Closed palango closed 5 years ago
@palango Actually you're correct, it's not matrix related. I misread the traceback that was posted in Gitter.
The most likely cause of a read timeout is that the node takes too long to generate a response to whatever was requested (likely a filter in this case).
The most likely cause of a read timeout is that the node takes too long to generate a response to whatever was requested (likely a filter in this case).
And the cause of this is probably that the whole blockrange from last_known_block
to 'latest'
is queried at once, whgich can get quite big on the first node start and leads to the timeout of the connection.
This happened again to some people on the EthCC booth. Can someone on linux try to reproduce this?
I will pick this up and fix asap.
Problem Definition
I was able to reproduce a problem forwarded to us. I only see this problem on mainnet, but I'm currently not sure if it's related to the
production
flag (that one works fine on rinkeby testnet) or somehow related to the matrix server (@ulope guess) or that we try to sync to big intervals (I had the problem of web3 requests timing out for the MS).System Description