Merge pull request #66488 from ClickHouse/ci_gh_retry

CI: Add retry for GH set_status_comment call

tests/ci/ci.py

@@ -6,6 +6,7 @@ import os
 import re
 import subprocess
 import sys
+import time
 from dataclasses import dataclass
 from pathlib import Path
 from typing import Any, Dict, List, Optional
@@ -550,7 +551,17 @@ def _update_gh_statuses_action(indata: Dict, s3: S3Helper) -> None:
             except Exception as e:
                 raise e
     print("Going to update overall CI report")
-    set_status_comment(commit, pr_info)
+    for retry in range(2):
+        try:
+            set_status_comment(commit, pr_info)
+            break
+        except Exception as e:
+            print(
+                f"WARNING: Failed to update CI Running status, attempt [{retry + 1}], exception [{e}]"
+            )
+            time.sleep(1)
+    else:
+        print("ERROR: All retry attempts failed.")
     print("... CI report update - done")
 
 
@@ -996,10 +1007,10 @@ def main() -> int:
             args.skip_jobs,
         )
 
+        ci_cache.print_status()
         if IS_CI and pr_info.is_pr and not ci_settings.no_ci_cache:
             ci_cache.filter_out_not_affected_jobs()
-
-        ci_cache.print_status()
+            ci_cache.print_status()
 
         if IS_CI and not pr_info.is_merge_queue:
             # wait for pending jobs to be finished, await_jobs is a long blocking call

tests/ci/ci_cache.py

@@ -714,7 +714,7 @@ class CiCache:
             if CI.is_test_job(job_name) and job_name != CI.JobNames.BUILD_CHECK:
                 if job_config.reference_job_name:
                     reference_name = job_config.reference_job_name
-                    reference_config = self.jobs_to_do[reference_name]
+                    reference_config = CI.JOB_CONFIGS[reference_name]
                 else:
                     reference_name = job_name
                     reference_config = job_config
@@ -745,7 +745,8 @@ class CiCache:
                 del self.jobs_to_do[job]
             if job in self.jobs_to_wait:
                 del self.jobs_to_wait[job]
-            self.jobs_to_skip.append(job)
+            if job in self.jobs_to_skip:
+                self.jobs_to_skip.remove(job)
 
     def await_pending_jobs(self, is_release: bool, dry_run: bool = False) -> None:
         """