Fix for LDL connection monitoring

As a service wrapped as a daemon, an issue where an LDL connection hadn't been received was masked. The action, no LDL connection received, is corrected in this release.

Inclusion of recent servers

Recent changes have brought a few new servers into service, including prod3 and cdx3. This update adds these server names to the monitor system so that their metrics are tracked.

Offline service removal and status report tidy

Commented out cdx1 and crawler07.bl.uk as two offline servers. Rewrote construction of wa_status dashboard to remove 'space taking' section titles. Also, reworked x,y positioning.

Alerts and ELK metrics

Fix for some broken alerts and addition of ELK metrics.

Correction to nohup command

Update to ensure nohup out on production (monitor) doesn't swell over time.

Update to LDL connection monitoring over time

Previously, the collected data seemed to remain and be stale, rather than be dropped. This update changes the reporting process and the monitoring.

Fix to time difference

Now using datetime to calculate gap between last and latest curl requests.

Amendment to LDL script environ code

As monitoring does not run on prod but monitor, amended environ code.

LDL connection monitoring

Inclusion of LDL connection tracking and monitoring code.

Each LDL VM curls ldukwa-proxy via a cronjob (every 6 or 10 minutes). ldukwa-proxy forwards these to this code, which accumulates them and pushes to prometheus gateway. From the wa_status dashboard, the connection count can be seen.

Removed offline server

Our server 'bigcdx' is no longer online - removing from monitoring.