Algorand Node on VM cannot sync - issue with space?

General
1

Hello everyone,

I am running an Algorand node on a VM on AWS and have encountered an issue. I couldn’t seem to connect to the node since the goal node status -d /var/lib/algorand would give this output:
Get "http://127.0.0.1:8080/versions": dial tcp 127.0.0.1:8080: connect: connection refused

It seems that I ran out of space. My VM currently has 8 GB of space, which seems to be enough in the requirements. Is this normal or should I adjust some settings to stop using so much space? How much space would definitely be enough for this not to happen again?

2

8GB of disk space seems a bit too small.
For a non-archival node, 40GB is somewhat necessary.

You can check using df -h how much space is left.

In addition, the disk needs to be a SSD, and you need to have at least 4GB of RAM.

3

Hi @fabrice,

Thank you for pointing this out - sorry I wasn’t specific enough when describing my VM: it has 8 GB of RAM and a general purpose SSD with 100 IOPS. Do you think this is an issue with an SSD then?

Thanks and sorry for not being clear the first time!

4

100 IOPS seems extremely low. Are you sure it’s really 100 IOPS?

To help debug, please provide the output of the following (assuming Algorand software is in the PATH):

export ALGORAND_DATA=... the correct path
goal node status
goal version -v
algod -v
cat "$ALGORAND_DATA/algod-err.log"
cat "$ALGORAND_DATA/algod-out.log"
cat "$ALGORAND_DATA/config.json"
tail -n 50 "$ALGORAND_DATA/node.log"
df -h
5

Hi Fabrice,

Just curious… Do algod-err.log and algod-out.log exist if there have been no errors?

My node has only two log files:

[/var/lib/algorand] $ ls *.log
node.archive.log node.log

6

It’s possible they don’t exist.
For sure they don’t contain meaningful information if your node started properly.