Context: On several occasions, the site became unusable due to the lack of disk space on the EBS storage. More specifically, the main Node.js server crashes frequently resulting in restarts from PM2 and therefore loss of user session data (related: #81).
There are several culprits to this, namely Redis persisting weather data on disk, along with Amazon Linux packages being automatically updated.
Related issues:
To prevent this monitoring could exist leveraging CloudWatch metrics to automatically create alarms to be proactive before outages hit.
Context: On several occasions, the site became unusable due to the lack of disk space on the EBS storage. More specifically, the main Node.js server crashes frequently resulting in restarts from PM2 and therefore loss of user session data (related: #81).
There are several culprits to this, namely Redis persisting weather data on disk, along with Amazon Linux packages being automatically updated.
Related issues:
To prevent this monitoring could exist leveraging CloudWatch metrics to automatically create alarms to be proactive before outages hit.