feat: enable T2 auto health check (#16825)

Description of PR Enable T2 auto health check and add running container info to return data. Summary: Fixes # (issue) Microsoft ADO 30293537 Approach What is the motivation for this PR? We want to enable T2 auto health check as we have now supported T2 auto recover. co-authorized by: [email protected]
sonic-net · Feb 12, 2025 · 4302661 · 4302661
1 parent 4a19378
commit 4302661
Showing 1 changed file with 5 additions and 8 deletions.
diff --git a/.azure-pipelines/testbed_health_check.py b/.azure-pipelines/testbed_health_check.py
@@ -218,14 +218,6 @@ def pre_check(self):
         if len(ipv4_not_exists_hosts) > 0:
             raise HostsUnreachable(self.check_result.errmsg)
 
-        # TODO: Refactor the following code to specify a "leader" T2 Testbed and skip the check on "followers"
-        # Retrieve the basic facts of the DUTs
-        if self.is_multi_asic:
-            errmsg = "Not support to perform checks on multi-asic DUT now."
-            logger.info(errmsg)
-
-            raise SkipCurrentTestbed(errmsg)
-
         logger.info("======================= pre_check ends =======================")
 
     def run_check(self):
@@ -461,6 +453,7 @@ def check_critical_containers_running(self, critical_containers: list = None):
             critical_containers = ["syncd", "swss", "bgp"]
 
         failed = False
+        running_containers_facts_on_hosts = {}
 
         logger.info("======================= check_critical_containers_running starts =======================")
 
@@ -479,6 +472,8 @@ def check_critical_containers_running(self, critical_containers: list = None):
             running_containers = sonichost.shell(r"docker ps -f 'status=running' --format \{\{.Names\}\}")[
                 'stdout_lines']
 
+            running_containers_facts_on_hosts[hostname] = running_containers
+
             containers_to_check = critical_containers
             if self.is_multi_asic:
                 if (self.is_chassis and
@@ -505,6 +500,8 @@ def check_critical_containers_running(self, critical_containers: list = None):
                     # Add errlog to check result errmsg
                     self.check_result.errmsg.append(errlog)
 
+        self.check_result.data["running_containers_facts_on_hosts"] = running_containers_facts_on_hosts
+
         logger.info("======================= check_critical_containers_running ends =======================")
 
         if failed: