quivr/volumes/logs/vector.yml
Stan Girard 87138635aa
feat(install): it now takes 30 seconds to install Quivr (#1780)
# Description

Please include a summary of the changes and the related issue. Please
also include relevant motivation and context.

## Checklist before requesting a review

Please delete options that are not relevant.

- [ ] My code follows the style guidelines of this project
- [ ] I have performed a self-review of my code
- [ ] I have commented hard-to-understand areas
- [ ] I have ideally added tests that prove my fix is effective or that
my feature works
- [ ] New and existing unit tests pass locally with my changes
- [ ] Any dependent changes have been merged

## Screenshots (if appropriate):
2023-12-02 00:26:06 +01:00

233 lines
7.1 KiB
YAML

api:
enabled: true
address: 0.0.0.0:9001
sources:
docker_host:
type: docker_logs
exclude_containers:
- supabase-vector
transforms:
project_logs:
type: remap
inputs:
- docker_host
source: |-
.project = "default"
.event_message = del(.message)
.appname = del(.container_name)
del(.container_created_at)
del(.container_id)
del(.source_type)
del(.stream)
del(.label)
del(.image)
del(.host)
del(.stream)
router:
type: route
inputs:
- project_logs
route:
kong: '.appname == "supabase-kong"'
auth: '.appname == "supabase-auth"'
rest: '.appname == "supabase-rest"'
realtime: '.appname == "supabase-realtime"'
storage: '.appname == "supabase-storage"'
functions: '.appname == "supabase-functions"'
db: '.appname == "supabase-db"'
# Ignores non nginx errors since they are related with kong booting up
kong_logs:
type: remap
inputs:
- router.kong
source: |-
req, err = parse_nginx_log(.event_message, "combined")
if err == null {
.timestamp = req.timestamp
.metadata.request.headers.referer = req.referer
.metadata.request.headers.user_agent = req.agent
.metadata.request.headers.cf_connecting_ip = req.client
.metadata.request.method = req.method
.metadata.request.path = req.path
.metadata.request.protocol = req.protocol
.metadata.response.status_code = req.status
}
if err != null {
abort
}
# Ignores non nginx errors since they are related with kong booting up
kong_err:
type: remap
inputs:
- router.kong
source: |-
.metadata.request.method = "GET"
.metadata.response.status_code = 200
parsed, err = parse_nginx_log(.event_message, "error")
if err == null {
.timestamp = parsed.timestamp
.severity = parsed.severity
.metadata.request.host = parsed.host
.metadata.request.headers.cf_connecting_ip = parsed.client
url, err = split(parsed.request, " ")
if err == null {
.metadata.request.method = url[0]
.metadata.request.path = url[1]
.metadata.request.protocol = url[2]
}
}
if err != null {
abort
}
# Gotrue logs are structured json strings which frontend parses directly. But we keep metadata for consistency.
auth_logs:
type: remap
inputs:
- router.auth
source: |-
parsed, err = parse_json(.event_message)
if err == null {
.metadata.timestamp = parsed.time
.metadata = merge!(.metadata, parsed)
}
# PostgREST logs are structured so we separate timestamp from message using regex
rest_logs:
type: remap
inputs:
- router.rest
source: |-
parsed, err = parse_regex(.event_message, r'^(?P<time>.*): (?P<msg>.*)$')
if err == null {
.event_message = parsed.msg
.timestamp = to_timestamp!(parsed.time)
.metadata.host = .project
}
# Realtime logs are structured so we parse the severity level using regex (ignore time because it has no date)
realtime_logs:
type: remap
inputs:
- router.realtime
source: |-
.metadata.project = del(.project)
.metadata.external_id = .metadata.project
parsed, err = parse_regex(.event_message, r'^(?P<time>\d+:\d+:\d+\.\d+) \[(?P<level>\w+)\] (?P<msg>.*)$')
if err == null {
.event_message = parsed.msg
.metadata.level = parsed.level
}
# Storage logs may contain json objects so we parse them for completeness
storage_logs:
type: remap
inputs:
- router.storage
source: |-
.metadata.project = del(.project)
.metadata.tenantId = .metadata.project
parsed, err = parse_json(.event_message)
if err == null {
.event_message = parsed.msg
.metadata.level = parsed.level
.metadata.timestamp = parsed.time
.metadata.context[0].host = parsed.hostname
.metadata.context[0].pid = parsed.pid
}
# Postgres logs some messages to stderr which we map to warning severity level
db_logs:
type: remap
inputs:
- router.db
source: |-
.metadata.host = "db-default"
.metadata.parsed.timestamp = .timestamp
parsed, err = parse_regex(.event_message, r'.*(?P<level>INFO|NOTICE|WARNING|ERROR|LOG|FATAL|PANIC?):.*', numeric_groups: true)
if err != null || parsed == null {
.metadata.parsed.error_severity = "info"
}
if parsed != null {
.metadata.parsed.error_severity = parsed.level
}
if .metadata.parsed.error_severity == "info" {
.metadata.parsed.error_severity = "log"
}
.metadata.parsed.error_severity = upcase!(.metadata.parsed.error_severity)
sinks:
logflare_auth:
type: 'http'
inputs:
- auth_logs
encoding:
codec: 'json'
method: 'post'
request:
retry_max_duration_secs: 10
uri: 'http://analytics:4000/api/logs?source_name=gotrue.logs.prod&api_key=your-super-secret-and-long-logflare-key'
logflare_realtime:
type: 'http'
inputs:
- realtime_logs
encoding:
codec: 'json'
method: 'post'
request:
retry_max_duration_secs: 10
uri: 'http://analytics:4000/api/logs?source_name=realtime.logs.prod&api_key=your-super-secret-and-long-logflare-key'
logflare_rest:
type: 'http'
inputs:
- rest_logs
encoding:
codec: 'json'
method: 'post'
request:
retry_max_duration_secs: 10
uri: 'http://analytics:4000/api/logs?source_name=postgREST.logs.prod&api_key=your-super-secret-and-long-logflare-key'
logflare_db:
type: 'http'
inputs:
- db_logs
encoding:
codec: 'json'
method: 'post'
request:
retry_max_duration_secs: 10
# We must route the sink through kong because ingesting logs before logflare is fully initialised will
# lead to broken queries from studio. This works by the assumption that containers are started in the
# following order: vector > db > logflare > kong
uri: 'http://kong:8000/analytics/v1/api/logs?source_name=postgres.logs&api_key=your-super-secret-and-long-logflare-key'
logflare_functions:
type: 'http'
inputs:
- router.functions
encoding:
codec: 'json'
method: 'post'
request:
retry_max_duration_secs: 10
uri: 'http://analytics:4000/api/logs?source_name=deno-relay-logs&api_key=your-super-secret-and-long-logflare-key'
logflare_storage:
type: 'http'
inputs:
- storage_logs
encoding:
codec: 'json'
method: 'post'
request:
retry_max_duration_secs: 10
uri: 'http://analytics:4000/api/logs?source_name=storage.logs.prod.2&api_key=your-super-secret-and-long-logflare-key'
logflare_kong:
type: 'http'
inputs:
- kong_logs
- kong_err
encoding:
codec: 'json'
method: 'post'
request:
retry_max_duration_secs: 10
uri: 'http://analytics:4000/api/logs?source_name=cloudflare.logs.prod&api_key=your-super-secret-and-long-logflare-key'