Closed andreclaro closed 1 month ago
iostat nvme1n1
Linux 6.2.0-1012-aws (m-xlayer-01) 07/17/24 _x86_64_ (4 CPU)
avg-cpu: %user %nice %system %iowait %steal %idle
31.60 0.01 2.53 14.24 0.00 51.61
Device tps kB_read/s kB_wrtn/s kB_dscd/s kB_read kB_wrtn kB_dscd
nvme1n1 846.55 49378.88 56419.74 0.00 115888644449 132413039148 0
it was quicker when I switched to a bigger instance type, from m6a.xlarger
to m6a.4xlarger
(64 Gb of memory)... probably due to the memory requirements of this step 🤔
Description
I have noticed this during X Layer testnet upgrade, where we had to start the node from a snapshot (1 day old)...
Today, I'm working on the mainnet upgrade and we also had to restart the node from a snapshot (2 days old), so node needs to download old blocks.
The node was started at 10:27:13 UTC and it still at
[9/16 IntermediateHashes] SMT regenerate progress: (58%)
now (12:15:30)I don't see any bootleneck (CPU, Memory or disk)...
System information
Cdk-erigon version: v1.1.5.8
OS & Version: Ubuntu
Node Config: https://github.com/okx/Deploy/blob/main/mainnet/setup-erigon-rpc.md
Steps to reproduce the behaviour
Follow https://github.com/okx/Deploy/blob/main/mainnet/setup-erigon-rpc.md
Full Node Logs