Stream closing as soon as opened

[DataKnox]

This issue began on March 8, 2024. See log output:

[2024-03-08T16:46:57.067975Z INFO  jupiter_core::geyser_client] stream opened
[2024-03-08T16:46:57.075502Z INFO  jupiter_core::geyser_client] subscribed to new accounts
[2024-03-08T16:46:57.075897Z ERROR jupiter_core::geyser_client] stream closed
[2024-03-08T16:46:57.577297Z INFO  jupiter_core::geyser_client] Retry to connect to subscribe to the geyser server
[2024-03-08T16:46:57.579563Z INFO  jupiter_core::geyser_client] subscribed to new accounts
[2024-03-08T16:46:57.579576Z INFO  jupiter_core::geyser_client] stream opened
[2024-03-08T16:46:57.585672Z ERROR jupiter_core::geyser_client] stream closed

It was working fine yesterday and overnight appears to have broken.

Jito Solana v1.17.22 Yellowstone geyser v1.13.0+solana.1.17.22

It appears to be related to Raydium as the following does not yield the error

RUST_LOG=info ./jupiter-swap-api --rpc-url http://127.0.0.1:8899/ --yellowstone-grpc-endpoint http://127.0.0.1:10000/ --snapshot-poll-interval-ms 50 --enable-new-dexes --exclude-dexes Raydium

[nika-efimova]

same problem

[burmisov]

hmm, experiencing the same issue with Solana 1.17.24, Yellowstone-GRPC 1.13+1.17.24 and jup-api 6.0.13

this is DEBUG level logs:

[2024-03-14T15:03:13.529797Z INFO  jupiter_core::geyser_client] subscribed to new accounts                                   
[2024-03-14T15:03:13.529804Z INFO  jupiter_core::geyser_client] stream opened   
[2024-03-14T15:03:13.536019Z DEBUG h2::codec::framed_write] send frame=Data { stream_id: StreamId(1) } 
..... multiple of these ^ v
[2024-03-14T15:03:13.536019Z DEBUG h2::codec::framed_write] send frame=Data { stream_id: StreamId(1) } 
[2024-03-14T15:03:13.536065Z DEBUG h2::codec::framed_read] received frame=Headers { stream_id: StreamId(1), flags: (0x5: END_HEADERS | END_STREAM) }                                                                                                      
[2024-03-14T15:03:13.536079Z DEBUG h2::codec::framed_read] received frame=Reset { stream_id: StreamId(1), error_code: NO_ERROR }                                                                                                                          
[2024-03-14T15:03:13.536088Z DEBUG h2::codec::framed_read] received frame=WindowUpdate { stream_id: StreamId(0), size_increment: 540672 }                                                                                                                 
[2024-03-14T15:03:13.536093Z DEBUG h2::codec::framed_read] received frame=WindowUpdate { stream_id: StreamId(0), size_increment: 360448 }                                                                                                                 
[2024-03-14T15:03:13.536093Z ERROR jupiter_core::geyser_client] stream closed 

Nothing particularly interesting on the Solana node side either:

[2024-03-14T15:21:24.502061184Z INFO  yellowstone_grpc_geyser::grpc] client #53: removed
[2024-03-14T15:21:25.004434662Z INFO  yellowstone_grpc_geyser::grpc] client #54: new
[2024-03-14T15:21:25.012579351Z INFO  yellowstone_grpc_geyser::grpc] client #54: removed
[2024-03-14T15:21:25.517151508Z INFO  yellowstone_grpc_geyser::grpc] client #55: new
[2024-03-14T15:21:25.527132852Z INFO  yellowstone_grpc_geyser::grpc] client #55: removed
[2024-03-14T15:21:26.029286870Z INFO  yellowstone_grpc_geyser::grpc] client #56: new
[2024-03-14T15:21:26.036787713Z INFO  yellowstone_grpc_geyser::grpc] client #56: removed

[dremovd]

I experience the same issue: ERROR jupiter_core::geyser_client] stream closed

[zxfgds]

same issue. yellowstone-grpc-geyser v1.13.0+solana.1.17.25

[Tongoko]

Is there a solution to this problem or maybe a temporary workaround (with Raydium)?

I'm having the same bug on:

Solana 1.17.28 yellowstone-grpc-geyser-1.14.1+solana.1.17.28 jupiter v6.0.15

[Arrowana]

Make sure you are running the geyser plugin with max_decoding_message_size_default set to 8MB

https://github.com/rpcpool/yellowstone-grpc/pull/288

I'll add this to the docs

[burmisov]

Doesn't seem to help in my case:

Here's a part of my yellowstone config:

{
    ...
    "grpc": {
        ....
        "max_decoding_message_size": "8_338_608"
    }
}

Is there any way to debug it further? Thanks.

Or, is it necessary to build the plugin with specifically the "default" set to 8MB and not just change the config file?...

UPD: maybe there's a suggested full yellowstone config file, I'm guessing other config params could be affecting? Though I think I've turned off all the limitations there.

[burmisov]

upgrading to the latest jup api version finally fixed it for me

[Daice]

To set up your own Jupiter-swap API, what are the requirements for the self hosted RPC? Does the RPC need to have account-index?