cluster

commit a160e1e93ee4c0f39a4aec6bd68918bb03793f2c
parent d47f1507dab472116cc5b9f4f180cbed49e78f02
Author: MTRNord <MTRNord@users.noreply.github.com>
Date:   Sun,  5 Apr 2026 15:52:41 +0200

improve velero and velero alerts

Signed-off-by: MTRNord <MTRNord@users.noreply.github.com>

Diffstat:
M
infrastructure_talos/configs/velero-alerts.yaml
 | 
2
+-
M
infrastructure_talos/controllers/velero/release.yaml
 | 
3
+++

2 files changed, 4 insertions(+), 1 deletion(-)
diff --git a/infrastructure_talos/configs/velero-alerts.yaml b/infrastructure_talos/configs/velero-alerts.yaml
@@ -34,7 +34,7 @@ spec:
         - alert: VeleroBackupMissing
           expr: |
             time() - velero_backup_last_successful_timestamp{schedule!=""} > 90000
-          for: 5m
+          for: 1h
           labels:
             severity: critical
           annotations:
diff --git a/infrastructure_talos/controllers/velero/release.yaml b/infrastructure_talos/controllers/velero/release.yaml
@@ -52,6 +52,9 @@ spec:
       clientBurst: 200
       # Increase timeout to avoid "exposed PVB is not ready" errors on slow volumes
       podVolumeOperationTimeout: 60m
+      # Increase CSI snapshot timeout — Longhorn queues snapshot creation behind its
+      # concurrent task limit (5), so with many volumes the default 10m is too short.
+      csiSnapshotTimeout: 30m
       backupStorageLocation:
         - name: default
           provider: aws