High cpu load caused by storj containers

[Balage76]

I see the same on my windows nodes.
Now I set retain.concurrency: 1, lazy FW false and startup scan true for one of the four nodes. Let’s see if it will end up again at the mentioned 100% CPU core use.
BTW, all of a sudden (within 2 hours after the restart) the node deleted 300+ GB of trash out of 1,2TB, which was there for more than a week.

So far it is ok, even though I see two BF in the retain directory:

[Alexey]

@gingerbread233

I wanted to know what that means, the link to github doesn’t work

Release preparation v1.108

This filter is now not hardcoded and can be changed to another value, like /25 in that unit test.

@gingerbread233

So you mean, I should “fix it” by editing the config.yaml, and set value 5 to 1 by myself?

for the 1.105.x - yes. In the 1.108.x it’s changed to 1 by default

Disk usage discrepancy?

In 1.108 it was changed to 1

[gingerbread233]

Changing “retain.concurrency” to 1 seem to have fixed the issue. It’s almost 24 hours, and the CPU load has not increased as before (after around 12 hours).

[Alexey]

I think, that the high CPU usage (stalling of some filewalkers…) is fixed in 1.108.x