If you are processing old block ranges with the RPC poller + merger and want to chunk up the block range into separate jobs, you need to overlap the different jobs by sufficient blocks to cover the lib. Else the last set of one-blocks will not be merged.
Add option --lib 0 or something to force the lib to be 0 instead of the default 200.
Or maybe just set the lib to 1 block automatically if the head_block_drift > 6 hrs.
Current one-block filenames show 200 block LIB, example:
If you are processing old block ranges with the RPC poller + merger and want to chunk up the block range into separate jobs, you need to overlap the different jobs by sufficient blocks to cover the lib. Else the last set of one-blocks will not be merged.
Add option
--lib 0
or something to force the lib to be 0 instead of the default 200.Or maybe just set the lib to 1 block automatically if the head_block_drift > 6 hrs.
Current one-block filenames show 200 block LIB, example: