Several pods stuck in CrashLoopBackoff

[Evan-Berryman]

Summary

I am attempting to run train ticket with Jaeger. However, most of the site is inaccessible due to pods being unable to connect to services and crashing.

Expected behaviour

All services should be accessible and pods shouldn't crash

Current behaviour

auth-service, config-service, consign-price-service, contacts-service, food-map-service, inside-payment-service, order-other-service, order-service, payment-service, price-service, route-service, security-service, station-service, train-service, travel-service, travel2-service, and user-service have all been crashing and restarting. They seem to be unable to connect to various parts of the API. When using the site, attempting to search returns a 502 error for travelservice/trips/left, logging in gets a 502 from users/login, and the advanced search returns a 500 for its options. It seems like something was not connected properly but I'm not sure what.

Steps to reproduce

The installation instructions weren't super clear to me. I want to use Jaeger, so first I ran the kubectl apply command on the Jaeger deployment files. I later used the make deploy command to add the naocs pods. This strange deployment method may be the cause of the errors, but I also remember having similar issues when I first installed train ticket with just the basic make deploy.

Your environment

OS(e.g: cat /etc/os-release): Ubuntu 22.04.2 LTS

docker version(e.g docker version): 4.16.2

Kubernetes version(use kubectl version): Client: 1.26.3, Server: 1.25.4

Additional context

I've heard resources could be an issue, but currently Docker has 7 CPUs and 27.3 gigs of RAM, and I don't have much more to give.

[Deep-Yellow]

Use the kubectl logs command to check if these pods are properly connected to the database. If not, an UnknownHostException error will appear, and you can try to redeploy TrainTicket under the default namespace

[Evan-Berryman]

I got an error, but it was an IllegalStateExcpetion that occurred right after a service tried to connect to Mongo. This message stood out to me:

Caused by: org.springframework.data.mongodb.UncategorizedMongoDbException: Query failed with error code 352 and error message 'Unsupported OP_QUERY command: find. The client driver may require an upgrade. For more details see https://dochub.mongodb.org/core/legacy-opcode-removal' on server ts-user-mongo:27017; nested exception is com.mongodb.MongoQueryException: Query failed with error code 352 and error message 'Unsupported OP_QUERY command: find. The client driver may require an upgrade. For more details see https://dochub.mongodb.org/core/legacy-opcode-removal' on server ts-user-mongo:27017

It looks like Mongo is out of date, but I just pulled the newest image for Docker. Is the Train Ticket application trying to use deprecated commands, or is there something else I could be missing?

[Evan-Berryman]

Update: It was an issue with Mongo. In the ts-deployment-part1.yaml file, the Mongo services need to pull the mongo:5.0.15 image, and not the latest.