Unable to see messages logged to STDERR in docker logs

Community Support
,
1

I try to docker run temporalio/auto-setup:1.18.1, it failed as unable to initialize system namespace, but i can’t see the error detail with docker logs as the error detail is output to STDERR. Any help?

2

Can you show your docker command?
Here is docker compose that maybe helps.

Also would use auto-setup only for local development/testing. Auto-setup starts up all 4 server roles in same process on same deployment so it’s really not suited for anything prod. If you wanted to try starting each service role in separate containers we have example here.

3

thx for reply, i use auto-setup just for convenience to initialize cassandra schema and create default namespace. Below is my docker command:

docker run -e CASSANDRA_SEEDS=xxxx \
      -e CASSANDRA_USER=xxx \
      -e CASSANDRA_PASSWORD="***" \
      -e KEYSPACE=temporal \
      -e VISIBILITY_KEYSPACE=temporal_visibility \
      -e CASSANDRA_REPLICATION_FACTOR=2 \
      -e DEFAULT_NAMESPACE_RETENTION=7 \
      -e LOG_LEVEL=debug,info \
      -e SKIP_SCHEMA_SETUP=true \
      -e SKIP_DEFAULT_NAMESPACE_CREATION=false \
      -e SKIP_ADD_CUSTOM_SEARCH_ATTRIBUTES=true \
      temporalio/auto-setup:1.18.1

Here is the error log, but nothing about why it failed to create the temporal-system namespace.

{"level":"fatal","ts":"2022-10-20T12:30:36.894Z","msg":"temporal-system namespace does not exist","service":"worker","error":"Namespace temporal-system is not found.","logging-call-at":"service.go:519","stacktrace":"go.temporal.io/server/common/log.(*zapLogger).Fatal\n\t/home/builder/temporal/common/log/zap_logger.go:151\ngo.temporal.io/server/service/worker.(*Service).ensureSystemNamespaceExists\n\t/home/builder/temporal/service/worker/service.go:519\ngo.temporal.io/server/service/worker.(*Service).Start\n\t/home/builder/temporal/service/worker/service.go:361\ngo.temporal.io/server/service/worker.ServiceLifetimeHooks.func1.1\n\t/home/builder/temporal/service/worker/fx.go:139"}

But now i got the reason throught local debug: “Invalid BATCH message type LOGGED, do not support this type batch”.

4

I knonw why now. I am not using the standard cassandra, but GaussDB (for Cassandra) from huaweicloud, and logged batch is not supported.