mirror of
https://github.com/kmein/niveum
synced 2026-03-16 10:11:08 +01:00
alertmanager: alert lassulus for mastodon outtime
This commit is contained in:
@@ -3,11 +3,13 @@
|
||||
config,
|
||||
pkgs,
|
||||
...
|
||||
}: let
|
||||
}:
|
||||
let
|
||||
lokiConfig = import ./loki.nix;
|
||||
blackboxConfig = import ./blackbox.nix;
|
||||
inherit (import ../../../lib) restic;
|
||||
in {
|
||||
in
|
||||
{
|
||||
services.grafana = {
|
||||
enable = true;
|
||||
settings = {
|
||||
@@ -80,143 +82,150 @@ in {
|
||||
}
|
||||
];
|
||||
|
||||
services.prometheus.rules = let
|
||||
diskFreeThreshold = 10;
|
||||
in [
|
||||
(builtins.toJSON {
|
||||
groups = [
|
||||
{
|
||||
name = "niveum";
|
||||
rules = [
|
||||
{
|
||||
alert = "HostSystemdServiceCrashed";
|
||||
expr = ''(node_systemd_unit_state{state="failed"} == 1) * on(instance) group_left (nodename) node_uname_info{nodename=~".+"}'';
|
||||
annotations = {
|
||||
description = "{{$labels.name}} failed on {{$labels.instance}}";
|
||||
};
|
||||
}
|
||||
{
|
||||
alert = "RootPartitionFull";
|
||||
for = "10m";
|
||||
expr = ''(node_filesystem_free_bytes{mountpoint="/"} * 100) / node_filesystem_size_bytes{mountpoint="/"} < ${toString diskFreeThreshold}'';
|
||||
annotations = {
|
||||
description = ''{{ $labels.instance }} running out of space: {{ $value | printf "%.2f" }}% < ${toString diskFreeThreshold}%'';
|
||||
};
|
||||
}
|
||||
{
|
||||
alert = "RootPartitionFullWeek";
|
||||
for = "1h";
|
||||
expr =
|
||||
''node_filesystem_free_bytes{mountpoint="/"} ''
|
||||
+ ''and predict_linear(node_filesystem_free_bytes{mountpoint="/"}[2d], 7*24*3600) <= 0'';
|
||||
annotations = {
|
||||
description = "{{$labels.instance}} running out of space in 7 days";
|
||||
};
|
||||
}
|
||||
{
|
||||
alert = "HighLoad";
|
||||
expr = ''node_load15 / on(job) count(node_cpu_seconds_total{mode="system"}) by (job) >= 1.0'';
|
||||
for = "10m";
|
||||
annotations = {
|
||||
description = "{{$labels.instance}} running on high load: {{$value}}";
|
||||
};
|
||||
}
|
||||
{
|
||||
alert = "HostUnusualNetworkThroughputIn";
|
||||
expr = ''(rate(node_network_receive_bytes_total[2m])) / 1024 / 1024 > 100'';
|
||||
for = "5m";
|
||||
annotations.description = "Host unusual network throughput in (instance {{ $labels.instance }})";
|
||||
}
|
||||
{
|
||||
alert = "HostUnusualNetworkThroughputOut";
|
||||
expr = ''(rate(node_network_transmit_bytes_total[2m])) / 1024 / 1024 > 100'';
|
||||
for = "5m";
|
||||
annotations.description = "Host unusual network throughput out (instance {{ $labels.instance }})";
|
||||
}
|
||||
{
|
||||
alert = "HostUnusualDiskReadRate";
|
||||
expr = ''(rate(node_disk_read_bytes_total[2m])) / 1024 / 1024 > 50'';
|
||||
for = "5m";
|
||||
annotations.description = "Host unusual disk read rate (instance {{ $labels.instance }})";
|
||||
}
|
||||
{
|
||||
alert = "HostUnusualDiskWriteRate";
|
||||
expr = ''(rate(node_disk_written_bytes_total[2m])) / 1024 / 1024 > 50'';
|
||||
for = "2m";
|
||||
annotations.description = "Host unusual disk write rate (instance {{ $labels.instance }})";
|
||||
}
|
||||
{
|
||||
alert = "HostOutOfInodes";
|
||||
expr = ''node_filesystem_files_free{fstype!="msdosfs"} / node_filesystem_files{fstype!="msdosfs"} * 100 < 10 and ON (instance, device, mountpoint) node_filesystem_readonly == 0'';
|
||||
for = "2m";
|
||||
annotations.description = "Host out of inodes (instance {{ $labels.instance }})";
|
||||
}
|
||||
{
|
||||
alert = "HostInodesWillFillIn24Hours";
|
||||
expr = ''node_filesystem_files_free{fstype!="msdosfs"} / node_filesystem_files{fstype!="msdosfs"} * 100 < 10 and predict_linear(node_filesystem_files_free{fstype!="msdosfs"}[1h], 24 * 3600) < 0 and ON (instance, device, mountpoint) node_filesystem_readonly{fstype!="msdosfs"} == 0'';
|
||||
for = "2m";
|
||||
annotations.description = "Host inodes will fill in 24 hours (instance {{ $labels.instance }})";
|
||||
}
|
||||
{
|
||||
alert = "HighRAM";
|
||||
expr = "node_memory_MemFree_bytes + node_memory_Buffers_bytes + node_memory_Cached_bytes < node_memory_MemTotal_bytes * 0.1";
|
||||
for = "1h";
|
||||
annotations.description = "{{$labels.instance}} using lots of RAM";
|
||||
}
|
||||
{
|
||||
alert = "UptimeMonster";
|
||||
expr = "time() - node_boot_time_seconds > 2592000";
|
||||
annotations.description = "uptime monster {{$labels.instance}} up for more than 30 days";
|
||||
}
|
||||
{
|
||||
alert = "HostDown";
|
||||
expr = ''up == 0'';
|
||||
for = "5m";
|
||||
annotations = {
|
||||
description = "{{ $labels.instance }} seeming down since 5 minutes";
|
||||
};
|
||||
}
|
||||
{
|
||||
alert = "Reboot";
|
||||
expr = "time() - node_boot_time_seconds < 300";
|
||||
annotations.description = "{{$labels.instance}} rebooted";
|
||||
}
|
||||
{
|
||||
alert = "ProbeFailed";
|
||||
expr = "probe_success == 0";
|
||||
for = "5m";
|
||||
annotations.description = "HTTP probe failed for {{$labels.instance}}";
|
||||
}
|
||||
{
|
||||
alert = "SlowProbe";
|
||||
expr = "avg_over_time(probe_http_duration_seconds[1m]) > 1";
|
||||
for = "5m";
|
||||
annotations.description = "HTTP probe slow for {{$labels.instance}}";
|
||||
}
|
||||
{
|
||||
alert = "HttpStatusCode";
|
||||
expr = "probe_http_status_code != 0 AND (probe_http_status_code <= 199 OR probe_http_status_code >= 400)";
|
||||
for = "5m";
|
||||
annotations.description = "status code {{$value}} for {{$labels.instance}}";
|
||||
}
|
||||
{
|
||||
alert = "SslExpirySoon";
|
||||
expr = "probe_ssl_earliest_cert_expiry - time() < 86400 * 30";
|
||||
for = "5m";
|
||||
annotations.description = "SSL certificate for {{$labels.instance}} expires in 30 days";
|
||||
}
|
||||
{
|
||||
alert = "SslExpiry";
|
||||
expr = "probe_ssl_earliest_cert_expiry - time() <= 0";
|
||||
for = "5m";
|
||||
annotations.description = "SSL certificate for {{$labels.instance}} has expired";
|
||||
}
|
||||
];
|
||||
}
|
||||
];
|
||||
})
|
||||
];
|
||||
|
||||
services.prometheus.rules =
|
||||
let
|
||||
diskFreeThreshold = 10;
|
||||
in
|
||||
[
|
||||
(builtins.toJSON {
|
||||
groups = [
|
||||
{
|
||||
name = "niveum";
|
||||
rules = [
|
||||
{
|
||||
alert = "HostSystemdServiceCrashed";
|
||||
expr = ''(node_systemd_unit_state{state="failed"} == 1) * on(instance) group_left (nodename) node_uname_info{nodename=~".+"}'';
|
||||
annotations = {
|
||||
description = "{{$labels.name}} failed on {{$labels.instance}}";
|
||||
};
|
||||
}
|
||||
{
|
||||
alert = "RootPartitionFull";
|
||||
for = "10m";
|
||||
expr = ''(node_filesystem_free_bytes{mountpoint="/"} * 100) / node_filesystem_size_bytes{mountpoint="/"} < ${toString diskFreeThreshold}'';
|
||||
annotations = {
|
||||
description = ''{{ $labels.instance }} running out of space: {{ $value | printf "%.2f" }}% < ${toString diskFreeThreshold}%'';
|
||||
};
|
||||
}
|
||||
{
|
||||
alert = "RootPartitionFullWeek";
|
||||
for = "1h";
|
||||
expr =
|
||||
''node_filesystem_free_bytes{mountpoint="/"} ''
|
||||
+ ''and predict_linear(node_filesystem_free_bytes{mountpoint="/"}[2d], 7*24*3600) <= 0'';
|
||||
annotations = {
|
||||
description = "{{$labels.instance}} running out of space in 7 days";
|
||||
};
|
||||
}
|
||||
{
|
||||
alert = "HighLoad";
|
||||
expr = ''node_load15 / on(job) count(node_cpu_seconds_total{mode="system"}) by (job) >= 1.0'';
|
||||
for = "10m";
|
||||
annotations = {
|
||||
description = "{{$labels.instance}} running on high load: {{$value}}";
|
||||
};
|
||||
}
|
||||
{
|
||||
alert = "HostUnusualNetworkThroughputIn";
|
||||
expr = ''(rate(node_network_receive_bytes_total[2m])) / 1024 / 1024 > 100'';
|
||||
for = "5m";
|
||||
annotations.description = "Host unusual network throughput in (instance {{ $labels.instance }})";
|
||||
}
|
||||
{
|
||||
alert = "HostUnusualNetworkThroughputOut";
|
||||
expr = ''(rate(node_network_transmit_bytes_total[2m])) / 1024 / 1024 > 100'';
|
||||
for = "5m";
|
||||
annotations.description = "Host unusual network throughput out (instance {{ $labels.instance }})";
|
||||
}
|
||||
{
|
||||
alert = "HostUnusualDiskReadRate";
|
||||
expr = ''(rate(node_disk_read_bytes_total[2m])) / 1024 / 1024 > 50'';
|
||||
for = "5m";
|
||||
annotations.description = "Host unusual disk read rate (instance {{ $labels.instance }})";
|
||||
}
|
||||
{
|
||||
alert = "HostUnusualDiskWriteRate";
|
||||
expr = ''(rate(node_disk_written_bytes_total[2m])) / 1024 / 1024 > 50'';
|
||||
for = "2m";
|
||||
annotations.description = "Host unusual disk write rate (instance {{ $labels.instance }})";
|
||||
}
|
||||
{
|
||||
alert = "HostOutOfInodes";
|
||||
expr = ''node_filesystem_files_free{fstype!="msdosfs"} / node_filesystem_files{fstype!="msdosfs"} * 100 < 10 and ON (instance, device, mountpoint) node_filesystem_readonly == 0'';
|
||||
for = "2m";
|
||||
annotations.description = "Host out of inodes (instance {{ $labels.instance }})";
|
||||
}
|
||||
{
|
||||
alert = "HostInodesWillFillIn24Hours";
|
||||
expr = ''node_filesystem_files_free{fstype!="msdosfs"} / node_filesystem_files{fstype!="msdosfs"} * 100 < 10 and predict_linear(node_filesystem_files_free{fstype!="msdosfs"}[1h], 24 * 3600) < 0 and ON (instance, device, mountpoint) node_filesystem_readonly{fstype!="msdosfs"} == 0'';
|
||||
for = "2m";
|
||||
annotations.description = "Host inodes will fill in 24 hours (instance {{ $labels.instance }})";
|
||||
}
|
||||
{
|
||||
alert = "HighRAM";
|
||||
expr = "node_memory_MemFree_bytes + node_memory_Buffers_bytes + node_memory_Cached_bytes < node_memory_MemTotal_bytes * 0.1";
|
||||
for = "1h";
|
||||
annotations.description = "{{$labels.instance}} using lots of RAM";
|
||||
}
|
||||
{
|
||||
alert = "UptimeMonster";
|
||||
expr = "time() - node_boot_time_seconds > 2592000";
|
||||
annotations.description = "uptime monster {{$labels.instance}} up for more than 30 days";
|
||||
}
|
||||
{
|
||||
alert = "HostDown";
|
||||
expr = ''up == 0'';
|
||||
for = "5m";
|
||||
annotations = {
|
||||
description = "{{ $labels.instance }} seeming down since 5 minutes";
|
||||
};
|
||||
}
|
||||
{
|
||||
alert = "Reboot";
|
||||
expr = "time() - node_boot_time_seconds < 300";
|
||||
annotations.description = "{{$labels.instance}} rebooted";
|
||||
}
|
||||
{
|
||||
alert = "Mastodon";
|
||||
expr = ''probe_success{instance="https://social.krebsco.de"}'';
|
||||
for = "5m";
|
||||
annotations.description = "Mastodon instance {{$labels.instance}} is down";
|
||||
}
|
||||
{
|
||||
alert = "ProbeFailed";
|
||||
expr = "probe_success == 0";
|
||||
for = "5m";
|
||||
annotations.description = "HTTP probe failed for {{$labels.instance}}";
|
||||
}
|
||||
{
|
||||
alert = "SlowProbe";
|
||||
expr = "avg_over_time(probe_http_duration_seconds[1m]) > 1";
|
||||
for = "5m";
|
||||
annotations.description = "HTTP probe slow for {{$labels.instance}}";
|
||||
}
|
||||
{
|
||||
alert = "HttpStatusCode";
|
||||
expr = "probe_http_status_code != 0 AND (probe_http_status_code <= 199 OR probe_http_status_code >= 400)";
|
||||
for = "5m";
|
||||
annotations.description = "status code {{$value}} for {{$labels.instance}}";
|
||||
}
|
||||
{
|
||||
alert = "SslExpirySoon";
|
||||
expr = "probe_ssl_earliest_cert_expiry - time() < 86400 * 30";
|
||||
for = "5m";
|
||||
annotations.description = "SSL certificate for {{$labels.instance}} expires in 30 days";
|
||||
}
|
||||
{
|
||||
alert = "SslExpiry";
|
||||
expr = "probe_ssl_earliest_cert_expiry - time() <= 0";
|
||||
for = "5m";
|
||||
annotations.description = "SSL certificate for {{$labels.instance}} has expired";
|
||||
}
|
||||
];
|
||||
}
|
||||
];
|
||||
})
|
||||
];
|
||||
|
||||
# ref https://github.com/Mic92/dotfiles/blob/f44bac5dd6970ed3fbb4feb906917331ec3c2be5/machines/eva/modules/prometheus/default.nix
|
||||
systemd.services.matrix-hook = {
|
||||
@@ -246,6 +255,33 @@ in {
|
||||
};
|
||||
};
|
||||
|
||||
systemd.services.matrix-hook-lassulus = {
|
||||
description = "Matrix Hook";
|
||||
after = [ "network.target" ];
|
||||
wantedBy = [ "multi-user.target" ];
|
||||
environment = {
|
||||
HTTP_ADDRESS = "[::1]";
|
||||
HTTP_PORT = "9089";
|
||||
MX_HOMESERVER = "https://matrix.4d2.org";
|
||||
MX_ID = "@lakai:4d2.org";
|
||||
MX_ROOMID = "!MJAGqBAOKZGMywzwkI:lassul.us";
|
||||
MX_MSG_TEMPLATE = "${pkgs.matrix-hook}/message.html.tmpl";
|
||||
};
|
||||
serviceConfig = {
|
||||
EnvironmentFile = [
|
||||
# format: MX_TOKEN=<token>
|
||||
config.age.secrets.matrix-token-lakai-env.path
|
||||
];
|
||||
Type = "simple";
|
||||
ExecStart = "${pkgs.matrix-hook}/bin/matrix-hook";
|
||||
Restart = "always";
|
||||
RestartSec = "10";
|
||||
DynamicUser = true;
|
||||
User = "matrix-hook";
|
||||
Group = "matrix-hook";
|
||||
};
|
||||
};
|
||||
|
||||
age.secrets = {
|
||||
matrix-token-lakai-env.file = ../../../secrets/matrix-token-lakai-env.age;
|
||||
};
|
||||
@@ -260,8 +296,23 @@ in {
|
||||
group_wait = "30s";
|
||||
repeat_interval = "24h";
|
||||
receiver = "matrix";
|
||||
routes = [
|
||||
{
|
||||
receiver = "lassulus";
|
||||
matchers = [ "alertname = \"Mastodon\"" ];
|
||||
}
|
||||
];
|
||||
};
|
||||
receivers = [
|
||||
{
|
||||
name = "lassulus";
|
||||
webhook_configs = [
|
||||
{
|
||||
url = "http://localhost:9089/alert";
|
||||
max_alerts = 5;
|
||||
}
|
||||
];
|
||||
}
|
||||
{
|
||||
name = "matrix";
|
||||
webhook_configs = [
|
||||
|
||||
Reference in New Issue
Block a user