Issues deploying Typesense to AWS EKS

Hi,

I deployed typesense to AWS EKS and it /health was giving {ok:true}

But when I tried to access, curl 'http://xyz:8108/keys' -X GET -H "X-TYPESENSE-API-KEY: MySuperstrongpassword#6876876578", all my pods went into Error. Here are the pod logs:

> kubectl logs typesense-dev-2                   
I20220803 10:20:41.334745     1 typesense_server_utils.cpp:352] Starting Typesense 0.23.1
I20220803 10:20:41.334774     1 typesense_server_utils.cpp:355] Typesense is using jemalloc.
I20220803 10:20:41.334929     1 typesense_server_utils.cpp:404] Thread pool size: 16
I20220803 10:20:41.336777     1 store.h:61] Initializing DB by opening state dir: /usr/share/typesense/data/db
I20220803 10:20:41.348294     1 store.h:61] Initializing DB by opening state dir: /usr/share/typesense/data/meta
I20220803 10:20:41.359787     1 typesense_server_utils.cpp:486] Starting API service...
I20220803 10:20:41.359961     1 http_server.cpp:177] Typesense has started listening on port 8108
I20220803 10:20:41.359997   103 batched_indexer.cpp:120] Starting batch indexer with 16 threads.
I20220803 10:20:41.360364   103 batched_indexer.cpp:126] BatchedIndexer skip_index: -9999
I20220803 10:20:41.363009   102 server.cpp:1045] Server[braft::RaftStatImpl+braft::FileServiceImpl+braft::RaftServiceImpl+braft::CliServiceImpl] is serving on port=8107.
I20220803 10:20:41.363023   102 server.cpp:1048] Check out  in web browser.
E20220803 10:20:41.391083   102 raft_server.cpp:169] Unable to resolve host: typesense-0.ts.typesense.svc.cluster.local
E20220803 10:20:41.407008   102 raft_server.cpp:169] Unable to resolve host: typesense-1.ts.typesense.svc.cluster.local
E20220803 10:20:41.415683   102 raft_server.cpp:169] Unable to resolve host: typesense-2.ts.typesense.svc.cluster.local
E20220803 10:20:41.415729   102 configuration.cpp:43] Fail to parse typesense-0.ts.typesense.svc.cluster.local:8107:8108
E20220803 10:20:41.415732   102 raft_server.cpp:51] Failed to parse nodes configuration: `typesense-0.ts.typesense.svc.cluster.local:8107:8108,typesense-1.ts.typesense.svc.cluster.local:8107:8108,typesense-2.ts.typesense.svc.cluster.local:8107:8108` --  will retry shortly...

Typesense deployment on k8s has some quirks because the raft library we are using supports only raw IPs. See this comment and the comment above in the issue for details: https://github.com/typesense/typesense/issues/465#issuecomment-1173536082

So, what is the best way to deploy typesense?

It's easy to deploy on plain EC2 instance(s)

But how do we achieve high availability?

But, can you help me figure out the issue that I;m facing with K8s?

I followed https://raw.githubusercontent.com/typesense/typesense-kubernetes/master/base/install.yaml

Sorry we don't have operational experience with k8s. Whatever work has been done in terms of chart etc. is community driven.

Okay, based on this, what is the API Key?

API key is a key you start Typesense server with which is used for authentication

I got my HELM working. Thanks for the help Kishore Nallan