[pull] master from DataDog:master

cisco_aci/README.md

@@ -21,8 +21,6 @@ By integrating ACI metrics with Datadog's infrastructure and application monitor
 
 ## Setup
 
-**The Cisco ACI integration is Generally Available. To learn more about billing implications, visit our [pricing page][19].**
-
 ### Installation
 
 The Cisco ACI check is packaged with the Agent, so simply [install the Agent][1] on a server within your network.
@@ -201,5 +199,4 @@ Contact [Datadog support][9].
 [15]: /dash/integration/242/cisco-aci---overview
 [16]: /logs
 [17]: https://docs.datadoghq.com/logs/log_collection/?tab=host#setup
-[18]: https://github.com/DataDog/integrations-core/blob/master/cisco_aci/tests/cisco_aci_query.py
-[19]: https://www.datadoghq.com/pricing/?product=network-monitoring&tab=network-device-monitoring#products
+[18]: https://github.com/DataDog/integrations-core/blob/master/cisco_aci/tests/cisco_aci_query.py

cisco_sdwan/README.md

@@ -20,8 +20,6 @@ By surfacing traffic and throughput metrics at the device and interface level, t
 
 ## Setup
 
-**The Cisco SD-WAN NDM integration is Generally Available. To learn more about billing implications, visit our [pricing page][8].**
-
 Follow the instructions below to install and configure this check for an Agent running on a host. For containerized environments, see the [Autodiscovery Integration Templates][3] for guidance on applying these instructions.
 
 ### Installation
@@ -63,4 +61,3 @@ Need help? Contact [Datadog support][7].
 [5]: https://docs.datadoghq.com/agent/guide/agent-commands/#start-stop-and-restart-the-agent
 [6]: https://github.com/DataDog/integrations-core/blob/master/cisco_sdwan/metadata.csv
 [7]: https://docs.datadoghq.com/help/
-[8]: https://www.datadoghq.com/pricing/?product=network-monitoring&tab=network-device-monitoring#products

ddev/changelog.d/23646.fixed

@@ -0,0 +1 @@
+Retry agent check invocations on transient failures to address SNMP E2E flake from autodiscovery reload races.

ddev/src/ddev/cli/env/agent.py

@@ -9,6 +9,26 @@
 
 if TYPE_CHECKING:
     from ddev.cli.application import Application
+    from ddev.e2e.agent.interface import AgentInterface
+
+
+def _invoke_check_with_retry(agent: AgentInterface, args: list[str], *, retries: int = 3, backoff: float = 0.5) -> None:
+    """Invoke ``agent check`` with bounded retry to absorb transient autodiscovery-reload races."""
+    import subprocess
+    import time
+
+    for attempt in range(retries + 1):
+        try:
+            agent.invoke(args)
+            return
+        except subprocess.CalledProcessError:
+            if attempt >= retries:
+                raise
+            click.echo(
+                f'agent check failed (attempt {attempt + 1}/{retries + 1}), retrying in {backoff:.1f}s...',
+                err=True,
+            )
+            time.sleep(backoff)
 
 
 @click.command(
@@ -54,7 +74,10 @@ def agent(app: Application, *, intg_name: str, environment: str, args: tuple[str
 
     if config_file is None or not trigger_run:
         try:
-            agent.invoke(full_args)
+            if trigger_run:
+                _invoke_check_with_retry(agent, full_args)
+            else:
+                agent.invoke(full_args)
         except subprocess.CalledProcessError as e:
             app.abort(code=e.returncode)
 
@@ -67,14 +90,14 @@ def agent(app: Application, *, intg_name: str, environment: str, args: tuple[str
     if not env_data.config_file.is_file():
         try:
             env_data.write_config(config)
-            agent.invoke(full_args)
+            _invoke_check_with_retry(agent, full_args)
         finally:
             env_data.config_file.unlink()
     else:
         temp_config_file = env_data.config_file.parent / f'{env_data.config_file.name}.bak.example'
         env_data.config_file.replace(temp_config_file)
         try:
             env_data.write_config(config)
-            agent.invoke(full_args)
+            _invoke_check_with_retry(agent, full_args)
         finally:
             temp_config_file.replace(env_data.config_file)

versa/README.md

@@ -22,8 +22,6 @@ This integration provides insight into how applications and users consume bandwi
 
 ## Setup
 
-**The Versa NDM integration is Generally Available. To learn more about billing implications, visit our [pricing page][10].**
-
 Follow the instructions below to install and configure this check for an Agent running on a host. For containerized environments, see the [Autodiscovery Integration Templates][3] for guidance on applying these instructions.
 
 ### Installation
@@ -70,5 +68,4 @@ Need help? Contact [Datadog support][9].
 [6]: https://docs.datadoghq.com/agent/configuration/agent-commands/#agent-status-and-information
 [7]: https://github.com/DataDog/integrations-core/blob/master/versa/metadata.csv
 [8]: https://github.com/DataDog/integrations-core/blob/master/versa/assets/service_checks.json
-[9]: https://docs.datadoghq.com/help/
-[10]: https://www.datadoghq.com/pricing/?product=network-monitoring&tab=network-device-monitoring#products
+[9]: https://docs.datadoghq.com/help/