Closed andreclaro closed 3 weeks ago
andreclaro
commented
3 weeks ago I also noticed the warning Pivot seemingly stale, moving message several times in the logs:
geth[3802696]: WARN [06-03|19:09:32.831] Pivot seemingly stale, moving old=15,326,091 new=15,326,148I have just tried using io2 with max iops and I don't see improvements...
andreclaro
commented
3 weeks ago I was able to spin-up the node using snap sync after cleaning the db and using an instance type with more cpu and memory.... Example for base mainnet:
op.geth:
/usr/local/bin/geth \
--op-network=base-mainnet \
--datadir=/var/lib/base/data \
--syncmode=snap \
--gcmode=archive \
--networkid="8453" \
--http \
--http.addr=0.0.0.0 \
--http.port=8545 \
--http.vhosts="*" \
--http.corsdomain="*" \
--http.api=web3,debug,eth,net,engine,geth \
--ws \
--ws.port=8546 \
--ws.addr=0.0.0.0 \
--ws.origins="*" \
--ws.api=debug,eth,net,engine \
--authrpc.addr=127.0.0.1 \
--authrpc.port=8551 \
--authrpc.jwtsecret=/var/lib/base/jwt.txt \
--authrpc.vhosts="*" \
--metrics \
--metrics.addr=0.0.0.0 \
--metrics.port=6060 \
--maxpeers=500 \
--port=30303 \
--discovery.port=30303 \
--rollup.disabletxpoolgossip=true \
--rollup.sequencerhttp=https://mainnet-sequencer.base.org \
--rollup.superchain-upgrades \
--rollup.halt=major \
--verbosity=3op-node:
/usr/local/bin/op-node \
--syncmode=execution-layer \
--l1.trustrpc \
--l1.rpckind=basic \
--l1=http://ETH_URL:8545 \
--l1.beacon=http://ETH_URL:5052 \
--l2=http://127.0.0.1:8551 \
--rpc.addr=127.0.0.1 \
--rpc.port=9545 \
--l2.jwt-secret=/var/lib/base/jwt.txt \
--network=base-mainnet \
--p2p.peerstore.path=/var/lib/base/data/opnode_peerstore_db \
--p2p.priv.path=/var/lib/base/data/opnode_p2p_priv.txt \
--p2p.discovery.path=/var/lib/base/data/opnode_discovery_db \
--rollup.load-protocol-versions=true \
--rollup.halt=major
Type=simple
System information
op_geth: "v1.101315.0" op_node: "op-node/v1.7.5"
OS & Version: Linux (Ubuntu:
Expected behaviour
I was expected snap sync to take a few days not weeks or months.
Actual behaviour
"Syncing: state healing in progress" is taking a lot to finish.... Op-geth is running for 7h 34min and it only download
331,784accounts:Steps to reproduce the behaviour
Start nodes with snap sync enabled
Example of the command for
base mainnet... for optimism it was used a similar command:Op-geth cli:
op-node cli:
Logs
op-geth in optimism mainnet:
op-geth in base mainnet: