Clean up data
Prune state
- Go to your docker-compose dir, adding the entrypoint for overwriting the container entrypoint
node:
image: ${NODE_IMAGE}
stop_grace_period: 5m
hostname: node
container_name: node
.....
entrypoint: ronin snapshot prune-state --datadir /ronin/data
....
ports: ...
- Run
docker-compose up -d
Example log:
docker logs node -f --tail 5
INFO [05-08|07:11:01.990] Initialized state bloom size=2.00GiB
INFO [05-08|07:11:02.895] Selecting bottom-most difflayer as the pruning target root=61d5f3..91e715 height=34,440,527
INFO [05-08|07:11:03.037] Deleted trie clean cache path=/ronin/data/ronin/triecache
INFO [05-08|07:11:03.052] Iterating state snapshot accounts=0 slots=0 elapsed="674.358µs"
INFO [05-08|07:11:11.052] Iterating state snapshot accounts=75398 slots=785 elapsed=8.000s eta=32m41.227s
- After the container process is finished, then remove the line added in step 1 and start the node again.
docker-compose up -d
Replace by Snapshot
- Download the latest snapshot link following this repo
cd /root/ronin/chaindata/data/ronin/snapshot
wget -q -O - <snapshot URL from the README file in the repo> | tar -I zstd -xvf -
- Stop your node
docker-compose down
- Remove the old chain data and replace it with snapshot
rm -rf chaindata
mv snapshot/chaindata ./
- Start your node
docker-compose up -d
tip
- Prune state will create a huge downtime. With current data size of Ronin, it will take more than one hour for finishing this progress. We suggest to use the snapshot (if you have enough disk space to download the snapshot) to reduce downtime.
- We should put a disk size alert with threshold 30% disk size left for starting the progressing maintenance for replacing disk( 4-5 months one time).