twoeths

>seems like this code is just copy-pasted from blob_sidecar, we do the same there @nflaig yes it's tracked in https://github.com/ChainSafe/lodestar/issues/7957

ok then the retry mechanism was already in place I think we only need metrics to see: - if the 1st call is success - if the 2nd call is...

in this case, nodes have to count on range sync and gossip does not work

as seen on node 2, the reason we can't proceed through gossip is `data_column_sidecar` messages being ignored the reason is only parent_unknown

stale issue, this is fusaka-devnet-4

this was tested in lodestar, see https://github.com/ChainSafe/lodestar/pull/7171#issuecomment-2425472397

prefer to get this PR #409 to be reviewed first

thanks @enyinnaya1234 for looking into this

current error: ``` May-31 08:34:21.148[network] verbose: Req error method=beacon_blocks_by_range, version=2, encoding=ssz_snappy, client=Lighthouse, peer=16...cwJnrM, requestId=269, code=REQUEST_ERROR_DIAL_TIMEOUT Error: REQUEST_ERROR_DIAL_TIMEOUT at new LodestarError (/usr/src/lodestar/packages/utils/lib/errors.js:6:9) at new RequestError (/usr/src/lodestar/packages/reqresp/lib/request/errors.js:39:9) at (/usr/src/lodestar/packages/reqresp/lib/request/index.js:63:27) at processTicksAndRejections (native:7:39)...

another error that could be related is ``` May-30 00:00:16.332[sync] ^[[36mverbose^[[39m: Batch download error id=Finalized, startEpoch=15495, status=Downloading, peer=16...nywWEc, code=REQUEST_ERROR_DIAL_ERROR, error=unexpected end of input Error: unexpected end of input at new...