This is Sirepo health check utility. It checks the availability of the light page of the specified servers (currently 2 local NSLS-II servers and RadiaSoft's alpha/beta servers. It creates a .json status file (stored in /tmp, via a crontab job, see the crontab file for details). On the first start the utility checks the current statuses of the machines and sends a notification to the specified addresses like:
On 9/6/2017 17:15, Sirepo Health Check wrote:
Monitoring started for:
- https://expdev.nsls2.bnl.gov/light (up)
- http://nsls2expdev1.bnl.gov:8000/light (up)
- https://alpha.sirepo.com/light (up)
- https://beta.sirepo.com/light (up)
If any status changes, it sends the corresponding notification like:
http://nsls2expdev1.bnl.gov:8000/light: status changed: up -> down (2017-09-06 17:30:02)
When the server is back, it sends the confirmation:
http://nsls2expdev1.bnl.gov:8000/light: status changed: down -> up (2017-09-06 17:35:02)
The crontab job runs every 5 minutes on weekdays from 8 am to 7:55 pm. If the machine is not restored, a reminder email will be sent in 2 hours after the last notification.
Run it in one line:
$ docker run -it --rm -e SLACK_WEBHOOK_URL -e SLACK_BOT_TOKEN -e SLACK_POSTING_CHANNEL nsls2/debian-with-miniconda:v0.1.2 bash -c "conda create -n sirepo-monitoring python=3.9 ipython phantomjs -c conda-forge -y && . /opt/conda/etc/profile.d/conda.sh && conda activate sirepo-monitoring && git clone https://github.com/NSLS-II/sirepo-healthcheck && cd sirepo-healthcheck/ && pip install -r requirements.txt && pip install -r requirements-dev.txt && python -VV && python health_check.py"