[go: up one dir, main page]

Cannot utilize new featue to keep additional conext in rolling mode history

Environment (Mainnet, test network, build from source, ...)

Mainnet, v10.2 5bfd311b

Summary

Failure when running the tezos node the first time after updating the config mode to change from a simple rolling node to a rolling node that keeps an additional 15 cycles of context

Expected behavior

expected the node to run as usual and to begin keeping more contexts

Actual behavior

I get a message that I need to use --force-history-mode-switch to force it to do so, but then when I use that additional option i get: Error: Failed to get the nth predecessor of BMTrxPyWjYGGbW5HQ9ByhiQfQy4se5gG2T5kAXnxJXt8QjpxTCz. The offset is invalid: 47386. What actually happened.

Steps to reproduce

From a rolling mode node, stop it use config update to specify --history-mode=rolling:15 start the node back with --force-history-mode-switch

History mode

rolling