github/node-problem-detector
1
0
Fork 0
mirror of https://github.com/kubernetes/node-problem-detector.git synced 2026-08-19 04:06:24 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix invalid config JSON and enforce canonical config format

Browse Source 
config/disk-log-message-filelog.json contained trailing commas and no
trailing newline, so it could not be parsed by encoding/json and NPD
would fail to start when configured with it.

- Fix the invalid JSON
- Add TestShippedConfigs, which verifies that every shipped JSON config
  under config/ (top level and one directory down) parses (catching
  trailing commas, UTF-8 BOMs and other syntax errors) and is in the
  canonical format (2-space indent, trailing newline), showing a diff on
  failure
- Add `make fmt-configs` to rewrite nonconforming files in place
  (UPDATE_EXPECTED=y)
- Normalize the existing config files to the canonical format
  (whitespace-only changes)
This commit is contained in:
Ciprian Hacman
2026-07-22 07:58:20 +03:00
parent fe3ab8c84e
commit 87dfadb120
18 changed files with 791 additions and 565 deletions
Download Patch File Download Diff File Expand all files Collapse all files
config/abrt-adaptor.json
+36 -36
View File
@@ -1,38 +1,38 @@
{
"plugin": "journald",
"pluginConfig": {
"source": "abrt-notification"
},
"logPath": "/var/log/journal",
"lookback": "5m",
"bufferSize": 10,
"source": "abrt-adaptor",
"conditions": [],
"rules": [
{
"type": "temporary",
"reason": "CCPPCrash",
"pattern": "Process \\d+ \\(\\S+\\) crashed in .*"
},
{
"type": "temporary",
"reason": "UncaughtException",
"pattern": "Process \\d+ \\(\\S+\\) of user \\d+ encountered an uncaught \\S+ exception"
},
{
"type": "temporary",
"reason": "XorgCrash",
"pattern": "Display server \\S+ crash in \\S+"
},
{
"type": "temporary",
"reason": "VMcore",
"pattern": "System encountered a fatal error in \\S+"
},
{
"type": "temporary",
"reason": "KernelOops",
"pattern": "System encountered a non-fatal error in \\S+"
}
]
"plugin": "journald",
"pluginConfig": {
"source": "abrt-notification"
},
"logPath": "/var/log/journal",
"lookback": "5m",
"bufferSize": 10,
"source": "abrt-adaptor",
"conditions": [],
"rules": [
{
"type": "temporary",
"reason": "CCPPCrash",
"pattern": "Process \\d+ \\(\\S+\\) crashed in .*"
},
{
"type": "temporary",
"reason": "UncaughtException",
"pattern": "Process \\d+ \\(\\S+\\) of user \\d+ encountered an uncaught \\S+ exception"
},
{
"type": "temporary",
"reason": "XorgCrash",
"pattern": "Display server \\S+ crash in \\S+"
},
{
"type": "temporary",
"reason": "VMcore",
"pattern": "System encountered a fatal error in \\S+"
},
{
"type": "temporary",
"reason": "KernelOops",
"pattern": "System encountered a non-fatal error in \\S+"
}
]
}
config/disk-log-message-filelog.json
+30 -27
View File
@@ -1,28 +1,31 @@
{
"plugin": "filelog",
"pluginConfig": {
"timestamp": "^.{15}",
"message": "(?i)Currently unreadable.*sectors|(?i)Offline uncorrectable sectors",
"timestampFormat": "Jan _2 15:04:05"
},
"logPath": "/var/log/messages",
"lookback": "10h",
"bufferSize": 1,
"source": "disk-monitor",
"skipList": [ " audit:", " audit[" ],
"conditions": [
{
"type": "DiskBadBlock",
"reason": "DiskBadBlock",
"message": "Disk no bad block"
},
],
"rules": [
{
"type": "permanent",
"condition": "DiskBadBlock",
"reason": "DiskBadBlock",
"pattern": ".*([1-9]\\d{2,}) (Currently unreadable.*sectors|Offline uncorrectable sectors).*"
},
]
}
"plugin": "filelog",
"pluginConfig": {
"timestamp": "^.{15}",
"message": "(?i)Currently unreadable.*sectors|(?i)Offline uncorrectable sectors",
"timestampFormat": "Jan _2 15:04:05"
},
"logPath": "/var/log/messages",
"lookback": "10h",
"bufferSize": 1,
"source": "disk-monitor",
"skipList": [
" audit:",
" audit["
],
"conditions": [
{
"type": "DiskBadBlock",
"reason": "DiskBadBlock",
"message": "Disk no bad block"
}
],
"rules": [
{
"type": "permanent",
"condition": "DiskBadBlock",
"reason": "DiskBadBlock",
"pattern": ".*([1-9]\\d{2,}) (Currently unreadable.*sectors|Offline uncorrectable sectors).*"
}
]
}
config/docker-monitor-filelog.json
+18 -18
View File
@@ -1,20 +1,20 @@
{
"plugin": "filelog",
"pluginConfig": {
"timestamp": "^time=\"(\\S*)\"",
"message": "msg=\"([^\n]*)\"",
"timestampFormat": "2006-01-02T15:04:05.999999999-07:00"
},
"logPath": "/var/log/docker.log",
"lookback": "5m",
"bufferSize": 10,
"source": "docker-monitor",
"conditions": [],
"rules": [
{
"type": "temporary",
"reason": "CorruptDockerImage",
"pattern": "Error trying v2 registry: failed to register layer: rename /var/lib/docker/image/(.+) /var/lib/docker/image/(.+): directory not empty.*"
}
]
"plugin": "filelog",
"pluginConfig": {
"timestamp": "^time=\"(\\S*)\"",
"message": "msg=\"([^\n]*)\"",
"timestampFormat": "2006-01-02T15:04:05.999999999-07:00"
},
"logPath": "/var/log/docker.log",
"lookback": "5m",
"bufferSize": 10,
"source": "docker-monitor",
"conditions": [],
"rules": [
{
"type": "temporary",
"reason": "CorruptDockerImage",
"pattern": "Error trying v2 registry: failed to register layer: rename /var/lib/docker/image/(.+) /var/lib/docker/image/(.+): directory not empty.*"
}
]
}
config/docker-monitor.json
+34 -34
View File
@@ -1,36 +1,36 @@
{
"plugin": "journald",
"pluginConfig": {
"source": "dockerd"
},
"logPath": "/var/log/journal",
"lookback": "5m",
"bufferSize": 10,
"source": "docker-monitor",
"metricsReporting": true,
"conditions": [
{
"type": "CorruptDockerOverlay2",
"reason": "NoCorruptDockerOverlay2",
"message": "docker overlay2 is functioning properly"
}
],
"rules": [
{
"type": "temporary",
"reason": "CorruptDockerImage",
"pattern": "Error trying v2 registry: failed to register layer: rename /var/lib/docker/image/(.+) /var/lib/docker/image/(.+): directory not empty.*"
},
{
"type": "permanent",
"condition": "CorruptDockerOverlay2",
"reason": "CorruptDockerOverlay2",
"pattern": "returned error: readlink /var/lib/docker/overlay2.*: invalid argument.*"
},
{
"type": "temporary",
"reason": "DockerContainerStartupFailure",
"pattern": "OCI runtime start failed: container process is already dead: unknown"
}
]
"plugin": "journald",
"pluginConfig": {
"source": "dockerd"
},
"logPath": "/var/log/journal",
"lookback": "5m",
"bufferSize": 10,
"source": "docker-monitor",
"metricsReporting": true,
"conditions": [
{
"type": "CorruptDockerOverlay2",
"reason": "NoCorruptDockerOverlay2",
"message": "docker overlay2 is functioning properly"
}
],
"rules": [
{
"type": "temporary",
"reason": "CorruptDockerImage",
"pattern": "Error trying v2 registry: failed to register layer: rename /var/lib/docker/image/(.+) /var/lib/docker/image/(.+): directory not empty.*"
},
{
"type": "permanent",
"condition": "CorruptDockerOverlay2",
"reason": "CorruptDockerOverlay2",
"pattern": "returned error: readlink /var/lib/docker/overlay2.*: invalid argument.*"
},
{
"type": "temporary",
"reason": "DockerContainerStartupFailure",
"pattern": "OCI runtime start failed: container process is already dead: unknown"
}
]
}
config/exporter/stackdriver-exporter.json
+6 -6
View File
@@ -1,8 +1,8 @@
{
"apiEndpoint": "monitoring.googleapis.com:443",
"exportPeriod": "60s",
"metadataFetchTimeout": "600s",
"metadataFetchInterval": "10s",
"panicOnMetadataFetchFailure": false,
"customMetricPrefix": ""
"apiEndpoint": "monitoring.googleapis.com:443",
"exportPeriod": "60s",
"metadataFetchTimeout": "600s",
"metadataFetchInterval": "10s",
"panicOnMetadataFetchFailure": false,
"customMetricPrefix": ""
}
config/guestosconfig/known-modules.json
+201 -67
View File
@@ -1,69 +1,203 @@
[
{ "moduleName": "xt_MASQUERADE"},
{ "moduleName": "xt_addrtype"},
{ "moduleName": "iptable_nat"},
{ "moduleName": "nf_nat"},
{ "moduleName": "br_netfilter"},
{ "moduleName": "ip6table_filter"},
{ "moduleName": "ip6_tables"},
{ "moduleName": "aesni_intel"},
{ "moduleName": "glue_helper"},
{ "moduleName": "crypto_simd"},
{ "moduleName": "cryptd"},
{ "moduleName": "virtio_balloon"},
{ "moduleName": "loadpin_trigger"},
{ "moduleName":"ip6table_filter"},
{ "moduleName":"ip6_tables"},
{ "moduleName":"iptable_filter"},
{ "moduleName":"bpfilter"},
{ "moduleName":"nls_iso8859_1"},
{ "moduleName":"intel_rapl_msr"},
{ "moduleName":"intel_rapl_common"},
{ "moduleName":"sb_edac"},
{ "moduleName":"rapl"},
{ "moduleName":"input_leds"},
{ "moduleName":"serio_raw"},
{ "moduleName":"pvpanic"},
{ "moduleName":"mac_hid"},
{ "moduleName":"sch_fq_codel"},
{ "moduleName":"ib_iser"},
{ "moduleName":"rdma_cm"},
{ "moduleName":"iw_cm"},
{ "moduleName":"ib_cm"},
{ "moduleName":"ib_core"},
{ "moduleName":"iscsi_tcp"},
{ "moduleName":"libiscsi_tcp"},
{ "moduleName":"libiscsi"},
{ "moduleName":"scsi_transport_iscsi"},
{ "moduleName":"virtio_rng"},
{ "moduleName":"ip_tables"},
{ "moduleName":"x_tables"},
{ "moduleName":"autofs4"},
{ "moduleName":"btrfs"},
{ "moduleName":"zstd_compress"},
{ "moduleName":"raid10"},
{ "moduleName":"raid456"},
{ "moduleName":"async_raid6_recov"},
{ "moduleName":"async_memcpy"},
{ "moduleName":"async_pq"},
{ "moduleName":"async_xor"},
{ "moduleName":"async_tx"},
{ "moduleName":"xor"},
{ "moduleName":"raid6_pq"},
{ "moduleName":"raid1"},
{ "moduleName":"raid0"},
{ "moduleName":"multipath"},
{ "moduleName":"linear"},
{ "moduleName":"crct10dif_pclmul"},
{ "moduleName":"crc32_pclmul"},
{ "moduleName":"ghash_clmulni_intel"},
{ "moduleName":"aesni_intel"},
{ "moduleName":"crypto_simd"},
{ "moduleName":"cryptd"},
{ "moduleName":"glue_helper"},
{ "moduleName":"psmouse"},
{ "moduleName":"virtio_net"},
{ "moduleName":"net_failover"},
{ "moduleName": "failover"},
{ "moduleName":"i2c_piix4"}
{
"moduleName": "xt_MASQUERADE"
},
{
"moduleName": "xt_addrtype"
},
{
"moduleName": "iptable_nat"
},
{
"moduleName": "nf_nat"
},
{
"moduleName": "br_netfilter"
},
{
"moduleName": "ip6table_filter"
},
{
"moduleName": "ip6_tables"
},
{
"moduleName": "aesni_intel"
},
{
"moduleName": "glue_helper"
},
{
"moduleName": "crypto_simd"
},
{
"moduleName": "cryptd"
},
{
"moduleName": "virtio_balloon"
},
{
"moduleName": "loadpin_trigger"
},
{
"moduleName": "ip6table_filter"
},
{
"moduleName": "ip6_tables"
},
{
"moduleName": "iptable_filter"
},
{
"moduleName": "bpfilter"
},
{
"moduleName": "nls_iso8859_1"
},
{
"moduleName": "intel_rapl_msr"
},
{
"moduleName": "intel_rapl_common"
},
{
"moduleName": "sb_edac"
},
{
"moduleName": "rapl"
},
{
"moduleName": "input_leds"
},
{
"moduleName": "serio_raw"
},
{
"moduleName": "pvpanic"
},
{
"moduleName": "mac_hid"
},
{
"moduleName": "sch_fq_codel"
},
{
"moduleName": "ib_iser"
},
{
"moduleName": "rdma_cm"
},
{
"moduleName": "iw_cm"
},
{
"moduleName": "ib_cm"
},
{
"moduleName": "ib_core"
},
{
"moduleName": "iscsi_tcp"
},
{
"moduleName": "libiscsi_tcp"
},
{
"moduleName": "libiscsi"
},
{
"moduleName": "scsi_transport_iscsi"
},
{
"moduleName": "virtio_rng"
},
{
"moduleName": "ip_tables"
},
{
"moduleName": "x_tables"
},
{
"moduleName": "autofs4"
},
{
"moduleName": "btrfs"
},
{
"moduleName": "zstd_compress"
},
{
"moduleName": "raid10"
},
{
"moduleName": "raid456"
},
{
"moduleName": "async_raid6_recov"
},
{
"moduleName": "async_memcpy"
},
{
"moduleName": "async_pq"
},
{
"moduleName": "async_xor"
},
{
"moduleName": "async_tx"
},
{
"moduleName": "xor"
},
{
"moduleName": "raid6_pq"
},
{
"moduleName": "raid1"
},
{
"moduleName": "raid0"
},
{
"moduleName": "multipath"
},
{
"moduleName": "linear"
},
{
"moduleName": "crct10dif_pclmul"
},
{
"moduleName": "crc32_pclmul"
},
{
"moduleName": "ghash_clmulni_intel"
},
{
"moduleName": "aesni_intel"
},
{
"moduleName": "crypto_simd"
},
{
"moduleName": "cryptd"
},
{
"moduleName": "glue_helper"
},
{
"moduleName": "psmouse"
},
{
"moduleName": "virtio_net"
},
{
"moduleName": "net_failover"
},
{
"moduleName": "failover"
},
{
"moduleName": "i2c_piix4"
}
]
config/kernel-monitor-filelog.json
+50 -50
View File
@@ -1,52 +1,52 @@
{
"plugin": "filelog",
"pluginConfig": {
"timestamp": "^.{15}",
"message": "kernel: \\[.*\\] (.*)",
"timestampFormat": "Jan _2 15:04:05"
},
"logPath": "/var/log/kern.log",
"lookback": "5m",
"bufferSize": 10,
"source": "kernel-monitor",
"conditions": [
{
"type": "KernelDeadlock",
"reason": "KernelHasNoDeadlock",
"message": "kernel has no deadlock"
}
],
"rules": [
{
"type": "temporary",
"reason": "OOMKilling",
"pattern": "Killed process \\d+ (.+) total-vm:\\d+kB, anon-rss:\\d+kB, file-rss:\\d+kB.*"
},
{
"type": "temporary",
"reason": "TaskHung",
"pattern": "task \\S+:\\w+ blocked for more than \\w+ seconds\\."
},
{
"type": "temporary",
"reason": "UnregisterNetDevice",
"pattern": "unregister_netdevice: waiting for \\w+ to become free. Usage count = \\d+"
},
{
"type": "temporary",
"reason": "KernelOops",
"pattern": "BUG: unable to handle kernel NULL pointer dereference at .*"
},
{
"type": "temporary",
"reason": "KernelOops",
"pattern": "divide error: 0000 \\[#\\d+\\] SMP"
},
{
"type": "permanent",
"condition": "KernelDeadlock",
"reason": "DockerHung",
"pattern": "task docker:\\w+ blocked for more than \\w+ seconds\\."
}
]
"plugin": "filelog",
"pluginConfig": {
"timestamp": "^.{15}",
"message": "kernel: \\[.*\\] (.*)",
"timestampFormat": "Jan _2 15:04:05"
},
"logPath": "/var/log/kern.log",
"lookback": "5m",
"bufferSize": 10,
"source": "kernel-monitor",
"conditions": [
{
"type": "KernelDeadlock",
"reason": "KernelHasNoDeadlock",
"message": "kernel has no deadlock"
}
],
"rules": [
{
"type": "temporary",
"reason": "OOMKilling",
"pattern": "Killed process \\d+ (.+) total-vm:\\d+kB, anon-rss:\\d+kB, file-rss:\\d+kB.*"
},
{
"type": "temporary",
"reason": "TaskHung",
"pattern": "task \\S+:\\w+ blocked for more than \\w+ seconds\\."
},
{
"type": "temporary",
"reason": "UnregisterNetDevice",
"pattern": "unregister_netdevice: waiting for \\w+ to become free. Usage count = \\d+"
},
{
"type": "temporary",
"reason": "KernelOops",
"pattern": "BUG: unable to handle kernel NULL pointer dereference at .*"
},
{
"type": "temporary",
"reason": "KernelOops",
"pattern": "divide error: 0000 \\[#\\d+\\] SMP"
},
{
"type": "permanent",
"condition": "KernelDeadlock",
"reason": "DockerHung",
"pattern": "task docker:\\w+ blocked for more than \\w+ seconds\\."
}
]
}
config/kernel-monitor.json
+98 -98
View File
@@ -1,100 +1,100 @@
{
"plugin": "kmsg",
"logPath": "/dev/kmsg",
"lookback": "5m",
"bufferSize": 10,
"source": "kernel-monitor",
"metricsReporting": true,
"conditions": [
{
"type": "KernelDeadlock",
"reason": "KernelHasNoDeadlock",
"message": "kernel has no deadlock"
},
{
"type": "XfsShutdown",
"reason": "XfsHasNotShutDown",
"message": "XFS has not shutdown"
},
{
"type": "CperHardwareErrorFatal",
"reason": "CperHardwareHasNoFatalError",
"message": "UEFI CPER has no fatal error"
}
],
"rules": [
{
"type": "temporary",
"reason": "OOMKilling",
"pattern": "Killed process \\d+ (.+) total-vm:\\d+kB, anon-rss:\\d+kB, file-rss:\\d+kB.*"
},
{
"type": "temporary",
"reason": "TaskHung",
"pattern": "task [\\S ]+:\\w+ blocked for more than \\w+ seconds\\."
},
{
"type": "temporary",
"reason": "UnregisterNetDevice",
"pattern": "unregister_netdevice: waiting for \\w+ to become free. Usage count = \\d+"
},
{
"type": "temporary",
"reason": "KernelOops",
"pattern": "BUG: unable to handle kernel NULL pointer dereference at .*"
},
{
"type": "temporary",
"reason": "KernelOops",
"pattern": "divide error: 0000 \\[#\\d+\\] SMP"
},
{
"type": "temporary",
"reason": "Ext4Error",
"pattern": "EXT4-fs error .*"
},
{
"type": "temporary",
"reason": "Ext4Warning",
"pattern": "EXT4-fs warning .*"
},
{
"type": "temporary",
"reason": "IOError",
"pattern": "Buffer I/O error .*"
},
{
"type": "permanent",
"condition": "XfsShutdown",
"reason": "XfsHasShutdown",
"pattern": "XFS .* Shutting down filesystem.?"
},
{
"type": "temporary",
"reason": "MemoryReadError",
"pattern": "CE memory read error .*"
},
{
"type": "temporary",
"reason": "CperHardwareErrorCorrected",
"pattern": ".*\\[Hardware Error\\]: event severity: corrected$"
},
{
"type": "temporary",
"reason": "CperHardwareErrorRecoverable",
"pattern": ".*\\[Hardware Error\\]: event severity: recoverable$"
},
{
"type": "permanent",
"condition": "CperHardwareErrorFatal",
"reason": "CperHardwareErrorFatal",
"pattern": ".*\\[Hardware Error\\]: event severity: fatal$"
},
{
"type": "permanent",
"condition": "KernelDeadlock",
"reason": "DockerHung",
"pattern": "task docker:\\w+ blocked for more than \\w+ seconds\\."
}
]
"plugin": "kmsg",
"logPath": "/dev/kmsg",
"lookback": "5m",
"bufferSize": 10,
"source": "kernel-monitor",
"metricsReporting": true,
"conditions": [
{
"type": "KernelDeadlock",
"reason": "KernelHasNoDeadlock",
"message": "kernel has no deadlock"
},
{
"type": "XfsShutdown",
"reason": "XfsHasNotShutDown",
"message": "XFS has not shutdown"
},
{
"type": "CperHardwareErrorFatal",
"reason": "CperHardwareHasNoFatalError",
"message": "UEFI CPER has no fatal error"
}
],
"rules": [
{
"type": "temporary",
"reason": "OOMKilling",
"pattern": "Killed process \\d+ (.+) total-vm:\\d+kB, anon-rss:\\d+kB, file-rss:\\d+kB.*"
},
{
"type": "temporary",
"reason": "TaskHung",
"pattern": "task [\\S ]+:\\w+ blocked for more than \\w+ seconds\\."
},
{
"type": "temporary",
"reason": "UnregisterNetDevice",
"pattern": "unregister_netdevice: waiting for \\w+ to become free. Usage count = \\d+"
},
{
"type": "temporary",
"reason": "KernelOops",
"pattern": "BUG: unable to handle kernel NULL pointer dereference at .*"
},
{
"type": "temporary",
"reason": "KernelOops",
"pattern": "divide error: 0000 \\[#\\d+\\] SMP"
},
{
"type": "temporary",
"reason": "Ext4Error",
"pattern": "EXT4-fs error .*"
},
{
"type": "temporary",
"reason": "Ext4Warning",
"pattern": "EXT4-fs warning .*"
},
{
"type": "temporary",
"reason": "IOError",
"pattern": "Buffer I/O error .*"
},
{
"type": "permanent",
"condition": "XfsShutdown",
"reason": "XfsHasShutdown",
"pattern": "XFS .* Shutting down filesystem.?"
},
{
"type": "temporary",
"reason": "MemoryReadError",
"pattern": "CE memory read error .*"
},
{
"type": "temporary",
"reason": "CperHardwareErrorCorrected",
"pattern": ".*\\[Hardware Error\\]: event severity: corrected$"
},
{
"type": "temporary",
"reason": "CperHardwareErrorRecoverable",
"pattern": ".*\\[Hardware Error\\]: event severity: recoverable$"
},
{
"type": "permanent",
"condition": "CperHardwareErrorFatal",
"reason": "CperHardwareErrorFatal",
"pattern": ".*\\[Hardware Error\\]: event severity: fatal$"
},
{
"type": "permanent",
"condition": "KernelDeadlock",
"reason": "DockerHung",
"pattern": "task docker:\\w+ blocked for more than \\w+ seconds\\."
}
]
}
config/readonly-monitor.json
+21 -21
View File
@@ -1,23 +1,23 @@
{
"plugin": "kmsg",
"logPath": "/dev/kmsg",
"lookback": "5m",
"bufferSize": 10,
"source": "readonly-monitor",
"metricsReporting": true,
"conditions": [
{
"type": "ReadonlyFilesystem",
"reason": "FilesystemIsNotReadOnly",
"message": "Filesystem is not read-only"
}
],
"rules": [
{
"type": "permanent",
"condition": "ReadonlyFilesystem",
"reason": "FilesystemIsReadOnly",
"pattern": "Remounting filesystem read-only"
}
]
"plugin": "kmsg",
"logPath": "/dev/kmsg",
"lookback": "5m",
"bufferSize": 10,
"source": "readonly-monitor",
"metricsReporting": true,
"conditions": [
{
"type": "ReadonlyFilesystem",
"reason": "FilesystemIsNotReadOnly",
"message": "Filesystem is not read-only"
}
],
"rules": [
{
"type": "permanent",
"condition": "ReadonlyFilesystem",
"reason": "FilesystemIsReadOnly",
"pattern": "Remounting filesystem read-only"
}
]
}
config/systemd-monitor.json
+27 -27
View File
@@ -1,29 +1,29 @@
{
"plugin": "journald",
"pluginConfig": {
"source": "systemd"
},
"logPath": "/var/log/journal",
"lookback": "5m",
"bufferSize": 10,
"source": "systemd-monitor",
"metricsReporting": true,
"conditions": [],
"rules": [
{
"type": "temporary",
"reason": "KubeletStart",
"pattern": "Started (Kubernetes kubelet|kubelet.service|kubelet.service - Kubernetes kubelet|kubelet.service - .*)."
},
{
"type": "temporary",
"reason": "DockerStart",
"pattern": "Starting (Docker Application Container Engine|docker.service|docker.service - Docker Application Container Engine)..."
},
{
"type": "temporary",
"reason": "ContainerdStart",
"pattern": "Starting (containerd container runtime|containerd.service|containerd.service - containerd container runtime)..."
}
]
"plugin": "journald",
"pluginConfig": {
"source": "systemd"
},
"logPath": "/var/log/journal",
"lookback": "5m",
"bufferSize": 10,
"source": "systemd-monitor",
"metricsReporting": true,
"conditions": [],
"rules": [
{
"type": "temporary",
"reason": "KubeletStart",
"pattern": "Started (Kubernetes kubelet|kubelet.service|kubelet.service - Kubernetes kubelet|kubelet.service - .*)."
},
{
"type": "temporary",
"reason": "DockerStart",
"pattern": "Starting (Docker Application Container Engine|docker.service|docker.service - Docker Application Container Engine)..."
},
{
"type": "temporary",
"reason": "ContainerdStart",
"pattern": "Starting (containerd container runtime|containerd.service|containerd.service - containerd container runtime)..."
}
]
}
config/windows-containerd-monitor-filelog.json
+28 -28
View File
@@ -1,30 +1,30 @@
{
"plugin": "filelog",
"pluginConfig": {
"timestamp": "^time=\"(\\S*)\"",
"message": "msg=\"?([^\n]*)\"?",
"timestampFormat": "2006-01-02T15:04:05.999999999Z"
},
"logPath": "C:\\etc\\kubernetes\\logs\\containerd.log",
"lookback": "5m",
"bufferSize": 10,
"source": "containerd",
"conditions": [],
"rules": [
{
"type": "temporary",
"reason": "ContainerCreationFailed",
"pattern": ".*failed to create containerd container.*error unpacking image.*wrong diff id calculated on extraction.*"
},
{
"type": "temporary",
"reason": "CorruptContainerImageLayer",
"pattern": ".*failed to pull and unpack image.*failed to extract layer.*archive/tar: invalid tar header.*"
},
{
"type": "temporary",
"reason": "HCSEmptyLayerchain",
"pattern": ".*Failed to unmarshall layerchain json - invalid character '\\x00' looking for beginning of value*"
}
]
"plugin": "filelog",
"pluginConfig": {
"timestamp": "^time=\"(\\S*)\"",
"message": "msg=\"?([^\n]*)\"?",
"timestampFormat": "2006-01-02T15:04:05.999999999Z"
},
"logPath": "C:\\etc\\kubernetes\\logs\\containerd.log",
"lookback": "5m",
"bufferSize": 10,
"source": "containerd",
"conditions": [],
"rules": [
{
"type": "temporary",
"reason": "ContainerCreationFailed",
"pattern": ".*failed to create containerd container.*error unpacking image.*wrong diff id calculated on extraction.*"
},
{
"type": "temporary",
"reason": "CorruptContainerImageLayer",
"pattern": ".*failed to pull and unpack image.*failed to extract layer.*archive/tar: invalid tar header.*"
},
{
"type": "temporary",
"reason": "HCSEmptyLayerchain",
"pattern": ".*Failed to unmarshall layerchain json - invalid character '\\x00' looking for beginning of value*"
}
]
}
config/windows-defender-monitor.json
+19 -20
View File
@@ -1,21 +1,20 @@
{
"plugin": "custom",
"pluginConfig": {
"invoke_interval": "10m",
"timeout": "5s",
"max_output_length": 80,
"concurrency": 3
},
"source": "windows-defender-custom-plugin-monitor",
"metricsReporting": true,
"conditions": [],
"rules": [
{
"type": "temporary",
"reason": "WindowsDefenderThreatsDetected",
"path": "C:\\etc\\kubernetes\\node-problem-detector\\config\\plugin\\windows_defender_problem.ps1",
"timeout": "3s"
}
]
}
"plugin": "custom",
"pluginConfig": {
"invoke_interval": "10m",
"timeout": "5s",
"max_output_length": 80,
"concurrency": 3
},
"source": "windows-defender-custom-plugin-monitor",
"metricsReporting": true,
"conditions": [],
"rules": [
{
"type": "temporary",
"reason": "WindowsDefenderThreatsDetected",
"path": "C:\\etc\\kubernetes\\node-problem-detector\\config\\plugin\\windows_defender_problem.ps1",
"timeout": "3s"
}
]
}
config/windows-health-checker-containerd.json
+33 -33
View File
@@ -1,34 +1,34 @@
{
"plugin": "custom",
"pluginConfig": {
"invoke_interval": "10s",
"timeout": "3m",
"max_output_length": 80,
"concurrency": 1
},
"source": "health-checker",
"metricsReporting": true,
"conditions": [
{
"type": "ContainerRuntimeUnhealthy",
"reason": "ContainerRuntimeIsHealthy",
"message": "Container runtime on the node is functioning properly"
}
],
"rules": [
{
"type": "permanent",
"condition": "ContainerRuntimeUnhealthy",
"reason": "ContainerdUnhealthy",
"path": "C:\\etc\\kubernetes\\node\\bin\\health-checker.exe",
"args": [
"--component=cri",
"--enable-repair=true",
"--cooldown-time=2m",
"--cri-timeout=10s",
"--health-check-timeout=60s"
],
"timeout": "3m"
}
]
}
"plugin": "custom",
"pluginConfig": {
"invoke_interval": "10s",
"timeout": "3m",
"max_output_length": 80,
"concurrency": 1
},
"source": "health-checker",
"metricsReporting": true,
"conditions": [
{
"type": "ContainerRuntimeUnhealthy",
"reason": "ContainerRuntimeIsHealthy",
"message": "Container runtime on the node is functioning properly"
}
],
"rules": [
{
"type": "permanent",
"condition": "ContainerRuntimeUnhealthy",
"reason": "ContainerdUnhealthy",
"path": "C:\\etc\\kubernetes\\node\\bin\\health-checker.exe",
"args": [
"--component=cri",
"--enable-repair=true",
"--cooldown-time=2m",
"--cri-timeout=10s",
"--health-check-timeout=60s"
],
"timeout": "3m"
}
]
}
config/windows-health-checker-docker.json
+32 -33
View File
@@ -1,34 +1,33 @@
{
"plugin": "custom",
"pluginConfig": {
"invoke_interval": "10s",
"timeout": "3m",
"max_output_length": 80,
"concurrency": 1
},
"source": "health-checker",
"metricsReporting": true,
"conditions": [
{
"type": "ContainerRuntimeUnhealthy",
"reason": "ContainerRuntimeIsHealthy",
"message": "Container runtime on the node is functioning properly"
}
],
"rules": [
{
"type": "permanent",
"condition": "ContainerRuntimeUnhealthy",
"reason": "DockerUnhealthy",
"path": "C:\\etc\\kubernetes\\node\\bin\\health-checker.exe",
"args": [
"--component=docker",
"--enable-repair=true",
"--cooldown-time=2m",
"--health-check-timeout=60s"
],
"timeout": "3m"
}
]
}
"plugin": "custom",
"pluginConfig": {
"invoke_interval": "10s",
"timeout": "3m",
"max_output_length": 80,
"concurrency": 1
},
"source": "health-checker",
"metricsReporting": true,
"conditions": [
{
"type": "ContainerRuntimeUnhealthy",
"reason": "ContainerRuntimeIsHealthy",
"message": "Container runtime on the node is functioning properly"
}
],
"rules": [
{
"type": "permanent",
"condition": "ContainerRuntimeUnhealthy",
"reason": "DockerUnhealthy",
"path": "C:\\etc\\kubernetes\\node\\bin\\health-checker.exe",
"args": [
"--component=docker",
"--enable-repair=true",
"--cooldown-time=2m",
"--health-check-timeout=60s"
],
"timeout": "3m"
}
]
}
config/windows-health-checker-kubelet.json
+32 -33
View File
@@ -1,34 +1,33 @@
{
"plugin": "custom",
"pluginConfig": {
"invoke_interval": "10s",
"timeout": "3m",
"max_output_length": 80,
"concurrency": 1
},
"source": "health-checker",
"metricsReporting": true,
"conditions": [
{
"type": "KubeletUnhealthy",
"reason": "KubeletIsHealthy",
"message": "kubelet on the node is functioning properly"
}
],
"rules": [
{
"type": "permanent",
"condition": "KubeletUnhealthy",
"reason": "KubeletUnhealthy",
"path": "C:\\etc\\kubernetes\\node\\bin\\health-checker.exe",
"args": [
"--component=kubelet",
"--enable-repair=true",
"--cooldown-time=1m",
"--health-check-timeout=10s"
],
"timeout": "3m"
}
]
}
"plugin": "custom",
"pluginConfig": {
"invoke_interval": "10s",
"timeout": "3m",
"max_output_length": 80,
"concurrency": 1
},
"source": "health-checker",
"metricsReporting": true,
"conditions": [
{
"type": "KubeletUnhealthy",
"reason": "KubeletIsHealthy",
"message": "kubelet on the node is functioning properly"
}
],
"rules": [
{
"type": "permanent",
"condition": "KubeletUnhealthy",
"reason": "KubeletUnhealthy",
"path": "C:\\etc\\kubernetes\\node\\bin\\health-checker.exe",
"args": [
"--component=kubelet",
"--enable-repair=true",
"--cooldown-time=1m",
"--health-check-timeout=10s"
],
"timeout": "3m"
}
]
}
config/windows-health-checker-kubeproxy.json
+32 -33
View File
@@ -1,34 +1,33 @@
{
"plugin": "custom",
"pluginConfig": {
"invoke_interval": "10s",
"timeout": "3m",
"max_output_length": 80,
"concurrency": 1
},
"source": "health-checker",
"metricsReporting": true,
"conditions": [
{
"type": "KubeProxyUnhealthy",
"reason": "KubeProxyIsHealthy",
"message": "kube-proxy on the node is functioning properly"
}
],
"rules": [
{
"type": "permanent",
"condition": "KubeProxyUnhealthy",
"reason": "KubeProxyUnhealthy",
"path": "C:\\etc\\kubernetes\\node\\bin\\health-checker.exe",
"args": [
"--component=kube-proxy",
"--enable-repair=true",
"--cooldown-time=1m",
"--health-check-timeout=10s"
],
"timeout": "3m"
}
]
}
"plugin": "custom",
"pluginConfig": {
"invoke_interval": "10s",
"timeout": "3m",
"max_output_length": 80,
"concurrency": 1
},
"source": "health-checker",
"metricsReporting": true,
"conditions": [
{
"type": "KubeProxyUnhealthy",
"reason": "KubeProxyIsHealthy",
"message": "kube-proxy on the node is functioning properly"
}
],
"rules": [
{
"type": "permanent",
"condition": "KubeProxyUnhealthy",
"reason": "KubeProxyUnhealthy",
"path": "C:\\etc\\kubernetes\\node\\bin\\health-checker.exe",
"args": [
"--component=kube-proxy",
"--enable-repair=true",
"--cooldown-time=1m",
"--health-check-timeout=10s"
],
"timeout": "3m"
}
]
}