You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
After upgrading to 2.1.9, our validator has experienced significant performance degradation related to staked connections (SWQoS). The primary issue observed is not being able to stablish a estable QUIC connection between the RPC and Validator, which return the following error log on the RPC side:
Note: Once the Validator is rolled back to 2.0.22, the normal behaviour is immediately established.
Proposed Solution
No actual proposal beyond assessing what is triggering this issue in 2.1.x, however I would urge to identify the root cause before Solana Foundation call the 2.1.x for the whole cluster or when 2.0.x is forked, as it is widely used by major RPC providers.
The text was updated successfully, but these errors were encountered:
StaRkeSolanaValidator
changed the title
Performance degradation for staked connections (SWQOS) in version 2.1.9
Performance degradation for staked connections (SWQoS) in version 2.1.9
Jan 28, 2025
Hey @StaRkeSolanaValidator can you tell me more about how your swqos partner is sending txs? What client are they using? Are they using a custom client? Is it open source so I can review it?
Hey @StaRkeSolanaValidator can you tell me more about how your swqos partner is sending txs? What client are they using? Are they using a custom client? Is it open source so I can review it?
Not sure the client they are using, I will dig into it. Thanks for replying!
Problem Description
After upgrading to 2.1.9, our validator has experienced significant performance degradation related to staked connections (SWQoS). The primary issue observed is not being able to stablish a estable QUIC connection between the RPC and Validator, which return the following error log on the RPC side:
"ERR: Application error 0x5 (remote): invalid_stream"
Validator details:
Note: Once the Validator is rolled back to 2.0.22, the normal behaviour is immediately established.
Proposed Solution
No actual proposal beyond assessing what is triggering this issue in 2.1.x, however I would urge to identify the root cause before Solana Foundation call the 2.1.x for the whole cluster or when 2.0.x is forked, as it is widely used by major RPC providers.
The text was updated successfully, but these errors were encountered: