Skip to content

Commit b8be291

Browse files
authored
docs: update CATALOG.md for observability-pod-disruption-budget (#3498)
Regenerate CATALOG.md to reflect the zone-aware PDB validation changes introduced in PR #3467. The observability-pod-disruption-budget test case now documents the zone-aware checks that ensure workloads can survive an entire zone going offline during platform upgrades. Signed-off-by: Gonzalo Reyero Ferreras <greyerof@redhat.com>
1 parent a22e31a commit b8be291

File tree

1 file changed

+3
-3
lines changed

1 file changed

+3
-3
lines changed

CATALOG.md

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -1208,11 +1208,11 @@ Test Cases are the specifications used to perform a meaningful test. Test cases
12081208
|Property|Description|
12091209
|---|---|
12101210
|Unique ID|observability-pod-disruption-budget|
1211-
|Description|Checks to see if pod disruption budgets have allowed values for minAvailable and maxUnavailable|
1212-
|Suggested Remediation|Ensure minAvailable is not zero and maxUnavailable does not equal the number of pods in the replica|
1211+
|Description|Checks to see if pod disruption budgets have allowed values for minAvailable and maxUnavailable, and verifies that PDBs are zone-aware (can tolerate an entire zone going offline during platform upgrades)|
1212+
|Suggested Remediation|Ensure minAvailable is not zero and maxUnavailable does not equal the number of pods in the replica. For multi-zone clusters, also ensure the PDB is zone-aware: set maxUnavailable >= ceil(replicas/zones) or minAvailable <= replicas - ceil(replicas/zones) to allow draining all pods in one zone during platform upgrades.|
12131213
|Best Practice Reference|https://redhat-best-practices-for-k8s.github.io/guide/#k8s-best-practices-upgrade-expectations|
12141214
|Exception Process|No exceptions|
1215-
|Impact Statement|Improper disruption budgets can prevent necessary maintenance operations or allow too many pods to be disrupted simultaneously.|
1215+
|Impact Statement|Improper disruption budgets can prevent necessary maintenance operations or allow too many pods to be disrupted simultaneously. Non-zone-aware PDBs can block platform upgrades when all workers in a zone need to be drained.|
12161216
|Tags|common,observability|
12171217
|**Scenario**|**Optional/Mandatory**|
12181218
|Extended|Mandatory|

0 commit comments

Comments
 (0)