hum interesting I have come up with another approach in my end which is probably less intrusive. what I did is too hackish, but basically I am allowing to find the ramdisk in memory via loader.efi memdisk=ipxe:memdisk where memdisk is the name of something already in memory: imgfetch --name "memedisk" http://pathtoimg/memdisk.img, with ~250 Lignes of code added on top of what you have already done. the first step needs a bit of refactoring, like https://reviews.freebsd.org/D59979

right now no, but I can do that once I have converted it do tree(3), but it will take some time ;)

about netstat -u I haven't thought about it, for sockstat I have a patch which I will put in review later where I show the abstract unix socket as '@path' @ instead of \0 we can probably do the same for netstat -u (this is what linux does in the ss command

don't worry, as I explained, there is a github with an alternative which I found complicated, and I wondered if I was missing something.

this is what we discussed about at eurobsdcon, my find this version very simple, given other implementation I found are more coplicated, probably I am missing some point, but it seems anyway pretty complete from my testing. (if we have an issue with performance the linked list may be converted to hash table.

it is not useless at all, it is in particular very useful for anyone who wants to setup the web frontend and more.

Fix the comment

Address @dteske comments

missed git add

yesh I pondered the tempfile, but this was longer, actually, but yes I think you are right, as printf will complain also about the ARG_MAX.

Tested with success!

https://reviews.freebsd.org/D59348

https://reviews.freebsd.org/D59351

Note There is another pdfork() which we don't have anycall on waitpid either in zygote.c, but those are long living processes which dies along with the main process and so get reaped normally. so I haven't added the flag on it.

I am really not confident about this one, I tried to put all my understanding of the situation in the commit log, so reviewer can see what I understood

I think I have now a clear reproducer:

even with this patch I continue to get zombies after few hours. so clearly this was the wrong direction, sorry for the noise.

cat /home/bapt/crap.sh

#!/bin/sh
id
echo plop
sleep 1