I had been running a several different nodes concurrently on my machine for a few months, and recently decided to add Turbo-geth to the mix. Since launching, my vpn connection has been totally really unstable and my other nodes have been unreliable. I got to the point where I shut down all of the nodes except for Turbo-geth and Prysm, assuming that maybe my SSD just couldn’t write fast enough to run Turbo-geth, IDChain, geth, and Goerli Geth at the same time. (I was never able to get both geth and turbo-geth in sync at the same time). But even with just Turbo-geth and Prysm running, the instability persists.
Has anyone else experienced this?
Any suggestions on what might be causing it?
Machine specs:
2x Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz
32GB DDR4 Ram
4TB Raid 0 SSD
1Gb fiber internet connection
I have not personally done such a resource intensive configuration on my DAppNode.
Let’s see if someone in the forum can help you, or maybe you can also post this in our Discord to promote it - looks like an interesting case.
Meanwhile I wonder if you have tried to find something in Turbo-Geth logs, or VPN core package logs that may shed a light on what is causing the instability.
Also, you should install DMS, and DAppNode Exporter to have a host and packages monitor system running. Check this tutorial in case you need to:
Meanwhile I wonder if you have tried to find something in Turbo-Geth logs, or VPN core package logs that may shed a light on what is causing the instability.
I’m honestly not sure what I’d be looking for.
I have both of these running.
The machine seemed to be handling the load quite well, CPU was never over 10% used, memory never over 50%, and plenty of storage.
There were some times when the write speeds spiked during Turbo-geth syncing, but once it was in sync the write speeds were well below max.
I scrapped Geth and started syncing Open Ethereum instead. It got through the snapshot phase and is now trying to catch up (stuck on 94% for a few days). I switched Turbo-geth back on two days ago, so far it’s been much more stable and is very close to being back in sync, about 50k blocks behind right now. Will see what happens if/when it catches up to the chain head and if OE can ever catch up.