fix(alerting): Remove alert rules for detecting high disk read/write (#…

…149) * Remove alert rules for detecting high disk read/write This check does not make a lot of sense as a canned alert rule as it is way too context dependent. * Remove the HostReadonlyFilesystem alert --------- Co-authored-by: Leon <[email protected]>
canonical · Nov 28, 2024 · 151460b · 151460b
1 parent 5691548
commit 151460b
Showing 1 changed file with 0 additions and 41 deletions.
diff --git a/src/prometheus_alert_rules/disk.rules b/src/prometheus_alert_rules/disk.rules
@@ -29,17 +29,6 @@ groups:
         Host filesystem '{{ $labels.mountpoint }}' is using {{ $value | printf "%.0f" }}% of the total space.
           VALUE = {{ $value }}
           LABELS = {{ $labels }}
-  - alert: HostReadonlyFilesystem
-    expr: node_filesystem_readonly{mountpoint!~"/snap/.*|/sys/fs/cgroup/.*"} > 0
-    for: 0m
-    labels:
-      severity: warning
-    annotations:
-      summary: Host filesystem '{{ $labels.mountpoint }}' is readonly (instance {{ $labels.instance }})
-      description: >-
-        Host filesystem '{{ $labels.mountpoint }}' is readonly.
-          VALUE = {{ $value }}
-          LABELS = {{ $labels }}
   - alert: HostXFSError
     expr: node_filesystem_device_error{fstype="xfs"} > 0
     for: 0m
@@ -51,33 +40,3 @@ groups:
          XFS error found for device '{{ $labels.device }}'.
            VALUE = {{ $value }}
            LABELS = {{ $labels }}
-  - alert: HostHighDiskReadRate
-    expr: irate(node_disk_read_bytes_total[2m]) / 1024 / 1024 > 50
-    for: 5m
-    labels:
-      severity: warning
-    annotations:
-      summary: Host high disk '{{ $labels.device }}' read rate (instance {{ $labels.instance }})
-      description: >-
-        Host disk '{{ $labels.device }}' is probably reading too much data ({{ $value | printf "%.0f" }} > 50 MB/s) for last 5m.
-          VALUE = {{ $value }}
-          LABELS = {{ $labels }}
-  - alert: HostHighDiskWriteRate
-    expr: irate(node_disk_written_bytes_total[2m]) / 1024 / 1024 > 50
-    for: 5m
-    labels:
-      severity: warning
-    annotations:
-      summary: Host high disk '{{ $labels.device }}' write rate (instance {{ $labels.instance }})
-      description: >-
-        Host disk '{{ $labels.device }}' is probably writing too much data ({{ $value | printf "%.0f" }} > 50 MB/s) for last 5m.
-          VALUE = {{ $value }}
-          LABELS = {{ $labels }}
-  - alert: DiskRemoval
-    expr: count(node_disk_info{model!="", serial!="", wwn!=""}) by (instance, device) < on(instance, device) group_left() (count(node_disk_info{model!="", serial!="", wwn!=""} offset 30d) by (instance, device))
-    for: 5m
-    labels:
-      severity: warning
-    annotations:
-      summary: "Disk removal detected on {{ $labels.instance }}"
-      description: "Device {{ $labels.device }} on {{ $labels.instance }} has been removed which can be a hardware failure."