prometheus related updates
[distro-setup] / filesystem / etc / default / prometheus
index 63d1ee3798fa0b2c109f67aa57718e9b2089e6c1..c4e807a1ef150262962ca780bc634ecb0a47a6e0 100644 (file)
 # iank: initial file from 2.24, added to empty ARGS.
 
-# Set the command-line arguments to pass to the server.
+# config.file and tsdb.path are to configure the upstream version to use
+# the default locations of the debian package.
 
-ARGS="--web.listen-address=127.0.0.1:9090"
+# Sets the command-line arguments to pass to the server. to see all options, run prometheus --help
 
+ARGS="--web.listen-address=127.0.0.1:9090
+--web.external-url=https://i.b8.nz:9091
+--log.level=info
+--config.file=/etc/prometheus/prometheus.yml
+--storage.tsdb.path=/var/lib/prometheus/metrics2/"
 
-
-# Prometheus supports the following options:
-#  --config.file="/etc/prometheus/prometheus.yml"
-#                             Prometheus configuration file path.
-#  --web.listen-address="0.0.0.0:9090"
-#                             Address to listen on for UI, API, and telemetry.
-#  --web.read-timeout=5m      Maximum duration before timing out read of the
-#                             request, and closing idle connections.
-#  --web.max-connections=512  Maximum number of simultaneous connections.
-#  --web.external-url=<URL>   The URL under which Prometheus is externally
-#                             reachable (for example, if Prometheus is served
-#                             via a reverse proxy). Used for generating
-#                             relative and absolute links back to Prometheus
-#                             itself. If the URL has a path portion, it will
-#                             be used to prefix all HTTP endpoints served by
-#                             Prometheus. If omitted, relevant URL components
-#                             will be derived automatically.
-#  --web.route-prefix=<path>  Prefix for the internal routes of web endpoints.
-#                             Defaults to path of --web.external-url.
-#  --web.local-assets="/usr/share/prometheus/web/"
-#                             Path to static asset/templates directory.
-#  --web.user-assets=<path>   Path to user asset directory, available at
-#                             /user.
-#  --web.enable-lifecycle     Enable shutdown and reload via HTTP request.
-#  --web.enable-admin-api     Enable API endpoints for admin control actions.
-#  --web.console.templates="/etc/prometheus/consoles"
-#                             Path to the console template directory,
-#                             available at /consoles.
-#  --web.console.libraries="/etc/prometheus/console_libraries"
-#                             Path to the console library directory.
-#  --web.page-title="Prometheus Time Series Collection and Processing Server"
-#                             Document title of Prometheus instance.
-#  --web.cors.origin=".*"     Regex for CORS origin. It is fully anchored.
-#                             Example: 'https?://(domain1|domain2)\.com'
-#  --storage.tsdb.path="/var/lib/prometheus/metrics2/"
-#                             Base path for metrics storage.
-#  --storage.tsdb.retention=15d
-#                             [DEPRECATED] How long to retain samples in
-#                             storage. This flag has been deprecated, use
-#                             "storage.tsdb.retention.time" instead
-#  --storage.tsdb.retention.time=15d
-#                             How long to retain samples in storage. When this
-#                             flag is set it overrides
-#                             "storage.tsdb.retention".
-#                             If neither this flag nor "storage.tsdb.retention"
-#                             nor "storage.tsdb.retention.size" is set, the
-#                             retention time defaults to 15d.
-#                             Units Supported: y, w, d, h, m, s, ms.
-#  --storage.tsdb.retention.size=
-#                             [EXPERIMENTAL] Maximum number of bytes that can
-#                             be stored for blocks. Units supported: KB, MB,
-#                             GB, TB, PB. This flag is experimental and can be
-#                             changed in future releases.
-#  --storage.tsdb.use-lockfile
-#                             Create a lockfile in data directory.
-#  --storage.tsdb.allow-overlapping-blocks
-#                             [EXPERIMENTAL] Allow overlapping blocks, which
-#                             in turn enables vertical compaction and
-#                             vertical query merge.
-#  --storage.tsdb.wal-compression
-#                             Compress the tsdb WAL.
-#  --storage.remote.flush-deadline=<duration>
-#                             How long to wait flushing sample on shutdown or
-#                             config reload.
-#  --storage.remote.read-sample-limit=5e7
-#                             Maximum overall number of samples to return via
-#                             the remote read interface, in a single query. 0
-#                             means no limit. This limit is ignored for
-#                             streamed response types.
-#  --storage.remote.read-concurrent-limit=10
-#                             Maximum number of concurrent remote read calls.
-#                             0 means no limit.
-#  --storage.remote.read-max-bytes-in-frame=1048576
-#                             Maximum number of bytes in a single frame for
-#                             streaming remote read response types before
-#                             marshalling. Note that client might have limit on
-#                             frame size as well. 1MB as recommended by
-#                             protobuf by default.
-#  --rules.alert.for-outage-tolerance=1h
-#                             Max time to tolerate prometheus outage for
-#                             restoring "for" state of alert.
-#  --rules.alert.for-grace-period=10m
-#                             Minimum duration between alert and restored "for"
-#                             state. This is maintained only for alerts with
-#                             configured "for" time greater than grace period.
-#  --rules.alert.resend-delay=1m
-#                             Minimum amount of time to wait before resending
-#                             an alert to Alertmanager.
-#  --alertmanager.notification-queue-capacity=10000
-#                             The capacity of the queue for pending
-#                             Alertmanager notifications.
-#  --alertmanager.timeout=10s
-#                             Timeout for sending alerts to Alertmanager.
-#  --query.lookback-delta=5m  The maximum lookback duration for retrieving
-#                             metrics during expression evaluations and
-#                             federation.
-#  --query.timeout=2m         Maximum time a query may take before being
-#                             aborted.
-#  --query.max-concurrency=20
-#                             Maximum number of queries executed concurrently.
-#  --query.max-samples=50000000
-#                             Maximum number of samples a single query can load
-#                             into memory. Note that queries will fail if they
-#                             try to load more samples than this into memory,
-#                             so this also limits the number of samples a query
-#                             can return.
-#  --log.level=info           Only log messages with the given severity or
-#                             above. One of: [debug, info, warn, error]
-#  --log.format=logfmt        Output format of log messages. One of: [logfmt,
-#                             json]
+# for testing
+#--query.lookback-delta=15s"