SNEWS_Operations

These are the useful bits for running the SNEWS 2.0 infrastructure.

Apptainer usage:

Apptainer is a handy container environment that doesn't require administrative access to install or execute.

Data storage locations:

~/apps
~/data
~/etc
~/run

Building the Apptainer container

$ cd SNEWS_Operations/apps/snews_cs
$ apptainer build snews_cs.img snews_cs.def

Executing the Apptainer container

$ apptainer run --app coinc snews_cs.img
$ apptainer run --app feedback snews_cs.img

Note: It is possible to configure the container to not use localhost as the smtp relay, as well as use smtp-auth. These items can be passed in through the environment so that things like passwords are not laying in plain text. They are also VERY environment specific and will differ depending on where this container is launched from.

apptainer run \
	--env smtp_server_addr=relay.physics.purdue.edu \
	--env [email protected] \
	--env snews_sender_pass=$(pass [email protected]) \
	--app coinc snews_cs.img firedrill

Log rotations

It will eventually become necessary to rotate your log files. An example configuration exists in SNEWS_Operations/etc/logrotate/logrotate.conf.

An okay crontab might be configured like:

7 1 * * * /usr/sbin/logrotate -l ~/data/logrotate/logrotate.log -s ~/run/logrotate/state ~/etc/logrotate/logrotate.conf

Grafana graph configurations

These settings are used to generate the dashboard. I believe this should be templated at some point, although it doesn't appear Grafana supports variables in the field name transformation, which could make things interesting.

1. Click dashboard
2. Add/Row
3. Name: XENONnT at a glance
4. Ok

On/Off status panel

1. Add/Visualization
2. Transform data
3. Add transformation
4. Filter by name (unfortunately, this doesn't support the usage of variables yet)
5. Field: Detector, Match: Is equal, Value: XENONnT

6. Visualizaions: Stat
7. itle: XENONnT
8. Value mappings: Value: on -> green, off -> red
9. Unit: Boolean - On/Off
10. Calculation: Last, Fields: Status
11. Apply/Save

Heartbeats data

1. Add/Visualization
    1. Transform data
    2. Add transformation
    3. Filter by name (unfortunately, this doesn't support the usage of variables yet)
    4. Field: Detector, Match: Is equal, Value: XENONnT

2. Query:
    1. Field: Recieved Times: Type Time
    2. Field: Stamped Times: Type Time
    3. Field: Status: Type String
    4. Field: Time After Last: Type Number

3. Visualizations: Table
    1. Title: XENONnT Heartbeats
    2. Decimals: 5

snews2 / snews_operations Goto Github PK