Thermi

Well, I don't know that, because I'm not that intimiate with Golang to figure that out.

A time stamp file is maintained per jail. The file's time stamp is the last time a ban occured. fail2ban processes the bans starting from the oldest to the newest...

Without the PR, the bans are reported multiple times which causes the service (blocklist.de) to send out abuse emails once again. Edit: This solution prevents duplicate reports also in case...

Do you still maintain older releases? If yes, this might be a good fix to add to those.

TBH I think that code is already a little bit too compacted to be easily readable. I'll make a new PR with slightly easier to read code.

It's irrelevant because none of the transport protocols use TCP or are part of userspace.

Check test logs, there's some way to get them, I don't quite remember how. But there should be a way to do that. I had similar issues with other tests...

> Note it is not convenient to use {start,dpd,close}_action = trap to delay IKE SA renegotiation, as it is less reliable (in contrary with documentation!) than {start,dpd,close}_action That is wrong....

Okay, sure, we can implement that. But it costs much more money to implement that than to just eat up the carrier charges, or optimize that first.

Without the change, at least one test case crashes because there's garbage in the field, so comparisons with NULL fail and other code that dereferences the invalid pointer access unmapped...