found a hole in a oneblock files

[adu-web3]

I'm trying to syncing the eos mainnet , and all components were started and all seemed working in a good condition until encountering this error:
found a hole in a oneblock files (merger/bundle.go:90) {"bundle_lower_block": 1784100, "missing_block_num": 1784169, "missing_block_id": "1f870f97"}

By checking the dfuse.log.json , I find blockfiles /home/adu/eostest/dfuse-data/storage/merged-blocks/0001784100.dbin.zs was not created :



indexer got stuck

[sduchesneau]

Hi, from the block ID I see that you are trying to sync EOS Mainnet from the start.
This requires a few customization flags, it does not work out of the box.

I suggest that you first read " Sync'ing a large chain #26 " and try with the "phase1.yaml" approach. The "merge-and-store-directly" flag is key, and also the fact that you need to run mindreader instance separately from the rest to improve your chance of processing 128 million blocks without a hiccup (Most of the dfuse components are not meant to run on a chain that is not "ready" (already synced))
Note that "merge-and-store-directly" does not support "stopping and restarting" the process, so if it stops, you will have to restore your nodeos from a snapshot.

Your mindreader node is probably broken (dirty state), and you can see this with "dfuseeos start mindreader -vvv", you will have to restore it from a state snapshot.

Closing this issue, because you are trying to do something that is "guaranteed to not work" (as discussed in issue #26)