")


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to