00:07:44 kind of the last place I would expect a good desktop experience 00:08:08 on the other hand I would say after debian and prolly ubuntu. 00:15:58 feurig: what 00:18:17 fbsd+xfce4 = the best desktop experience ever 00:36:09 I think there's a newer maybe the exact same thing as lsop in the ports tree... does this port ring a bell? lsop will scan for outdated binaries and report in hopes you restart the pid/process/ 00:36:20 For some reason I'm thinking it starts with a c? 00:49:56 Think I found it... checkrestart. 00:54:17 Seems like it's either doing it wrong or doesn't have the false positivies that lsop would trigger. See how it works out. 01:01:20 once upon a few weeks ago, i tried to setup freebsd on my desktop, so i could just use xen and have freebsd and linux working together, and gradually move to freebsd maybe, but freebsd seemed to choke on xen 01:01:37 dom0 or domu? 01:01:42 dom0 01:01:51 immediate panic 01:12:59 welcome to freebsd the true free and open source OS 01:21:05 skered: would be nice if rc's status said that shit needs restart 01:51:07 is it safe to install runit on fbsd? 01:53:14 why 01:54:15 i guess it is if it is available in the ports 02:50:24 is there a simple list of commands to boot from grub or things to try? i have a machine that can only boot syslinux or drop into grub cli, uefi doesnt work so ive tried insmod bsd, kfreebsd the kernel and boot it but it freezes the screen and nothing, im guessing im missing some video firmare or modules or what not but hard to see what precisly (openbsd boots for install and I can run the install 02:50:26 but rebooting into the installation brigs vertical stripes garbled screen) 02:50:31 woups sorry 03:06:57 cedb: https://bpa.st/7CRRC 03:06:58 Title: View paste 7CRRC 03:40:19 mason: thanks will try 08:47:04 https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=270632 we could just remove ext4fs from https://man.freebsd.org/ext2fs(5) 08:47:07 Title: 270632 – [ext2fs] files <4096 bytes are corrupted on ext4 filesystems 08:52:09 guid morvening 08:54:27 _xor: https://github.com/valpackett/capsicumizer is pretty awesome, but I don't recall any pledge/unveil specific work 08:54:28 Title: GitHub - valpackett/capsicumizer: Run anything (like full blown GTK apps) under Capsicum 09:41:47 Anyone has experience with ravynos in bhyve? 09:46:31 meena: why? ext4 is supported 12:48:35 I'm more inclined to wonder if something broke during a patch, because fsu@ has been working on ext2fs pretty consistently. 14:03:06 is there some trick to get poudriere-devel to actually fetch a pkg? I just want it to pull rust, and no matter what I do, it refuses to just download it. all default config, etc. should it be removed from all my lists, configs, etc? 14:07:22 i specified "-b latest" with my poudriere command, when i did the options I selected all of them with the "+" (assuming those are default), its not in my port list, but it still won't just pull it down 14:14:47 if you want the defaults except for some small changes, it is easier to create a -make.conf and use _{UN,}SET in there (but iirc if you created those options files with poudriere options (or make config for plain ports) those override everything, making it annoying to work with, defaults change, ports gain new options, ...) 14:18:34 afaik, that + means it is a new option not covered by the current options file and to know if it is default you would need to read the port (or look at freshports) 14:19:42 i just did a -v and it told me my options differ, so fixing and attempting to redo and see if it pulls it this time 14:21:59 if you want the defaults, why have you created the options file? 14:23:09 when you run poudriere options it brings it up since its a dep of a port i'm building 14:26:13 iirc you can press alt+c there to not create it, but as said already, easier to manage with -make.conf files instead of poudriere options 14:29:43 changing the options almost fixed it, I need to pull in openssl and curl.. i have curl setup but not openssl, so assuming once I add that, it should pull it. I'll look into editing my make.conf as well for the future 14:43:30 well i can't pull in openssl since it appears to be part of fbsd and there don't seem to be a port. how it's listed as a dep then baffles me. 14:46:13 hey, i managed to compile a native bps patcher called 'beat' for freebsd. its used to apply patches to various romhacks for older games 14:46:39 should i work on an official port or nah? would this be interesting to anybody? 14:52:50 there is an openssl port, but normaly a port depends on the base one, you can control that via DEFAULT_VERSIONS I think 14:53:15 yeah i found it, whereis failed me 14:53:39 gzar: seems like a good idea for the ports tree 14:54:34 alright, i've never done porting before so i will read up on it 14:55:38 r0ni you need some stuff set in poudriere.conf 14:55:42 what have you got atm? 14:56:20 PACKAGE_FETCH_BRANCH="latest" 14:56:20 PACKAGE_FETCH_BLACKLIST="collectd*" 14:56:34 thats all I have, and it looks like `-b ...` can accommodate the first one 14:57:31 and if you do `poudriere options -cn ...` you can set options only for the top-level port you want to modify 15:00:32 ya ive set all that, and the default WHITELIST with gcc rust llvm all in it 15:00:53 r0ni: the only other reasons are might not fetch are: 15:01:15 - your /usr/ports git commit is no longer lined up with whatever the pkg builders are at 15:01:21 - you have custom options set 15:02:16 ya my port-list wasn't pulling in openssl tho its a dep of a dep, I have to explicably list it in order for poudriere to see i have it 15:03:25 i'm letting it check options atm, then i'll try a new bulk to see 15:04:01 poudriere shouldn't need that, it is very very good at working out the dependencies. its kinda the main thing it does 15:12:29 01:17] Package fetch: Skipping rust-1.68.2: deps wanted: curl-7.88.1 openssl-1.1.1t_2,1 15:12:32 [00:01:17] Package fetch: Skipping rust-1.68.2: deps remote: curl-7.88.1 15:12:32 [00:01:17] Package fetch: No packages eligible to fetch 15:13:00 this is what i get even adding openssl to my ports list... i'm at a loss for what it really wants from me 15:14:57 it has built the openssl pkg (its with all the rest of them) so i dunno what more it's looking for, it's all there 15:18:54 is it posible to boot freebsd using ipxe ? 15:21:29 it's aliiiiiive! 15:23:23 la_mettrie: finnish? 15:23:26 me too 15:23:29 moi 15:24:35 neirac: yes 15:25:43 i have some questions about firewall 15:25:50 i know I am a newbie 15:25:52 but 15:26:47 if I have NAT router and in the settings it says it has some kind of a firewall on 15:27:19 do i still need to switch the ipfw on if I have no services listening to to net 15:28:05 and are the "workstation" defaults fine in ifpw 15:29:11 am I safe? 15:35:26 dubiousness I don't seem to find a doc that works, on the diskless installation document is not clear where I download pxeboot , I was trying ipxe but is not working 15:38:36 neirac: preliminary search says, yes to ipxe boot 15:46:26 meena Could not boot: Exec format error (http://ipxe.org/2e008081) trying to boot using ipxe 15:46:27 Title: iPXE - open source boot firmware [err:2e0080] 16:12:18 LapsangS: why? well, persistent user process, systemd has ~ systemd/ which allows normal users to have services started by the supervision system itself 16:25:21 neirac: Your best best is to try it in a virtual environment first if you have the capabilities to 16:26:12 dubiousness I was booting ipxe smartos successfully now I'm trying FreeBSD, seems there is no clear docs on ipxe for this case. 16:28:21 neirac: Are you just trying to PXE boot FreeBSD? Are you using mfsBSD? 16:32:59 mason: Regarding the epair bug, epair is a cloned interface (ifconfig -C) capable of processing VLANs. It is documented in epair(4). They are really useful for testing, but also a nasty edge case. Soon-ish it'll gain a few more tests. 16:34:01 jlduran: Ah, thanks. I'd looked for the concept of "sub-interface" and missed it, but I didn't dig into how it works with vlans as yet, so that's probably going to be enlightening. 16:34:13 ek, yes I'm fetching a mfBSD iso , I'm using imgfetch to get it , could be my ipxe script that's wrong 16:37:47 neirac: It could be. I know it isn't iPXE-specific, but these instructions are the same as I use for mfsBSD booting with standard PXELINUX: https://eerielinux.wordpress.com/2021/02/05/multi-os-pxe-booting-from-freebsd-12-pxe-menu-and-bsd-pt-3/ 16:37:49 Title: Multi-OS PXE-booting from FreeBSD 12: PXE menu and *BSD (pt. 3) – eerielinux 16:41:40 mason: I'm sorry, I might have pulled that term from my Cisco days. Sub-interface roughly translates to VLAN interface in this case. 16:42:05 ek those are using pxeboot not ipxe, I'll try that, where I download pxe to chain it? 16:42:43 jlduran: Aha! I never got much further than a surface familiarity with IOS. Thank you again. 16:46:18 neirac: The download and configuration of the PXELINUX (Syslinux) setup is also included in that same part of the article. It uses the FBSD pkg to extract the needed items. 17:03:35 what can one use with getopt instead of "--", which does not seem to be handled? 17:03:38 getopt -n vtsgram -o 't:p:b:s:?' -- -p100 -b300 -s '-z60 -Z-30' /mnt/nas/vlf44/events/1676926423.vt 17:03:58 it does not seem to accept -- 17:13:28 CCFL_Man: what do you mean instead of "--"? "--" is the end of options option 17:16:17 TommyC: that's the convention, yes 17:16:27 and that's not how you use getopt, at least the one in the base system 17:16:45 (-n -o suggest you are looking at linux one?) 17:16:57 rm -- -f # delete file named -f 17:19:44 wouldn't it be cool if basic shit like getopt were the same between Linux and other systems 17:20:07 otoh, maybe we should just stop writing shell 17:20:25 yuripv: so that was pulled fron a script designed to run on linux and -- specifies the end of options. is there a way to specify it differently with bsd's getopt 17:20:48 meeena: it should be 17:21:18 TommyC: yeah, -- to specify the end of options doesn't work on bsd's getopt 17:23:20 looks like devel/util-linux has the getopt you want (didn't check) 17:25:19 wait, "--" doesn't work in fbsd? https://man.freebsd.org/cgi/man.cgi?getopt seems to imply it works 17:25:20 Title: getopt 17:25:22 yuripv: will installing it break anything? 17:25:40 TommyC: it does, but the placement of it seems critical 17:25:56 -- usually means stdin 17:26:55 CCFL_Man: why should it? :) it's just a port, you'll just need to specify full path to that getopt binary 17:27:18 CrtxReavr: right 17:27:39 TommyC: i might have to call it by a different name 17:27:42 '-' means stdin, '--' is the end-of-options 17:28:23 CCFL_Man: you need to use "--" to denote the end of options at the end of all of your options 17:28:42 like the rm example meeena gave, rm -- -rf, vs rm -rf -- 17:30:18 dd if=/dev/urandom of=/lib/\* 17:30:37 like getopt -- -n vtsgram -o 't:p:b:s:?' -p100 -b300 -s '-z60 -Z-30' /mnt/nas/vlf44/events/1676926423.vt 17:31:23 wait wait, is "vtsgram" the script and you're calling it via: getopt -- -n vtsgram ? 17:31:46 yes 17:32:07 the script itself uses getopt 17:32:17 right, but you don't use getopt on vtsgram 17:32:31 or is that actually how you are calling it? 17:34:09 it is calling getopt. here is vtsgram: https://bpa.st/LMHR4 17:34:10 Title: View paste LMHR4 17:34:27 it stops at line45 due to getopt 17:35:28 that's not how you use getopt 17:36:09 not on BSD, anyway 17:36:13 just install the linux one. 17:36:39 there's also a lot of evals...I don't like that 17:37:15 yuripv: that might be the best option here 17:37:36 TommyC: that's done to keep the script as portable as possible 17:38:37 why not skips the OPTS= line and just use a while loop? 17:39:56 yuripv: i see you wwere a port maintainer for it 17:40:06 TommyC: i'm not sure 17:40:18 how is the linuc-utils one called? 17:41:57 CCFL_Man: no idea 17:42:46 it looks like it's just named getopt but i don't know how it would be called instead of the BSD version 17:42:54 honestly I don't even use getopt, I use getopt*s* 17:43:27 CCFL_Man: nope, that's not me 17:44:33 oh 19:52:16 Is there a easy way to get nfsv4 with sec=sys working on a freebsd server and linux client? I can get the exports mounted on the client but get input/output errors when trying to read it 19:54:22 its a shame freebsd v3 doesn't have crossmnt 20:19:21 ugh annoying that linux to linux v4 works without issues 20:25:22 have you read nfsv4(4)? 20:26:57 sec=sys is using uid/gid, so you would need uid/gid values to be the same across the systems. 20:28:16 yup uid's match 20:30:43 also make sure the nfs domains match (or your users will be translated to nobody;) 20:30:44 it also mentions the sysctl needed to be turned on if you don't run nfsuserd 20:31:50 anyhow, IO error would hint about something weird going on 20:32:49 yup the sysctl's are enabled but from the page its more for letting you get the "real" ids instead of nobody nobody