Ancient Data Prune is a new feature in bsc v1.1.8
Special thanks to BNB48Club on contributing another dump of snapshot, you can also refer here to download.
Step 1: Preparation
- Make sure your hardware meets the suggested requirement.
- A disk with enough free storage, at least twice the size of the snapshot.
Step 2: Download && Uncompress
- Copy the above snapshot URL.
- Download:
wget -O geth.tar.lz4 "<paste snapshot URL here>". It will take one or two hours to download the snapshot, you can put it in the backgroud bynohup wget -O geth.tar.gz "<paste snapshot URL here>" &
If you need to speedup download, just use aria2c
aria2c -o geth.tar.lz4 -s14 -x14 -k100M https://download.bsc-snapshot.workers.dev/{filename}
But aria2c may fail sometimes, you need to rerun the download command. To make it convient, you can use the follow script
#!/bin/bash
if [ $# -eq 1 ]; then
dir=$(pwd)
elif [ $# -eq 2 ]; then
dir=$2
else
echo "Usage: $0 <uri> [filepath] "
exit 1
fi
uri=$1
filename=$(basename "$uri")
status=-1
while (( status != 0 ))
do
PIDS=$(pgrep aria2c)
if [ -z "$PIDS" ]; then
aria2c -d $dir -o $filename -s14 -x14 -k100M $uri
fi
status=$?
pid=$(pidof aria2c)
wait $pid
echo aria2c exit.
case $status in
3)
echo file not exist.
exit 3
;;
9)
echo No space left on device.
exit 9
;;
*)
continue
;;
esac
done
echo download succeed.
exit 0
- Uncompress:
tar -I lz4 -xvf geth.tar.lz4. It will take more than two hours to uncompress. You can put it in the background bynohup tar -I lz4 -xvf geth.tar.lz4 & - You can combine the above steps by running a script:
wget -O geth.tar.lz4 "<paste snapshot URL here>"
tar -I lz4 -xvf geth.tar.lz4
- If you do not need to store the archive for use with other nodes, you may also extract it while downloading to save time and disk space:
wget -q -O - <snapshot URL> | tar -I lz4 -xvf -
Step 3: Replace Data
- First, stop the running bsc client if you have one by
kill {pid}, and make sure the client has shut down. - Consider backing up the original data:
mv ${BSC_DataDir}/geth/chaindata ${BSC_DataDir}/geth/chaindata_backup; mv ${BSC_DataDir}/geth/triecache ${BSC_DataDir}/geth/triecache_backup - Replace the data:
mv server/data-seed/geth/chaindata ${BSC_DataDir}/geth/chaindata; mv server/data-seed/geth/triecache ${BSC_DataDir}/geth/triecache - Start the bsc client again and check the logs
erigon version v1.0.8
For more granular upload & download to avoid big files error, split the files into several chunks, so please download them together and concatenate finally.
Step 1: Preparation
- Make sure your hardware meets the suggested requirement.
- BSC Archive: 8TB. BSC Full: 2TB.
Step 2: Download && Concatenate && Uncompress
sudo yum install aria2c
aria2c -s14 -x14 -k100M https://pub-60a193f9bd504900a520f4f260497d1c.r2.dev/chunk/{filename}
cat /chunk* > mdbx.tar.lz4
tar -I lz4 -xvf mdbx.tar.lz4
Step 3: Replace Data And Restart erigon
- Stop the running erigon client by
kill {pid} - Backing up the original data:
mv ${erigon_datadir}/chaindata/mdbx.dat ${erigon_datadir}/chaindata/mdbx.dat - Replace the data:
mv ${erigon_snapshot_dir}/erigon/chaindata/mdbx.dat ${erigon_datadir}/chaindata/mdbx.dat - Start the erigon client again and check logs
./build/bin/erigon --sentry.drop-useless-peers --p2p.protocol=66 --txpool.disable --metrics.addr=0.0.0.0 --log.console.verbosity=dbug --db.pagesize=16k --datadir ${erigon_dir/data} --private.api.addr=localhost:9090 --chain=bsc --metrics --log.dir.path ${erigon_dir/log}