From dabd9f0f02b44b048b6355184fa64612201db72d Mon Sep 17 00:00:00 2001 From: lassulus Date: Sat, 28 Apr 2018 09:41:43 +0200 Subject: l monitoring: open ports --- lass/2configs/monitoring/node-exporter.nix | 6 ++++-- lass/2configs/monitoring/prometheus-server.nix | 6 ++++++ 2 files changed, 10 insertions(+), 2 deletions(-) (limited to 'lass/2configs/monitoring') diff --git a/lass/2configs/monitoring/node-exporter.nix b/lass/2configs/monitoring/node-exporter.nix index 8c27e90d4..561e3a25c 100644 --- a/lass/2configs/monitoring/node-exporter.nix +++ b/lass/2configs/monitoring/node-exporter.nix @@ -1,7 +1,9 @@ { config, lib, pkgs, ... }: { - networking.firewall.allowedTCPPorts = [ 9100 ]; - + krebs.iptables.tables.filter.INPUT.rules = [ + { predicate = "-i retiolum -p tcp --dport 9100 -s ${config.krebs.hosts.prism.nets.retiolum.ip4.addr}"; target = "ACCEPT"; v6 = false; } + { predicate = "-i retiolum -p tcp --dport 9100 -s ${config.krebs.hosts.prism.nets.retiolum.ip6.addr}"; target = "ACCEPT"; v4 = false; } + ]; services.prometheus.exporters = { node = { enable = true; diff --git a/lass/2configs/monitoring/prometheus-server.nix b/lass/2configs/monitoring/prometheus-server.nix index d56d7e552..c5c97412d 100644 --- a/lass/2configs/monitoring/prometheus-server.nix +++ b/lass/2configs/monitoring/prometheus-server.nix @@ -9,6 +9,12 @@ # useDHCP = true; #}; + krebs.iptables.tables.filter.INPUT.rules = [ + { predicate = "-i retiolum -p tcp --dport 3000"; target = "ACCEPT"; } + { predicate = "-i retiolum -p tcp --dport 9090"; target = "ACCEPT"; } + { predicate = "-i retiolum -p tcp --dport 9093"; target = "ACCEPT"; } + ]; + services = { prometheus = { enable = true; -- cgit v1.2.3 From a0862fa505ba8fb1d94c8bdac69a2077ba89bcdc Mon Sep 17 00:00:00 2001 From: lassulus Date: Sat, 28 Apr 2018 09:43:17 +0200 Subject: l monitoring: monitor more hosts --- lass/2configs/monitoring/prometheus-server.nix | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) (limited to 'lass/2configs/monitoring') diff --git a/lass/2configs/monitoring/prometheus-server.nix b/lass/2configs/monitoring/prometheus-server.nix index c5c97412d..92bb0519f 100644 --- a/lass/2configs/monitoring/prometheus-server.nix +++ b/lass/2configs/monitoring/prometheus-server.nix @@ -130,11 +130,10 @@ static_configs = [ { targets = [ - "localhost:9100" - ]; - labels = { - alias = "prometheus.example.com"; - }; + ] ++ map (host: "${host}:9100") (lib.attrNames (lib.filterAttrs (_: host: host.owner.name == "lass" && host.monitoring) config.krebs.hosts)); + #labels = { + # alias = "prometheus.example.com"; + #}; } ]; } -- cgit v1.2.3 From da44ae1115af80bb71f38de20b7421d08e435ea7 Mon Sep 17 00:00:00 2001 From: lassulus Date: Sat, 28 Apr 2018 09:43:51 +0200 Subject: l monitoring: print alarms to irc --- lass/2configs/monitoring/prometheus-server.nix | 33 ++++++++++++++++++++++++++ 1 file changed, 33 insertions(+) (limited to 'lass/2configs/monitoring') diff --git a/lass/2configs/monitoring/prometheus-server.nix b/lass/2configs/monitoring/prometheus-server.nix index 92bb0519f..1f9419e1a 100644 --- a/lass/2configs/monitoring/prometheus-server.nix +++ b/lass/2configs/monitoring/prometheus-server.nix @@ -181,4 +181,37 @@ security = import ; # { AdminUser = ""; adminPassword = ""} }; }; + services.logstash = { + enable = true; + inputConfig = '' + http { + port => 14813 + host => "127.0.0.1" + } + ''; + filterConfig = '' + if ([alerts]) { + ruby { + code => ' + lines = [] + event["alerts"].each {|p| + lines << "#{p["labels"]["instance"]}#{p["annotations"]["summary"]} #{p["status"]}" + } + event["output"] = lines.join("\n") + ' + } + } + ''; + outputConfig = '' + file { path => "/tmp/logs.json" codec => "json_lines" } + irc { + channels => [ "#noise" ] + host => "irc.r" + nick => "alarm" + codec => "json_lines" + format => "%{output}" + } + ''; + #plugins = [ ]; + }; } -- cgit v1.2.3 From 4190562d1233e40b3364c1bd812f2702a0748e49 Mon Sep 17 00:00:00 2001 From: lassulus Date: Sat, 28 Apr 2018 16:12:10 +0200 Subject: l monitoring: delete legacy config --- lass/2configs/monitoring/client.nix | 26 --------- lass/2configs/monitoring/monit-alarms.nix | 44 ---------------- lass/2configs/monitoring/server.nix | 87 ------------------------------- 3 files changed, 157 deletions(-) delete mode 100644 lass/2configs/monitoring/client.nix delete mode 100644 lass/2configs/monitoring/monit-alarms.nix delete mode 100644 lass/2configs/monitoring/server.nix (limited to 'lass/2configs/monitoring') diff --git a/lass/2configs/monitoring/client.nix b/lass/2configs/monitoring/client.nix deleted file mode 100644 index b8c245215..000000000 --- a/lass/2configs/monitoring/client.nix +++ /dev/null @@ -1,26 +0,0 @@ -{pkgs, config, ...}: -with import ; -{ - services.telegraf = { - enable = true; - - extraConfig = { - agent.interval = "1s"; - outputs = { - influxdb = { - urls = ["http://prism:8086"]; - database = "telegraf_db"; - user_agent = "telegraf"; - }; - }; - inputs = { - cpu = { - percpu = false; - totalcpu = true; - }; - mem = {}; - net = {}; - }; - }; - }; -} diff --git a/lass/2configs/monitoring/monit-alarms.nix b/lass/2configs/monitoring/monit-alarms.nix deleted file mode 100644 index 2cfc292e5..000000000 --- a/lass/2configs/monitoring/monit-alarms.nix +++ /dev/null @@ -1,44 +0,0 @@ -{pkgs, config, ...}: -with import ; -let - echoToIrc = msg: - pkgs.writeDash "echo_irc" '' - set -euf - export LOGNAME=prism-alarm - ${pkgs.irc-announce}/bin/irc-announce \ - irc.r 6667 ${config.networking.hostName}-alarm \#noise "${msg}" >/dev/null - ''; - -in { - krebs.monit = { - enable = true; - http.enable = true; - alarms = { - nirwanabluete = { - test = "${pkgs.curl}/bin/curl -sf 'https://nirwanabluete.de/'"; - alarm = echoToIrc "test nirwanabluete failed"; - }; - ubik = { - test = "${pkgs.curl}/bin/curl -sf 'https://ubikmedia.de'"; - alarm = echoToIrc "test ubik failed"; - }; - cac-panel = { - test = "${pkgs.curl}/bin/curl -sf 'https://panel.cloudatcost.com/login.php'"; - alarm = echoToIrc "test cac-panel failed"; - }; - radio = { - test = pkgs.writeBash "check_stream" '' - ${pkgs.curl}/bin/curl -sif http://lassul.us:8000/radio.ogg \ - | ${pkgs.gawk}/bin/awk '/^\r$/{exit}{print $0}' \ - | ${pkgs.gnugrep}/bin/grep -q "200 OK" || exit "''${PIPESTATUS[0]}" - ''; - alarm = echoToIrc "test radio failed"; - }; - }; - }; - - krebs.iptables.tables.filter.INPUT.rules = [ - { predicate = "-p tcp -i retiolum --dport 9093"; target = "ACCEPT"; } - ]; -} - diff --git a/lass/2configs/monitoring/server.nix b/lass/2configs/monitoring/server.nix deleted file mode 100644 index adaecde2c..000000000 --- a/lass/2configs/monitoring/server.nix +++ /dev/null @@ -1,87 +0,0 @@ -{pkgs, config, ...}: -with import ; -{ - services.influxdb.enable = true; - - services.influxdb.extraConfig = { - meta.hostname = config.krebs.build.host.name; - # meta.logging-enabled = true; - http.bind-address = ":8086"; - admin.bind-address = ":8083"; - http.log-enabled = false; - monitoring = { - enabled = false; - # write-interval = "24h"; - }; - collectd = [{ - enabled = true; - typesdb = "${pkgs.collectd}/share/collectd/types.db"; - database = "collectd_db"; - port = 25826; - }]; - }; - - krebs.kapacitor = - let - db = "telegraf_db"; - echoToIrc = pkgs.writeDash "echo_irc" '' - set -euf - data="$(${pkgs.jq}/bin/jq -r .message)" - export LOGNAME=prism-alarm - ${pkgs.irc-announce}/bin/irc-announce \ - irc.r 6667 prism-alarm \#noise "$data" >/dev/null - ''; - in { - enable = true; - alarms = { - cpu = { - database = db; - text = '' - var data = batch - |query(${"'''"} - SELECT mean("usage_user") AS mean - FROM "${db}"."default"."cpu" - ${"'''"}) - .period(10m) - .every(1m) - .groupBy('host') - data |alert() - .crit(lambda: "mean" > 90) - .exec('${echoToIrc}') - data |deadman(1.0,5m) - .stateChangesOnly() - .exec('${echoToIrc}') - ''; - }; - ram = { - database = db; - text = '' - var data = batch - |query(${"'''"} - SELECT mean("used_percent") AS mean - FROM "${db}"."default"."mem" - ${"'''"}) - .period(10m) - .every(1m) - .groupBy('host') - data |alert() - .crit(lambda: "mean" > 90) - .exec('${echoToIrc}') - ''; - }; - }; - }; - - services.grafana = { - enable = true; - addr = "0.0.0.0"; - auth.anonymous.enable = true; - security = import ; # { AdminUser = ""; adminPassword = ""} - }; - - krebs.iptables.tables.filter.INPUT.rules = [ - { predicate = "-p tcp -i retiolum --dport 8086"; target = "ACCEPT"; } - { predicate = "-p tcp -i retiolum --dport 3000"; target = "ACCEPT"; } - { predicate = "-p udp -i retiolum --dport 25826"; target = "ACCEPT"; } - ]; -} -- cgit v1.2.3 From 28d6704a0d617ca7d379b836ab9fdd4d6a0be868 Mon Sep 17 00:00:00 2001 From: lassulus Date: Thu, 3 May 2018 18:04:38 +0200 Subject: l monitoring: use correct logstash url --- lass/2configs/monitoring/prometheus-server.nix | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'lass/2configs/monitoring') diff --git a/lass/2configs/monitoring/prometheus-server.nix b/lass/2configs/monitoring/prometheus-server.nix index 1f9419e1a..e16d421a0 100644 --- a/lass/2configs/monitoring/prometheus-server.nix +++ b/lass/2configs/monitoring/prometheus-server.nix @@ -164,7 +164,7 @@ ]; "webhook_configs" = [ { - "url" = "https://example.com/prometheus-alerts"; + "url" = "http://127.0.0.1:14813/prometheus-alerts"; "send_resolved" = true; } ]; -- cgit v1.2.3