summaryrefslogtreecommitdiffstats
path: root/health
diff options
context:
space:
mode:
authorIlya Mashchenko <ilya@netdata.cloud>2021-04-14 13:23:33 +0300
committerGitHub <noreply@github.com>2021-04-14 13:23:33 +0300
commit78b540bb75adc0298da08327a9aa95650540d2c5 (patch)
tree0f4697d1d8e0e70b74cc654ed946a9db96dec36a /health
parent563c6554efb48a7cfae28610aeecc60651ee3077 (diff)
health: add systemdunits alarms (#10906)
Diffstat (limited to 'health')
-rw-r--r--health/Makefile.am1
-rw-r--r--health/health.d/systemdunits.conf112
2 files changed, 113 insertions, 0 deletions
diff --git a/health/Makefile.am b/health/Makefile.am
index 4b26d6ae64..b963ea0cd1 100644
--- a/health/Makefile.am
+++ b/health/Makefile.am
@@ -90,6 +90,7 @@ dist_healthconfig_DATA = \
health.d/stiebeleltron.conf \
health.d/synchronization.conf \
health.d/swap.conf \
+ health.d/systemdunits.conf \
health.d/tcp_conn.conf \
health.d/tcp_listen.conf \
health.d/tcp_mem.conf \
diff --git a/health/health.d/systemdunits.conf b/health/health.d/systemdunits.conf
new file mode 100644
index 0000000000..b566d8e0e8
--- /dev/null
+++ b/health/health.d/systemdunits.conf
@@ -0,0 +1,112 @@
+## Check if the are any systemd units in the failed state (crashed).
+## States: 1 - active, 2 - inactive, 3 - activating, 4 - deactivating, 5 - failed.
+
+## Service units
+template: systemd_service_units_state
+ on: systemd.service_units_state
+ lookup: max -1s min2max
+ units: ok/failed
+ every: 10s
+ warn: $this != nan AND $this == 5
+ delay: down 5m multiplier 1.5 max 1h
+ info: one or more systemd service units are in the failed state
+ to: sysadmin
+
+## Socket units
+template: systemd_socket_units_state
+ on: systemd.socket_unit_state
+ lookup: max -1s min2max
+ units: ok/failed
+ every: 10s
+ warn: $this != nan AND $this == 5
+ delay: down 5m multiplier 1.5 max 1h
+ info: one or more systemd socket units are in the failed state
+ to: sysadmin
+
+## Target units
+template: systemd_target_units_state
+ on: systemd.target_unit_state
+ lookup: max -1s min2max
+ units: ok/failed
+ every: 10s
+ warn: $this != nan AND $this == 5
+ delay: down 5m multiplier 1.5 max 1h
+ info: one or more systemd target units are in the failed state
+ to: sysadmin
+
+## Path units
+template: systemd_path_units_state
+ on: systemd.path_unit_state
+ lookup: max -1s min2max
+ units: ok/failed
+ every: 10s
+ warn: $this != nan AND $this == 5
+ delay: down 5m multiplier 1.5 max 1h
+ info: one or more systemd path units are in the failed state
+ to: sysadmin
+
+## Device units
+template: systemd_device_units_state
+ on: systemd.device_unit_state
+ lookup: max -1s min2max
+ units: ok/failed
+ every: 10s
+ warn: $this != nan AND $this == 5
+ delay: down 5m multiplier 1.5 max 1h
+ info: one or more the systemd device units are in the failed state
+ to: sysadmin
+
+## Mount units
+template: systemd_mount_units_state
+ on: systemd.mount_unit_state
+ lookup: max -1s min2max
+ units: ok/failed
+ every: 10s
+ warn: $this != nan AND $this == 5
+ delay: down 5m multiplier 1.5 max 1h
+ info: one or more the systemd mount units are in the failed state
+ to: sysadmin
+
+## Automount units
+template: systemd_automount_units_state
+ on: systemd.automount_unit_state
+ lookup: max -1s min2max
+ units: ok/failed
+ every: 10s
+ warn: $this != nan AND $this == 5
+ delay: down 5m multiplier 1.5 max 1h
+ info: one or more systemd automount units are in the failed state
+ to: sysadmin
+
+## Swap units
+template: systemd_swap_units_state
+ on: systemd.swap_unit_state
+ lookup: max -1s min2max
+ units: ok/failed
+ every: 10s
+ warn: $this != nan AND $this == 5
+ delay: down 5m multiplier 1.5 max 1h
+ info: one or more systemd swap units are in the failed state
+ to: sysadmin
+
+## Scope units
+template: systemd_scope_units_state
+ on: systemd.scope_unit_state
+ lookup: max -1s min2max
+ units: ok/failed
+ every: 10s
+ warn: $this != nan AND $this == 5
+ delay: down 5m multiplier 1.5 max 1h
+ info: one or more systemd scope units are in the failed state
+ to: sysadmin
+
+## Slice units
+template: systemd_slice_units_state
+ on: systemd.slice_unit_state
+ lookup: max -1s min2max
+ units: ok/failed
+ every: 10s
+ warn: $this != nan AND $this == 5
+ delay: down 5m multiplier 1.5 max 1h
+ info: one or more systemd slice units are in the failed state
+ to: sysadmin