RDMurray
commented
1 year ago Yes and those logs are 99% useless information. They could definitely be cleaned up. I was looking at what the Scottish Tech Army did with the backend. I like that they separated the ingest service into initial import and update, rather than having a long running process. This should be easier to manage and we could schedule the updates at less busy times to avoid performance issues if necessary. A single container should be sufficient though, the scripts can be started with docker compose run.
The tile service exposes a few methods to check its health (as noted in #67), but to my knowledge the only way to see how the ingest service is doing is to ssh into the server and look at the container logs. It would be nice to see when the last data import ran, what its status was, and how long it took. I'm open to suggestions as to how to do that.