Virtual Node No Space Left

I’m running a vnode. It is no longer starting. I’m seeing this error where I try to start the docker image:

Error response from daemon: mkdir /var/lib/docker/overlay2/a9e44b6d43615ea2c06206e3dced342952b1b487bab64e09a1c7cd42af4114fc/merged: no space left on device
Error: failed to start containers: inc_mainnet

My machine appears to have used all 66.0 GB of hard drive space. According to the setup article I should only need 50 GB of space? How to host a Virtual Node

What are the requirements for a virual node? I can up the sized of my hard drive, but Is there something I should do to free up space? Are the minimum requirements from that article out of date?

I have this same exact problem, I thought it was just me.

The setup guide is half a year old. The data of main net is now over 100GB already.
Here’s my node:
image
You must upgrade your storage.

1 Like

I don’t think that’s what’s going on. You are saying that its over 100GB now…well mine all topped out at 27GB before the drive was too full and as of a few days ago all my nodes were working just fine. So you are saying that it went from less than 27GB to over 100GB in a few days and that’s normal? I don’t think so…

I think @support needs to look in to this and get back to us.

hey @brico84

Rocky is one of support team members and what he said is true.
The blockchain data can be grown up to 100GBs a year since there is many transactions.
I suggest you upgrade your disk to 300GBs

1 Like

I forgot to mention that 100GB is the whole chain’s data (running your node as a full node)

If your node is a validator node, then the data would be less than that. Because the whole network has 8 shards plus beacons, a validator node only have to sync data of the shard that it is currently a committee of, the data of that node would be around 100/8~13GB + beacons data. But a node will not always be a committee of only one shard, thus it will have to sync data of other shards if needed. So at some point, your node will have data of the whole network which is a lot.

Bottom line: upgrading your node’s storage is a must.

Just a little tip: You can make some space by deleting the log files in the data folder if there’s any. It’s not much but the little things do mater sometime :smiley:

2 Likes

Deleted

1 Like

A validator node can also run with the flag --relayshards all, which will make it have data of the whole network, just like a full node, and the node will have certain advantages compared to other nodes. But that’s not the case.

This is the case:

Maybe the mentioned post on how to setup a vNode can be edited and have the requirements for a vNode stated.

1 Like