Storagenode crashes instantly with allmost zero logs

[HGPlays]

Hello - last night a node went down - and i have not been able to get it back up.

I checked the following:
Speedtested the disk - its ok it seems.
Log level set to info
Checked for some currupted cache or database files, could not find this.

Logs are only these lines before crash:

2025-09-17T08:40:58+02:00 INFO Configuration loaded {“Location”: “C:\Program Files\Storj10\Storage Node\config.yaml”}
2025-09-17T08:40:58+02:00 INFO Anonymized tracing enabled
2025-09-17T08:40:58+02:00 INFO Operator email {“Address”: “xxx”}
2025-09-17T08:40:58+02:00 INFO Operator wallet {“Address”: “xxxx”}

I have 1.136.4
I can accces, read, write to disk fine.
Maybe someone here knowes more than me?

Best, HG

[Alexey]

Perhaps related to this, if you migrated to hashstore:

Windows GUI node consume big amount of virtual memory

I discovered that for some reason my server with 17 nodes converted to hashstore, nodes turn off several time a day without log entry why node is off. With grate help of littleskunk and Alexey we discovered that Error is Out of memory. So I started to monitor this stuff, and even when i have half of RAM free, nodes are steel go offline without log entry. My discovery that node allocate lot of virtual memory. [Screenshot 2025-09-11 152031] looks like node directly not use this memory but it …

Otherwise you may try to run from the command line, but you need to provide a path to at least the config file, maybe also identity:

"C:\Program Files\Storj10\Storage Node\storagenode.exe" run --config-dir "C:\Program Files\Storj10\Storage Node\config.yaml"

[HGPlays]

I have tried a restart of the entire system.
The system has 32GB of ram available at the moment.

[Alexey]

It’s not about RAM, it’s about a virtual memory - the swap file is too small and the node likely fail with the Out of memory error.
You may see the exact error, if you would run it from the second cmd.exe run in the first one cmd.exe (because when this happen, it also exit the shell). But in the case with a double cmd.exe the first one will remain.

[HGPlays]

Hi Alexey,

Thanks for the help! I followed your suggestion and ran the node from a double cmd window so the crash message would stay visible. That revealed a Badger initIndex panic.

After some testing I found out it wasn’t orders.db or piece_expiration.db in my case, but actually the filestatcache database that was corrupted. Once I moved that out, the node started up normally and is now running fine.

Might be useful info for others hitting the same type of crash.

Best,
HG

[Alexey]

It is always was. It helps also when you specified a wrong path for the logs, for example or specified double options.