openzipkin / zipkin-helm Goto Github PK

I created this repo from a combination of git commands to preserve history. The paths are no longer correct and may need some buffing. I'm not sure who's best to lead, but I think @jonkerj has been on a lot of threads. also pinging @optional303 in case there is interest from FT.

The main goal is to get the charts working, and some way documented to access them, ideally something like our docker examples in the main repo.

When things are working even at a basic level, remove the redundant resources from the main zipkin repo

cc'ing some org volunteers @reta @anuraaga and @shakuzen as well.

Feature

Add elastic search option to manage configuration like ES_HOSTS or ES_INDEX.

ES_SSL_NO_VERIFY
ES_USERNAME
ES_PASSWORD

Rationale

Without the options, all the scenario was not covered.

Describe the Bug

The pod don't start beacuse readines probe with error 503. I have configured elasticsearch env variables to connect to a elastic 8.2 cluster. The log ends with

2022-05-23 10:12:49.392  INFO 1 --- [           main] c.l.a.c.Flags                            : Using OpenSSL: BoringSSL, 0x1010007f
2022-05-23 10:12:50.678  INFO 1 --- [           main] c.l.a.s.d.DocStringExtractor             : Using com.linecorp.armeria.thrift.jsonDir: META-INF/armeria/thrift
2022-05-23 10:12:50.720  INFO 1 --- [           main] c.l.a.c.u.SystemInfo                     : Hostname: zipkin-server-75d9b5b4f8-mshps (from /proc/sys/kernel/hostname)
2022-05-23 10:12:51.133  INFO 1 --- [oss-http-*:9411] c.l.a.s.Server                           : Serving HTTP at /0.0.0.0:9411 - http://127.0.0.1:9411/
2022-05-23 10:12:51.138  INFO 1 --- [           main] c.l.a.s.ArmeriaAutoConfiguration         : Armeria server started at ports: {/0.0.0.0:9411=ServerPort(/0.0.0.0:9411, [http])}
2022-05-23 10:12:51.210  INFO 1 --- [           main] z.s.ZipkinServer                         : Started ZipkinServer in 6.383 seconds (JVM running for 8.539)
2022-05-23 10:16:09.208  INFO 1 --- [orker-epoll-2-1] c.l.a.i.JavaVersionSpecific              : Using the APIs optimized for: Java 9+

So I guess it starts correct but I can't see more errors to fix the problem.

I use this image version:
openzipkin/zipkin:2.17.1

Please, add cassandra storage support when deploying zipkin using helm chart.

Right now zipkin deployments by default avoid istio pod injection, while this is right or not we should be able to add some rationale to such behaviour.

https://github.com/openzipkin/zipkin/blob/master/charts/zipkin/values.yaml#L37

Ping @jonkerj @chirauki