Bad hdd scanning

Hi team,

this scanning for the whole disk data on every startup is so bad.
Any ideas or startup options to avoid this?
It takes so long and makes cheap hosts / disks unresponsible…


That’s how the Storj software works. There is no way to disable it.
That’s how it knows what’s stored, and is particularly useful when/if something goes wrong with databases.

However, maybe it’d be cool for this filewalker process to run only when DBs are KO, or if the node did not stop correctly, I dunno.

I do agree it’s a massive pain, especially on big nodes running on SMR drives… :confused:


Make sure on Linux systems to mount with the noatime option. This will ensure that it’s only reads and doesn’t have to update the access time every time. This helps a lot to speed that part up.

But it would definitely be nice if this could be limited so it doesn’t run every time, but only after an interval since the last run has expired.


I am using noatime but it is still bad…