Merge branch 'master' into feature/prometheus_apis

docs/configuration/defining-playbooks/index.rst

@@ -17,4 +17,4 @@ Learn how to define Robusta playbooks.
 List of All Triggers and Actions
 ---------------------------------
 
-Refer to :ref:`Playbook Reference<playbook-reference>`
+Refer to :ref:`Triggers Reference` or :ref:`Actions Reference`

docs/configuration/sinks/slack.rst

@@ -117,10 +117,10 @@ If you can't use the `official Slack app <https://slack.com/apps/A0214S5PHB4-rob
 your own. This is not recommended for most companies due to the added complexity.
 
 1. `Create a new Slack app. <https://api.slack.com/apps?new_app=1>`_
-2. Enable Socket mode in your Slack App and copy the websocket token into the Robusta deployment yaml.
-3. Under "OAuth and Permissions" add the following scopes: chat:write, files:write, incoming-webhook, and channels:history
-4. Under "Event Subscriptions" add bot user events for message.channels and press "Save Changes"
-5. Click "Install into Workspace"
+2. Enable Socket mode in your Slack App.
+3. Under "OAuth and Permissions" add the following scopes: chat:write, chat:write.public, files:write, incoming-webhook, and channels:history.
+4. Under "Event Subscriptions" add bot user events for message.channels and press "Save Changes".
+5. Click "Install into Workspace".
 6. Copy the ``Bot User OAuth Token`` from "OAuth and Permissions".
 7. Add the token to SinksConfig in your `generated_values.yaml` file.
 

playbooks/robusta_playbooks/k8s_resource_enrichments.py

@@ -48,6 +48,8 @@ class RelatedContainer(BaseModel):
     status: Optional[str] = None
     created: Optional[str] = None
     ports: List[Any] = []
+    statusMessage: Optional[str] = None
+    statusReason: Optional[str] = None
 
 
 class RelatedPod(BaseModel):
@@ -64,6 +66,7 @@ class RelatedPod(BaseModel):
     addresses: str
     containers: List[RelatedContainer]
     status: Optional[str] = None
+    statusReason: Optional[str] = None
 
 
 supported_resources = ["Deployment", "DaemonSet", "ReplicaSet", "Pod", "StatefulSet", "Job", "Node"]
@@ -87,6 +90,7 @@ def to_pod_row(pod: Pod, cluster_name: str) -> List:
         addresses,
         len(pod.spec.containers),
         pod.status.phase,
+        pod.status.reason,
     ]
 
 
@@ -95,7 +99,6 @@ def get_related_pods(resource) -> list[Pod]:
     if kind not in supported_resources:
         raise ActionException(ErrorCodes.RESOURCE_NOT_SUPPORTED, f"Related pods is not supported for resource {kind}")
 
-    pods = []
     if kind == "Job":
         job_pods = get_job_all_pods(resource)
         pods = job_pods if job_pods else []
@@ -128,6 +131,7 @@ def to_pod_obj(pod: Pod, cluster: str) -> RelatedPod:
         addresses=addresses,
         containers=get_pod_containers(pod),
         status=pod.status.phase,
+        statusReason=pod.status.reason,
     )
 
 
@@ -156,6 +160,8 @@ def get_pod_containers(pod: Pod) -> List[RelatedContainer]:
                 memoryRequest=requests.memory,
                 restarts=getattr(containerStatus, "restartCount", 0),
                 status=stateStr,
+                statusMessage=getattr(state, "messsage", None) if state else None,
+                statusReason=getattr(state, "reason", None) if state else None,
                 created=getattr(state, "startedAt", None),
                 ports=[port.to_dict() for port in container.ports] if container.ports else [],
             )
@@ -197,6 +203,7 @@ def related_pods(event: KubernetesResourceEvent, params: RelatedPodParams):
                         "addresses",
                         "containers",
                         "status",
+                        "status reason",
                     ],
                     rows=rows,
                 )

playbooks/robusta_playbooks/krr.py

@@ -352,15 +352,15 @@ def krr_scan(event: ExecutionBaseEvent, params: KRRParams):
     except json.JSONDecodeError:
         logging.error(f"*KRR scan job failed. Expecting json result.*\n\n Result:\n{logs}")
         return
-    except ValidationError as e:
-        logging.error(f"*KRR scan job failed. Result format issue.*\n\n {e}")
+    except ValidationError:
+        logging.error("*KRR scan job failed. Result format issue.*\n\n", exc_info=True)
         logging.error(f"\n {logs}")
         return
     except Exception as e:
         if str(e) == "Failed to reach wait condition":
-            logging.error(f"*KRR scan job failed. The job wait condition timed out ({params.timeout}s)*")
+            logging.error(f"*KRR scan job failed. The job wait condition timed out ({params.timeout}s)*", exc_info=True)
         else:
-            logging.error(f"*KRR scan job unexpected error.*\n {e}")
+            logging.error(f"*KRR scan job unexpected error.*\n {e}", exc_info=True)
         return
 
     scan_block = ScanReportBlock(

playbooks/robusta_playbooks/pod_investigator_enricher.py

@@ -1,7 +1,7 @@
 import datetime
 import logging
 from enum import Enum
-from typing import List, Optional
+from typing import List, Optional, Tuple
 
 from hikaru.model.rel_1_26 import Pod, PodList
 from robusta.api import (
@@ -59,8 +59,10 @@ def pod_issue_investigator(event: KubernetesResourceEvent):
         logging.info(f"No pod issues discovered for {resource.kind} {resource.metadata.name}")
         return
     # Investigate first issue found
-    pod_issue = detect_pod_issue(pods_with_issues[0])
-    report_pod_issue(event, pods_with_issues, pod_issue)
+    first_pod = pods_with_issues[0]
+    pod_issue = detect_pod_issue(first_pod)
+    message, reason = get_pod_issue_message_and_reason(first_pod)
+    report_pod_issue(event, pods_with_issues, pod_issue, message, reason)
 
 
 def detect_pod_issue(pod: Pod) -> PodIssue:
@@ -75,6 +77,18 @@ def detect_pod_issue(pod: Pod) -> PodIssue:
     return PodIssue.NoneDetected
 
 
+def get_pod_issue_message_and_reason(pod: Pod) -> Tuple[Optional[str], Optional[str]]:
+    # Works/should work only or KubeContainerWaiting and KubePodNotReady
+    # Note: in line with the old code in pod_issue_investigator, we only get the message for
+    # the first of possibly many misbehaving containers.
+    if pod.status.containerStatuses:
+        if pod.status.containerStatuses[0].state.waiting:
+            return (
+                pod.status.containerStatuses[0].state.waiting.message,
+                pod.status.containerStatuses[0].state.waiting.reason,
+            )
+
+
 def is_pod_pending(pod: Pod) -> bool:
     return pod.status.phase.lower() == "pending"
 
@@ -124,7 +138,9 @@ def has_image_pull_issue(pod: Pod) -> bool:
     return len(image_pull_statuses) > 0
 
 
-def report_pod_issue(event: KubernetesResourceEvent, pods: List[Pod], issue: PodIssue):
+def report_pod_issue(
+    event: KubernetesResourceEvent, pods: List[Pod], issue: PodIssue, message: Optional[str], reason: Optional[str]
+):
     # find pods with issues
     pods_with_issue = [pod for pod in pods if detect_pod_issue(pod) == issue]
     pod_names = [pod.metadata.name for pod in pods_with_issue]
@@ -145,6 +161,11 @@ def report_pod_issue(event: KubernetesResourceEvent, pods: List[Pod], issue: Pod
         blocks.extend(additional_blocks)
         event.add_enrichment(blocks)
 
+    if reason:
+        if message is None:
+            message = "unknown"
+        event.add_enrichment([MarkdownBlock(f"\n\n{reason}: {message}")])
+
 
 def get_expected_replicas(event: KubernetesResourceEvent) -> int:
     resource = event.get_resource()

src/robusta/integrations/prometheus/models.py

@@ -113,7 +113,7 @@ def get_prometheus_query(self) -> str:
             parsed_url = urlparse(url)
             query_params = parse_qs(parsed_url.query)
 
-            q_expr = query_params.get('g0.expr', [])
+            q_expr = query_params.get("g0.expr", [])
             if len(q_expr) < 1 or not q_expr[0]:
                 return ""