empty replay for csrf-token

[shihui-tang]

• Step1: install pulsar-manager with docker with version v0.3.0

docker pull apachepulsar/pulsar-manager:v0.3.0
docker run -it \
    -p 9527:9527 -p 7750:7750 \
    -e SPRING_CONFIGURATION_FILE=/pulsar-manager/pulsar-manager/application.properties \
    apachepulsar/pulsar-manager:v0.3.0

• Step2: access ui with url: http://localhost:9529

-- 2.1: login in with user pulsar/pulsar -- 2.2: login in response with: 404

• Step3: get token

-- 3.1 curl http://localhost:7750/pulsar-manager/csrf-token -- 3.2 curl: (52) Empty reply from server

Question is:

What's wrong with my steps or anything wrong with pulsar-manager v0.3.0 ?

[iamnivekx]

1. Run the pulsar-manager

docker run --rm -it \
    -p 9527:9527 -p 7750:7750 \
    -e SPRING_CONFIGURATION_FILE=/pulsar-manager/pulsar-manager/application.properties \
    apachepulsar/pulsar-manager:v0.3.0

2. get token, exec container ca632f969680

docker exec -it ca632f969680 sh
curl http://127.0.0.1:7750/pulsar-manager/csrf-token

=> b29ba8fc-be3e-4604-b2f3-9e954a91f105

It works fine for me.

[Kurounin]

You're not doing anything wrong. The backend service that listens to port 7750 takes around 2 minutes after the start of the container to begin responding to requests.

[dmx-on]

Hi. I'm under the impression that the pulsar-manager project is simply dead. It's quite surprising that in the year 2023, a basic web app takes 2 minutes or even longer to start. In my local Docker environment, it starts in under one minute about 1 in 10 times, but on other occasions, it does not start in more than 20 minutes with the CPU usage consistently at 100%. It would be advisable to have at least a health endpoint included. For example:

healthcheck: test: ["CMD", "curl", "-f", "http://localhost:7750/pulsar-manager/csrf-token"]