Restart of the docker container also restarts the indexing of blocks

I followed the steps as described below;
- installed docker
- added my account to the docker group (so I can issue docker commands without having to sudo)
- git clone https://github.com/Blockstream/esplora && cd esplora
- docker build -t esplora .
- docker run -p 50001:50001 -p 8080:80 \
--volume $PWD/data_bitcoin_mainnet:/data \
--rm -i -t esplora \
bash -c "/srv/explorer/run.sh bitcoin-mainnet explorer"

The container starts, I see a lot of console output. Based on the output I assume that it is indexing blocks which I think is weird, because I already have a fully synched node outside of the docker container. Eventually I see that the indexing of blocks has completed (took nearly 24 hours) but it got really quiet after that. Trying to connect to http://localhost:8080 gives me the dreaded "Esplora is currently unavailable, please try again later".

I decided to wait but that didn't change anything. Finally I shut down the docker container and restarted it, and the whole indexing process just starts from 0.000% again.

Why?

Not only "why does the indexing start over again?", but "why isn't there more documentation on what goes on inside the docker container?". "It runs its own bitcoind? I already have one running on the same machine, why another one?", "Why not more information on why the Esplora service is unavailable?", "Why isn't there a way to just restart the Esplora service inside the docker container?"