use container names to handle multiple at once

integration/scripts/test_configure.py

@@ -38,17 +38,19 @@ def run_test_windows(remote_host: u.Host, env_vars: dict) -> None:
     if _diagnose_result_is_successful(result):
         print(" ✅ observe-agent -> observe validation passed! ")
     else:
-        print(result)
+        u.print_remote_result(result)
         raise ValueError(
             f"❌ Failed: observe-agent -> observe validation (regex on diagnose output did not match)"
         )
 
 
 @u.print_test_decorator
 def run_test_docker(remote_host: u.Host, env_vars: dict) -> None:
+    container_name = u.get_random_test_name()
+    env_vars["container_name"] = container_name
     start.run_test_docker(remote_host, env_vars)
 
-    container_id = u.get_docker_container(remote_host)
+    container_id = u.get_docker_container(remote_host, container_name)
     exec_prefix = f"sudo docker exec {container_id} ./observe-agent"
     init_command = "{} init-config --token {} --observe_url {}".format(
         exec_prefix, env_vars["observe_token"], env_vars["observe_url"]
@@ -57,13 +59,22 @@ def run_test_docker(remote_host: u.Host, env_vars: dict) -> None:
 
     # Set up correct config with observe url and token
     result = remote_host.run_command(init_command)
+    if result.exited != 0 or result.stderr:
+        u.print_remote_result(result)
+        raise ValueError("❌ Error in init-config")
+
+    # Restart the process
+    result = remote_host.run_command(f"sudo docker restart {container_id}")
+    if result.exited != 0 or result.stderr:
+        u.print_remote_result(result)
+        raise ValueError("❌ Error in restart")
 
     # Check diagnose command
     result = remote_host.run_command(diagnose_command)
     if _diagnose_result_is_successful(result):
         print(" ✅ observe-agent -> observe validation passed! ")
     else:
-        print(result)
+        u.print_remote_result(result)
         raise ValueError(
             f"❌ Failed: observe-agent -> observe validation (regex on diagnose output did not match)"
         )
@@ -97,7 +108,7 @@ def run_test_linux(remote_host: u.Host, env_vars: dict) -> None:
     if _diagnose_result_is_successful(result):
         print(" ✅ observe-agent -> observe validation passed! ")
     else:
-        print(result)
+        u.print_remote_result(result)
         raise ValueError(
             f"❌ Failed: observe-agent -> observe validation (regex on diagnose output did not match)"
         )

integration/scripts/test_start.py

@@ -93,7 +93,8 @@ def run_test_windows(remote_host: u.Host, env_vars: dict) -> None:
 
 @u.print_test_decorator
 def run_test_docker(remote_host: u.Host, env_vars: dict) -> None:
-    docker_prefix = u.get_docker_prefix(remote_host, True)
+    container_name = env_vars.get("container_name", u.get_random_test_name())
+    docker_prefix = u.get_docker_prefix(remote_host, container_name, True)
     start_command = "start"
     start_timeout = 30  # how long to wait for observe-agent to start
 
@@ -105,7 +106,7 @@ def run_test_docker(remote_host: u.Host, env_vars: dict) -> None:
         print("✅ Observe Agent started successfully: " + result.stdout)
 
     # Get Observe Agent Container ID
-    container_id = u.get_docker_container(remote_host)
+    container_id = u.get_docker_container(remote_host, container_name)
     status_command = f"sudo docker exec {container_id} ./observe-agent status"
 
     # Check Agent Status

integration/scripts/test_version.py

@@ -70,7 +70,7 @@ def run_test_windows(remote_host: u.Host, env_vars: dict) -> None:
 
 @u.print_test_decorator
 def run_test_docker(remote_host: u.Host, env_vars: dict) -> None:
-    docker_prefix = u.get_docker_prefix(remote_host, False)
+    docker_prefix = u.get_docker_prefix(remote_host, u.get_random_test_name(), False)
     config_file_linux = "/etc/observe-agent/observe-agent.yaml"
     version_pattern = re.compile(r"^\d+\.\d+\.\d+(-[A-Za-z0-9-]+)?$")
     home_dir = "/home/{}".format(env_vars["user"])

integration/scripts/utils.py

@@ -5,6 +5,7 @@
 from paramiko.ssh_exception import AuthenticationException, NoValidConnectionsError
 
 import os
+import random
 import sys
 import time
 
@@ -247,6 +248,12 @@ def test_conection(self, timeout=60) -> None:
         raise RuntimeError(" ❌ The SSH connection failed")
 
 
+def get_random_test_name() -> str:
+    random.seed(time.time())
+    num_digits = 6
+    return "test-" + str(random.randint(10 ** (num_digits - 1), (10**num_digits) - 1))
+
+
 def get_docker_image(remote_host: Host) -> str:
     result = remote_host.run_command(
         'sudo docker images --format "{{.Repository}}:{{.Tag}}"'
@@ -258,7 +265,7 @@ def get_docker_image(remote_host: Host) -> str:
     return images[0]
 
 
-def get_docker_prefix(remote_host: Host, detach: bool) -> str:
+def get_docker_prefix(remote_host: Host, name: str, detach: bool) -> str:
     image = get_docker_image(remote_host)
     return f'sudo docker run {"-d --restart always" if detach else ""} \
         --mount type=bind,source=/proc,target=/hostfs/proc,readonly \
@@ -268,41 +275,36 @@ def get_docker_prefix(remote_host: Host, detach: bool) -> str:
         --mount type=bind,source=/var/log,target=/hostfs/var/log,readonly \
         --mount type=bind,source=/var/lib/docker/containers,target=/var/lib/docker/containers,readonly \
         --mount type=bind,source=$(pwd)/observe-agent.yaml,target=/etc/observe-agent/observe-agent.yaml \
+        --name {name} \
         --pid host {image}'
 
 
-def get_docker_container(remote_host: Host, retries: int = 3) -> str:
-    get_container_command = 'sudo docker ps --filter "status=running" --format "{{.ID}} {{.Image}} {{.CreatedAt}}"'
+def get_docker_container(remote_host: Host, name: str) -> str:
+    get_container_command = (
+        'sudo docker ps --filter "status=running" --filter "name='
+        + name
+        + '" --format "{{.ID}} {{.Image}} {{.CreatedAt}}"'
+    )
     result = remote_host.run_command(get_container_command)
     running = [
         line.strip() for line in result.stdout.splitlines() if "SNAPSHOT" in line
     ]
     if len(running) == 0:
-        if retries > 0 and result.stderr == "":
-            output = result.stdout.strip()
-            print(
-                "❌ Error in finding observe-agent container; retrying in 1s..."
-                + ("\n" + output if output != "" else ""),
-                file=sys.stderr,
-            )
-            time.sleep(1)
-            return get_docker_container(remote_host, retries - 1)
-        elif retries == 0:
-            # No retries left, see if we can get logs from stopped containers to help debug.
-            result = remote_host.run_command('sudo docker ps --format "{{.ID}}"')
-            if result.stdout != "":
-                container_ids = result.stdout.splitlines()
-                for container_id in container_ids:
-                    print(
-                        "Logs for container {}:".format(container_id),
-                        file=sys.stderr,
-                    )
-                    result = remote_host.run_command(
-                        "sudo docker logs {}".format(container_id)
-                    )
-                    print_remote_result(result)
-            else:
+        # No container matched our filter. Get logs from all containers to help debug.
+        result = remote_host.run_command('sudo docker ps --format "{{.ID}}"')
+        if result.stdout != "":
+            container_ids = result.stdout.splitlines()
+            for container_id in container_ids:
+                print(
+                    "Logs for container {}:".format(container_id),
+                    file=sys.stderr,
+                )
+                result = remote_host.run_command(
+                    "sudo docker logs {}".format(container_id)
+                )
                 print_remote_result(result)
+        else:
+            print_remote_result(result)
         die(
             "❌ Error in finding observe-agent container; command output:\n{}\ncommand error:\n{}".format(
                 result.stdout or "<empty>",
@@ -312,7 +314,7 @@ def get_docker_container(remote_host: Host, retries: int = 3) -> str:
         return ""
     if len(running) > 1:
         die(
-            "❌ Error in finding observe-agent container, to many snapshots running:\n"
+            "❌ Error in finding observe-agent container, too many snapshots running:\n"
             + result.stdout
         )
     # Only one snapshot running; return the ID from the first line.