Commit f45e743b authored by Jérome Perrin's avatar Jérome Perrin

grafana

parent 657c16d6
......@@ -15,7 +15,7 @@
[instance-profile]
filename = instance.cfg.in
md5sum = 33e6510d84c83a46a2edf217b4f1beb5
md5sum = 6ff69aef5a7185f880359a78965e072e
[influxdb-config-file]
filename = influxdb-config-file.cfg.in
......
{
"$schema": "http://json-schema.org/draft-07/schema#",
"$schema": "https://json-schema.org/draft/2020-12/schema",
"description": "Parameters to instantiate Grafana",
"type": "object",
"additionalProperties": false,
"$defs": {
"type": {
"description": "Type of the application. With `SlapOS` type, some metrics are collected from supervisor and from some known partition types (for example: ERP5's mariadb or ERP5's zopes). With `system` type, only log files are ingested.",
"type": "string",
"default": "SlapOS",
"enum": [
"SlapOS",
"system"
]
},
"name": {
"description": "Name of this application",
"type": "string"
},
"urls": {
"description": "URLs to monitor for availability and certificate lifetime",
"type": "array",
"items": {
"type": "string"
}
},
"log-file-patterns": {
"type": "array",
"items": {
"type": "string"
},
"description": "Glob patterns for watched log files."
},
"static-tags": {
"type": "object",
"description": "Static tags for this partition",
"examples": [
{
"region": "eu",
"data-center": "abc123"
}
]
}
},
"properties": {
"smtp-server": {
"description": "SMTP server used by grafana to send emails (in host:port format). Leaving this empty will disable email sending.",
"description": "SMTP server used by Grafana to send emails (in host:port format). Leaving this empty will disable email sending.",
"type": "string"
},
"smtp-username": {
......@@ -33,133 +72,171 @@
"description": "Applications to monitor",
"type": "array",
"items": {
"type": "object",
"required": [
"name",
"instance-root",
"partitions"
],
"properties": {
"name": {
"description": "Name of this application",
"type": "string"
},
"instance-root": {
"description": "Directory containing SlapOS partitions.",
"type": "string"
},
"urls": {
"description": "URLs to monitor for availability and certificate lifetime",
"type": "array",
"items": {
"type": "string"
}
},
"partitions": {
"description": "SlapOS partitions to monitor",
"type": "array",
"items": {
"type": "object",
"properties": {
"name": {
"type": "string",
"description": "Friendly name of the partition",
"examples": [
"mariadb",
"zope-activity"
]
},
"reference": {
"type": "string",
"description": "Reference of the partition",
"examples": [
"slappart1",
"slappart2"
]
},
"type": {
"type": "string",
"description": "Type of the partition. Known types have extra metrics and logs collected",
"enum": [
"erp5/mariadb",
"erp5/balancer",
"erp5/zope-activity",
"erp5/zope-front",
"erp5/zeo",
"mariadb",
"default"
]
},
"file-path": {
"type": "string",
"description": "Glob for the files to watch. This mostly makes sense for `default` type"
},
"static-tags": {
"oneOf": [
{
"type": "object",
"additionalProperties": false,
"description": "Configuration for SlapOS type application",
"required": [
"type",
"name",
"instance-root",
"partitions"
],
"properties": {
"type": {
"$ref": "#/$defs/type",
"const": "SlapOS"
},
"name": {
"$ref": "#/$defs/name"
},
"urls": {
"$ref": "#/$defs/urls"
},
"instance-root": {
"description": "Directory containing SlapOS partitions.",
"type": "string"
},
"partitions": {
"description": "SlapOS partitions to monitor",
"type": "array",
"items": {
"type": "object",
"description": "Static tags for this partition",
"required": [
"name",
"reference"
],
"properties": {
"name": {
"type": "string",
"description": "Friendly name of the partition",
"examples": [
"mariadb",
"zope-activity"
]
},
"reference": {
"type": "string",
"description": "Reference of the partition",
"examples": [
"slappart1",
"slappart2"
]
},
"type": {
"type": "string",
"description": "Type of the partition. Known types have metrics and logs collected",
"enum": [
"erp5/mariadb",
"erp5/balancer",
"erp5/zope-activity",
"erp5/zope-front",
"erp5/zeo",
"mariadb",
"default"
],
"default": "default"
},
"log-file-patterns": {
"$refs": "#/$defs/log-file-patterns",
"description": "Glob pattern for log files to watch. This mostly makes sense for `default` partition type"
},
"static-tags": {
"$refs": "#/$defs/static-tags"
}
},
"examples": [
{
"region": "eu",
"data-center": "abc123"
"name": "zope-backoffice",
"type": "erp5/zope-front",
"reference": "slappart1",
"static-tags": {
"instance": "instance-name"
}
},
{
"name": "mariadb",
"type": "erp5/mariadb",
"reference": "slappart2"
},
{
"name": "Theia",
"type": "default",
"log-file-patterns": [
".slappart*log"
]
}
]
}
}
}
},
{
"type": "object",
"description": "Configuration for `system` type application",
"required": [
"type",
"name"
],
"properties": {
"type": {
"$ref": "#/$defs/type",
"const": "system"
},
"name": {
"$ref": "#/$defs/name"
},
"anyOf": [
{
"urls": {
"$ref": "#/$defs/urls"
},
"partitions": {
"type": "array",
"items": {
"type": "object",
"properties": {
"type": {
"const": "default"
"name": {
"type": "string",
"description": "Friendly name of the partition",
"examples": [
"syslog",
"email"
]
},
"log-file-patterns": {
"$refs": "#/$defs/log-file-patterns"
},
"static-tags": {
"$refs": "#/$defs/static-tags"
}
},
"required": [
"name",
"file-path"
]
},
{
"properties": {
"type": {
"not": {
"const": "default"
}
"examples": [
{
"name": "syslog",
"log-file-patterns": [
"/var/log/syslog"
]
},
{
"name": "kernel",
"log-file-patterns": [
"/var/log/kern.log",
"/var/log/messages"
]
},
{
"name": "re6stnet",
"log-file-patterns": [
"/var/log/re6stnet/*.log"
]
}
},
"required": [
"name",
"reference"
]
}
],
"examples": [
{
"name": "zope-backoffice",
"type": "erp5/zope-front",
"reference": "slappart1",
"static-tags": {
"instance": "instance-name"
}
},
{
"name": "mariadb",
"type": "erp5/mariadb",
"reference": "slappart2"
},
{
"name": "syslog",
"type": "default",
"file-path": "/var/log/syslog"
}
]
}
}
}
}
]
}
},
"promtail-extra-scrape-config": {
"description": "Raw promtail config (experimental parameter, see https://github.com/grafana/loki/blob/v0.3.0/docs/promtail.md#scrape-configs for detail)",
"default": "",
"type": "string"
}
}
}
......@@ -251,7 +251,7 @@ init =
inputs = collections.defaultdict(list)
processors = collections.defaultdict(list)
slapparameter_dict = self.options["slapparameter-dict"]
for application in slapparameter_dict.get('applications', []):
for application in slapparameter_dict.get("applications", []):
partition_mapping = {}
for partition in application.get("partitions", []):
partition.setdefault("type", "default")
......@@ -351,50 +351,49 @@ init =
"tags": {"app": application["name"]},
})
# TODO: don't run more than one per instance_root
telegraf_slapos_input_config_file = os.path.join(
self.options['location'],
f"telegraf-input-slapos-{application['name']}.cfg")
self._config_files[telegraf_slapos_input_config_file] = toml.dumps({
"inputs": {
"slapos": [{
"instance_root": application['instance-root']}]}})
# TODO: supervisor process finder for
# https://github.com/influxdata/telegraf/tree/master/plugins/inputs/procstat ?
telegraf_slapos_input_command = self.options['telegraf-input-slapos-bin']
inputs["execd"].append({
"name_override": f"{application['name']}-processes",
"command": [telegraf_slapos_input_command, '-config', telegraf_slapos_input_config_file],
"tags": {"app": application["name"]},
})
# "cleanup" slapos process names, remove hash from wrappers and -on-watch suffix
processors["regex"].append({
"namepass": [f"{application['name']}-processes"],
"order": 1,
"tags": [{
"key": "name",
"pattern": "^(.*)-.{32}",
# XXX we concatenate strings so that we don't have to escape them for buildout
"replacement": "$" + "{1}",
}]})
processors["regex"].append({
"namepass": [f"{application['name']}-processes"],
"order": 2,
"tags": [{
"key": "name",
"pattern": "^(.*)-on-watch$",
"replacement": "$" + "{1}",
}]})
processors["enum"].append({
"namepass": [ f"{application['name']}-processes"],
"mapping": [{
# "tag": "group", # TODO: rename this in input plugin
"tag": "slappart",
"dest": "partition",
"value_mappings": partition_mapping,
}]})
if application.get("type") == "SlapOS":
telegraf_slapos_input_config_file = os.path.join(
self.options['location'],
f"telegraf-input-slapos-{application['name']}.cfg")
self._config_files[telegraf_slapos_input_config_file] = toml.dumps({
"inputs": {
"slapos": [{
"instance_root": application["instance-root"]}]}})
# TODO: supervisor process finder for
# https://github.com/influxdata/telegraf/tree/master/plugins/inputs/procstat ?
telegraf_slapos_input_command = self.options['telegraf-input-slapos-bin']
inputs["execd"].append({
"name_override": f"{application['name']}-processes",
"command": [telegraf_slapos_input_command, '-config', telegraf_slapos_input_config_file],
"tags": {"app": application["name"]},
})
# "cleanup" slapos process names, remove hash from wrappers and -on-watch suffix
processors["regex"].append({
"namepass": [f"{application['name']}-processes"],
"order": 1,
"tags": [{
"key": "name",
"pattern": "^(.*)-.{32}",
# XXX we concatenate strings so that we don't have to escape them for buildout
"replacement": "$" + "{1}",
}]})
processors["regex"].append({
"namepass": [f"{application['name']}-processes"],
"order": 2,
"tags": [{
"key": "name",
"pattern": "^(.*)-on-watch$",
"replacement": "$" + "{1}",
}]})
processors["enum"].append({
"namepass": [ f"{application['name']}-processes"],
"mapping": [{
# "tag": "group", # TODO: rename this in input plugin # XXX I don't remember what this means
"tag": "slappart",
"dest": "partition",
"value_mappings": partition_mapping,
}]})
# TODOs:
# - [ ] slapos input
......@@ -404,6 +403,7 @@ init =
# - [ ] alert dashboard
# - [ ] inclu "jerome-dev" partout ???
# - [ ] apdex
# - [ ] "job" is bad name in Explore
options["extra-config"] = toml.dumps({
"inputs": inputs,
......@@ -497,8 +497,6 @@ command-line =
wrapper-path = ${directory:service}/${:_buildout_section_name_}
extra-command-line =
# level=error ts=2022-09-24T14:40:13.636615531Z caller=scheduler_processor.go:182 org_id=fake msg="error notifying frontend about finished query" err="rpc error: code = ResourceExhausted desc = grpc: received message larger than max (4200411 vs. 4194304)" frontend=10.0.44.65:9099
[loki-listen-promise-macro]
<= check-url-available-promise
......@@ -703,6 +701,7 @@ install =
},
{
"regex": {
# TODO don't include the ----
"expression": "^------\\n(?P<timestamp>\\d{4}-\\d{2}-\\d{2}\\s\\d{1,2}\\:\\d{2}\\:\\d{2}\\,\\d{3}) (?P<level>\\S+) (?P<component>\\S+) (?P<message>.*)"
}
},
......@@ -714,9 +713,7 @@ install =
},
{
"labels": {
"level": None,
# XXX do we really want `component` ? it may cause lots of cardinality
# "component": None
"level": None
}
}
]
......@@ -873,7 +870,6 @@ install =
{
"labels": {
"level": None,
"component": None
}
}
]
......
......@@ -63,6 +63,13 @@ go.importpath = github.com/perrinjerome/telegraf-input-slapos
repository = https://github.com/perrinjerome/telegraf-input-slapos
revision = v0.0.1-0-gf8981f3
[go_github.com_prometheus_prometheus]
<= go-git-package
go.importpath = github.com/prometheus/prometheus
repository = https://github.com/prometheus/prometheus
revision = v0.41.0-0-gc0d8a56c6
# [go_github.com_jaegertracking_jaeger]
# <= go-git-package
# go.importpath = github.com/jaegertracking/jaeger
......@@ -70,8 +77,7 @@ revision = v0.0.1-0-gf8981f3
# revision = v1.20.0-623-gcac21f82
[gowork]
# Fails with current default golang1.18
golang = ${golang1.17:location}
golang = ${golang1.19:location}
install =
${go_github.com_grafana_loki:location}:./cmd/loki
${go_github.com_grafana_loki:location}:./clients/cmd/promtail
......@@ -79,6 +85,7 @@ install =
${go_github.com_influxdata_telegraf:location}:./cmd/...
${go_github.com_influxdata_influxdb:location}:./cmd/...
${go_github.com_perrinjerome_slapos_telegraf_input:location}:./...
${go_github.com_prometheus_prometheus:location}:./cmd/...
# disable cgo, to prevent loki/promtail from using go-systemd
environment =
......
......@@ -82,7 +82,7 @@ class TestGrafana(GrafanaTestCase):
self.assertEqual(requests.codes.ok, resp.status_code)
self.assertEqual(1, resp.json()['id'])
def test_grafana_datasource_povisinonned(self):
def test_grafana_datasource_provisioned(self):
# data sources are provisionned
connection_params = self.computer_partition.getConnectionParameterDict()
resp = requests.get(
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment