00:37:18 Hello all. I have a firewall/gateway/wireguard gateway at home. Suffice it to say, things are pretty complex: my ISP uses DHCPv6-PD and the prefix changes every time I reboot, but I also like to be able to full wg tunnel from my laptop on the go, so I use a he.net v6 tunnel to have a static v6 prefix. I am using pf, using rtable to a second FIB that has the he.net default route so wireguard 00:37:20 clients can use it. It works fine except ICMPv6 Packet too big egresses out the WAN interface instead of through the he.net gif, complete with the gif's source address. I'm at my wit's end here, any suggestions would be appreciated! 00:42:04 I've also tried adding a reply-to to the he.net gw on the wg tunnel ingress rule (the one with the rtable clause). This prevents it leaking out the WAN interface, but I don't see it egress anywhere. 00:48:02 i aspire to one day understand anything you just said. but gl solving it 00:49:20 Lol yeah I know, it's mind numbing to think about even for me. The shit home users are forced to do to save money. 00:50:19 i mean i struggle to install openwrt on a tplink or cisco router. i can't imagine configuring pf or freebsd to do the big heavy stuff like that 00:52:28 pf honestly is a pleasure to work with. I've been using it in this role for 20 years. 00:52:51 But policy-based routing is very hairy in FreeBSD 01:03:02 it's hairy on linux too. i don't love it anywhere 01:12:40 rtprio: that is interesting. I don't think I could stomach moving to iptables for this, but I was under the impression that it would be one way to get it working :| 01:12:42 amigan, I'm re-reading your messages. is it *all* ICMPv6 packets which egress out of the WAN, or just ones that are too big? 01:16:08 edenist: it is all ICMP 01:16:22 Echo replies too 01:17:24 I even put a pass out on $wan route-to (gif0 $hegw) from to any, but that doesn't work either 01:22:35 I haven't had any experience with wireguard. This only occurs with ICMP to/from remote clients? 01:27:26 Yeah, but only by virtue of them being the only ones that use a different fib 01:28:21 I can ping wireguard clients from the firewall or other machines on the LAN and things work fine. It's just when the ICMP is a reply to a TCP or ICMP packet coming in from an external host that would route out the WAN default route normally 01:28:53 It's like ICMP responses have no idea about the pf state they are attached to and do not obey its PBR 01:32:56 But echo replies that are not generated by the FreeBSD box are routing incorrectly too. The wg ingress state creation rule (with the rtable clause) covers tcp,udp,icmp6 01:34:26 ok I gotcha. geez this is painful 😕 01:34:48 Yeah. I feel like I had this issue before, and then I got everything working perfectly, and then it broke again 01:35:09 I've been using this particular topology for a year with on and off issues. 01:39:20 so my ThinkPad T495 arrived today, planning on using it to replace this E14 that was having issues. I'd like to avoid having to go through all of the initial install/setup of FreeBSD on this T495 since I've got a mostly set-up install already. not sure the best way to clone the install 01:40:49 tm512: boot live usb, ssh sourcehost dump -C64 -b64 -f - / | (cd /mnt && restore -rf -) 01:41:08 If you're using zfs, zfs send | zfs recv 01:41:54 I'm thinking from a live image, after I get a ZFS pool set up, I could do zfs send/recv, though I dunno about how to make the system bootable 01:42:19 I have to set up an EFI FAT32 partition and install the FreeBSD bootloader there, right? 01:42:20 tm512, could always go low level and take a full dump of the drive 01:42:21 tm512: just install a vanilla zfs boot install and then overwrite the zfs pool with the send 01:42:27 or at least the part of the bootloader that's a direct EFI payload 01:42:33 Yeah, what edenist said too 01:42:54 Probably easiest to just do a block-level image and resize as necessary 01:43:21 it's fairly brute force, but it's my go-to when I just want it to work 01:43:27 edenist++ 01:43:57 assuming they both support the same boot type [legacy/efi etc] 01:46:19 amigan, so by the sounds of things you've got a better handle of pf than I do [my experience is largely hacked together until it works, heh.... but maybe thats for everyone? ;-) ] 01:46:36 presumably you have a separate rule for this wg ICMP traffic? 01:46:41 lmao, I think we're in the same league 01:46:51 not entirely sure how to resize partitions on FreeBSD. also home network is pretty slow (the old router I have only does 100Mbit) so sending a full disk image over the network would probably be quite painful. I suppose I could cut out the middle-man and directly wire the laptops together 01:46:56 Yes, the icmpv6 is allowed to egress 01:47:08 I mean doing proto {tcp,udp,icmp6} basically makes 3 rules 01:47:10 zfs send would send over less data though 01:47:15 So separate rule fo ringress too 01:47:47 tm512: not necessarily, if zfs has been trimming. dd conv=sparse 01:48:04 Unless this is HDD, then yeah maybe 01:49:19 tm512, if the drive is mostly empty, adding the '-C' option to the ssh/scp/zfs send will compress all those zeroes down pretty well [assuming they are zeroes and not old data] 01:49:54 amigan, it almost sounds like 'keep state' or something isn't being applied for those ICMP packets? 01:50:11 I'm not entirely convinced zfs send doesn't already do run-length encoding of zero blocks 01:50:38 edenist: precisely, yes 01:50:49 I never zeroed out the entire drive, though perhaps that was done by the seller 01:50:56 I haven't tested, but it sounds like something easy to do. Heck, any sort of compression on the dataset will take care of that too 01:51:02 I thought zfs send would just send only the data that's been allocated 01:51:27 tm512: unallocated and allocated but filled with zeroes are different things :) 01:51:56 Sparse blocks will definitely not be sent over as zeroes, but specifically talking about blocks that actually just have zeroes 01:53:17 But dd using conv=sparse will actually just seek if it deblocks an input block filled with zeroes. Just make sure you have bs=512 02:45:45 amigan: I might just end up going with the idea of installing FreeBSD normally on the T495 and then using zfs send on the E14, then doing the config changes I'll need to do since these are quite different machines 02:46:05 like I'll need to disable wifibox, and have it load amdgpu instead of i915 02:46:39 guess I also need to remove coretemp from the loaded modules 02:47:54 I'd kinda prefer a more direct way without having to perform a redundant install but it's probably not worth the trouble 02:52:37 tm512: honestly install it, install all packages, copy your etc/s, /boot/*conf, /home, done 02:53:00 If you really want to get dirty, just copy /var/db/pkg and all of /usr/local/ 02:59:59 well I also did stuff with this installation like configuring it for pkgbase 03:01:59 though this Ryzen 5 PRO 3500U is quite a bit more powerful than the i3-10110U in this other laptop, so perhaps it'll be less painful to just keep STABLE updated the oldschool way of just building from source 03:02:49 is there some sort of non bypassable domain blocking software out there 03:21:01 thegman: dns/blocky with pf rdr pass on blah proto {tcp, udp} from any to any port 53 -> myblockyserver port 53 03:36:03 Just wondering, is it theoretically possible to chroot into my Linux SSD from my FreeBSD SSD? Or is rebooting between them still the only way? 03:40:06 but when i say non bypassable i mean something like a password protected configuration that cant be changed 03:40:56 only way i could think of doing it was encrypting the ssd then making /etc/hosts immutable then enabling secure mode 03:52:31 remiliascarlet, you can do this now with linuxjails, which effectively uses the linixulator to emulate the linux kernel. 03:52:32 https://wiki.freebsd.org/LinuxJails 03:52:33 Title: LinuxJails - FreeBSD Wiki 03:53:23 thing is, I wouldn't trust a production drive to this, it has R/W access and it is likely to get into a state which the actual linux kernel might not be able to handle when you try to do a native boot 03:54:31 edenist: I mean more like, I already have a Linux installation. The reason for that is so I can compile packages for the distro without having to reboot for this. 03:55:44 Just for a 20 second or so time save. 03:56:05 thegman, there are modules like zenarmor which use netmap to do what you [I think] are wanting to do. They support things like pfsense/opnsense but I believe also support plain freebsd installs 03:56:15 they are proprietary though 03:58:11 remiliascarlet I'd probably just setup a bhyve VM and boot it. Linixulator is built and tested around a small rather fixed config, unless you're running a rather default install of ubuntu or redhat etc, you're going to run into issues with libs and such. Only certain glibc versions are supported, last I checked. 03:58:40 how well does bhyve work with steam 03:58:55 my end goal is to migrate over to freebsd completely 03:59:07 I'm running CRUX Linux, which is much more BSD-like than most other distro's anyway, and very simplistic. But can take weeks to set up. 03:59:30 i just need some sort of nfsw site blocking thing (i have one that only works on linux) and the ability to use steam 04:00:06 I've never done anything requiring GPU passthrough with bhyve, but in theory it can be done. Honestly it seems better just to use wine/proton and run the windows builds on freebsd. There are a few projects which help to streamline this 04:00:28 id rather use proton if that works 04:00:30 Steam is why I chose FreeBSD over OpenBSD for my desktop. Then it turned out that FreeBSD can't run Steam at all (neither the Linux version nor the Windows version), so I keep the Linux installation for games. 04:00:40 i just dont know how well proton works with jails 04:01:01 And I ended up running all 3 of them. 04:01:33 it really is a shame that companies dont lift a finger to support non windows operating systems these days 04:01:42 Linux for gaming, FreeBSD for day job, and OpenBSD for night job. 04:02:02 sounds complicated 04:02:34 Well, I run DWM on all 3 of them, so there is some degree of uniformity between them. 04:02:46 linux for gaming freebsd for everything else and windows for toilet paper 04:03:44 FreeBSD for day job, OpenBSD for night job, and Linux for hand job (sorry, couldn't resist this joke). 04:04:37 im gonna assume i dont want to know what that means 04:06:44 Men most likely have experience with this. I can't even do hand job. 04:07:42 if you have the standalone installs, games on freebsd work fine with wine/proton. But yeah steam itself is an absolute dog to get running, not worth the trouble imo 04:08:27 I already tried to running Proton, couldn't even get it to work. 04:13:32 what is it about jails that dont work 04:13:36 missing syscalls or something 04:13:47 Speaking of which, does FreeBSD have support for FFS2 (OpenBSD's default filesystem)? 04:15:30 Currently I keep important files in sync with NFS. 04:17:49 Not going to lie, I'm still on Linux 04:18:32 thegman: Not just companies, it's pretty apparent with programming languages as well. C and Go have pretty much everyone covered, but Zig for example is pretty much Linux/Windows/Mac-only, FreeBSD support is a maybe, and OpenBSD support only exists if ported by Theo and co. 04:19:11 And V is like Zig, but without FreeBSD support at all, but at least it's compilable. 04:20:16 I'm often left wondering why Rust is becoming so well adopted, and Go is normally not even considered 04:21:20 Right now there is Rust going into the Linux Kernel, the Cosmic Desktop by System76 is Rust. There was just a Government agency speaking about the merits of using Rust for their code 04:31:01 The one thing that keeps Go from becoming a serious consideration is its garbage collection, which works fine for websites, but you don't want to have it in games or kernels. 04:32:49 Rust sucks in many ways. Learning curve, syntax, compile times, woke foundation, toxic community (common side effect of pandering to the woke), its users being so religious about it. 04:34:00 Meanwhile, the only "bad" thing I see in Go is the focus on simplicity and just using the standard libraries as much as possible, which I think are actually good things that other languages really need to take note of. 04:35:36 SponiX: a lot of wayland utilities are rust 04:35:47 rust is like c++ but they removed the ability to do anything useful in the name of good programming practices 04:36:33 the benefit is you get to feel good about yourself after you jumped through a bunch of hoops to do something in 30 lines that could be done in 5 lines in c 04:37:27 you need to clone an entire struct just to assign a variable to an element in that struct 04:37:45 unless the element supports the "clone" implementation i guess 05:17:39 thegman sorry it's a late reply, but with the jails it's not just syscalls [this might actually be complete now], but it's the compatibility with things like glibc. The old way was that a base linux system was installed based on something like centos, and that limited things because the libc was so old. 05:17:58 But now you can in theory install any base system. It's the compatibility with the core linux libraries and the emulated syscalls which I believe can cause issues. It's only really been tested with ubuntu as far as I can tell. 05:18:50 rats 05:19:16 also as far as the domain blocking system goes i cant use anything router based 05:19:29 since im stuck with this crappy spectrum router 05:20:08 even if it worked, I don't think emulated linux software has access to the native network stack on freebsd 05:20:22 you've been able to install any base even since the c6-in-ports days, but nobody really took advantage of the diversity (and it probably would die quickly with a relatively modern glibc/musl and some applications to go with it back then, but nothing would have prevented others) 05:20:22 can you throw it in a VM and route your traffic through it? 05:21:46 nothing's fundamentally changed about linux stuff here in a long time, but more recently newer syscalls have been implemented and some folks have actually taken to running a linux test suite to validate the implementation 05:22:23 maybe i could route it through a vm if i could make unable to connect to the internet without that vm 05:22:54 kevans, that was my assumption. The c6 implementation was comprised of components which were at least "supported" in the linuxulator, and that the implementation is now complete enough that that constraint isn't necessarily needed 05:24:02 maybe if i set up that vm as the router and disconnected the old router 05:24:04 il try that 05:24:18 yeah, pretty much 05:27:47 im gonna try to install freebsd again 05:41:08 why are you installing again? 05:41:14 and why can't you use your own router? 05:55:24 rtprio: "a lot of wayland utilities are rust" Yet another reason to refrain from using Wayland. 05:55:39 Or "Gayland" like a friend of mine calls it. 06:01:33 >gayland 06:02:26 sounds like you want people to use wayland ;3 06:04:35 well 06:04:38 didnt work 06:04:43 it still cant connect to the internet 06:34:05 miko: I let people make up their own minds, but being called "gay" is definitely nothing positive. 06:34:21 Unless you are gay, then that's fine. 06:37:48 are your friends 10 years olds from the 90s? 06:55:39 feels like I'm being transported back to the mid to late 00s when I was in middle and high school 06:58:58 yeah, the fucks that bullshit? 06:59:23 you stuck in /g/ or something? 07:08:21 yah, that's not a very good attitude 07:11:44 What's the problem? 07:13:34 Feels like I'm back in the #gnu channel all over again. 07:16:34 i wonder what could possibly be wrong with trying to make the word gay have negative connotations 07:16:38 god i really can't figure that one out 07:16:40 i really can't 07:17:33 "Oh no, she has a different opinion, the sky is falling!" 07:18:21 feel free to have your own opinion, but it's fucking stupid 07:18:35 I think you're overestimating how important your opinion is to any of us, tbh 07:19:21 it's still something to call out though 07:19:26 I'm not the one forcing you to have a certain opinion here. 07:19:51 calling you out for having a stupid opinion isn't forcing you to have another opinion 07:20:12 There is no such a thing as a "stupid opinion". 07:20:20 there absolutely is 07:20:57 Only if you want others to think the way you do, you can call it "stupid opinion". But then again, all you're doing is proving my point. 07:23:02 *wanting* people to not be homophobic or permissive of homophobia is not the same thing as forcing that 07:23:08 what, so one shouldn't call opinions based on stupid superstitions stupid? 07:23:14 ^ #freebsd-social 07:24:50 tm512: Just like you're free to find me a "stupid bitch" or a "good person", you should be free to find homosexuality good or bad. 07:25:29 So I can't see the problem here. 07:25:36 and we're free to call your shit out 07:25:47 Yes. 07:26:36 Although I don't see any shit being called out, just a bunch of people being mad over an opinion I hold. 07:27:22 im literally calling you out for having a stupid, shit opinion 07:27:39 That sounds very authoritarian. 07:28:12 that's authoritarian, huh? 07:28:24 comparing "Gayland" to shit I heard all the time from literal children back when I was in school, in a less progressive time, isn't "being mad", it's mockery 07:28:26 Sure, you're free to find my opinion stupid or shit, doesn't mean it's a fact. 07:28:32 do we need this offtopic discussion here? 07:29:44 remiliascarlet: and who exactly said that it's a fact? do i really need to prepend everything with "i think {}"? 07:29:45 well, I don't care about this enough to join a different channel so if it's too off-topic then I'll just dip 07:29:46 seriously? 07:30:40 Whatever, I said what had to be said. 07:31:09 So I'm more or less finished with this topic. 10:26:36 unfortunate, first experience with FreeBSD on the ThinkPad T495 I just got was running into this bug: https://forums.freebsd.org/threads/t495-thinkpad-hangs-on-boot-acpi0.85650/ 10:26:37 Title: Solved - T495 Thinkpad hangs on boot acpi0 | The FreeBSD Forums 10:27:50 suppose I will try updating the UEFI firmware. while that seems less risky than flashing new firmware on BIOS machines it still makes me a little bit uncomfortable 11:21:35 during the second freebsd-update install to upgrade to 13.3: rm: ///usr/include/c++/v1/__tuple: is a directory <- should I be worried?.. 11:22:53 https://x0.at/fMiZ.txt 11:23:41 (date now is 2024-03-05 12:23) 11:24:50 not really it's a special case in the upgrade where what was a file in earlier versions became a directory after upgrading 11:25:05 it was a bit of a blocking issue for the 14.0 release 11:25:26 if the machien doesn't reboot, I'm blaming you ;P 11:31:55 ok, reboot seems to have gone smoothly! 14:28:59 > ZFS: i/o Error all block copies unavailable 14:29:04 what a glorious way to start the week! 14:29:15 very nice 14:29:59 I am slightly terrified I will admit 14:30:33 I will try to reinstall and re-import the pool 14:34:14 wtf error is that 14:35:13 but i admit i've done same 14:35:42 did recover the pool, but not all the files 14:36:25 ketas: might be root pool corruption, I just upgraded to 13.3-RELEASE 14:36:36 in the bright side, zfs told what was corrupted 14:36:38 oh that 14:36:44 no storage fail? 14:37:12 god that's bad 14:37:36 it's not telling much alas 14:37:58 ketas: what do you make of this? https://imgur.com/a/rVXVxE8 14:38:00 :D 14:38:49 are the devices with the pool actually present? did you update the pool without updating the loader? etc. just boot off a thumb drive and try to import see if it's even recognized. 14:39:56 isley: I only followed the steps there: https://www.freebsd.org/releases/13.3R/installation/#upgrade-binary 14:39:58 Title: FreeBSD 13.3-RELEASE Installation Instructions | The FreeBSD Project 14:41:29 so yeah, I'm probably going to perform a reinstall with the help of https://gist.github.com/ctsrc/9a72bc9a0229496aab5e4d3745af0bb9 14:41:30 Title: Install FreeBSD 13.2 on Hetzner · GitHub 14:41:43 (btw if anyone knows a nice hosting company that supports FreeBSD I'm all ears) 14:44:20 Hi. How to debug the output of "service myservice start"? My service doesn't start with that way, however when I call it directly, it works. I'd like to see what's the matter.. 14:44:51 $PATH ? 14:49:42 ketas: -bash: /sbin:/bin:/usr/sbin:/usr/bin:/usr/local/sbin:/usr/local/bin:/root/bin: No such file or directory 14:50:30 within service init? 14:50:38 wild guess is path still 14:50:54 Yes, the binary is in /usr/local/sbin/ 14:51:02 which service? 14:51:12 any common thing? 14:51:56 owncast, but from: https://pkg.skunkwerks.at/FreeBSD%3A14%3Aamd64/All/owncast-0.1.2.pkg 14:52:24 Installed it with: pkg install owncast-0.1.2.pkg (the current one on FreeBSD 14/amd64 is almost 3 years old, extremely buggy and never gets updated) 14:54:31 eh 14:54:59 i would still bet it's env / conf :p 14:59:10 aah is this mine tercaL ? 14:59:26 so when it didn't start for me, it was because I needed to: 14:59:37 - create user/group owncast:owncast (done by installing port) 14:59:44 - add /var/db/owncast (manual) 14:59:55 - chown owncast:owncast /var/db/owncast (manual) 15:00:40 add this to rc.conf 15:00:40 owncast_enable=YES 15:00:40 owncast_path=/var/db/owncast 15:00:48 then it ran fine 15:00:59 dch: Oh, was it you? :) Thanks a lot for the binary! I will try these now. 15:01:01 the current port does what I consider a POLA violation 15:01:15 - it has /usr/local/share/www/owncast/ dirs as writable by the user 15:01:29 there may be a good reason for this but as I dont use the port how would i know 15:01:41 but I don't expect user data to end up in /usr/local/share/ 15:01:52 why is the port old 15:01:56 that should be root:wheel and readonly for users, updated by/with the port 15:02:15 because maintainers have lives, jobs, sickness, holidays, kids ... 15:02:25 sometimes we are on top of things, and sometimes we are ... not 15:02:41 tercaL: anyway if you have the permissions right, this should just work (tm) 15:03:10 tercaL: IIRC the first run i did as follows, without the daemon 15:03:21 dch: I agree, but the port's maintainer never had time in 2 years and half, just strange.. anyway.. It's extremely buggy.. By the way, the steps above solved it! Thank you very, very much! 15:03:32 su -m owncast -c 'owncast -enableVerboseLogging -adminpassword l33t' 15:03:47 for a first-time user I think this is necessary to set a default password? not sure 15:03:51 bbl, family times 15:04:10 https://www.freshports.org/www/owncast/ 15:04:11 Title: FreshPorts -- www/owncast: Self-hosted streaming service written in Go 15:04:12 please update https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=275190 with any findings, so we can commit it *if* it works 15:04:14 Title: 275190 – www/owncast: Update to 0.1.2 15:04:16 2 years? 15:06:33 I there, is there a way to install php5 on bsd ? 15:06:41 need to resurect an old project ... 15:07:40 manually i'd guess 15:12:38 nerozero: it will be hard. Find when that port disappeared, spin up whatever FreeBSD jail was "latest release" at that time, and fork the ports tree and build in poudriere off that 15:12:49 MOVED file says `lang/php56|lang/php72|2019-01-01|Has expired: Security Support ends on 31 Dec 2018` 15:13:10 so whatever FreeBSD was around on January 2019 is what you need to build against 15:13:28 you can also expect many of the tarballs needed from that time may no longer be available on the internet 15:13:31 this could be painful 15:13:41 .................... 15:14:32 or maybe you can try building it manually from source, old school way 15:15:51 Can bsdinstall's auto ZFS option detect my pre-existing ZFS pool? 15:16:32 dont even know if that old php sources exists or not ... 15:17:22 i bet php repo has it 15:18:20 and my distfiles archive :p 15:18:27 wooow 15:18:46 wait looking for obsolete drives with bsd installed there ... 15:19:46 https://museum.php.net/php5/ 15:19:48 Title: PHP: Release Archives (museum) 15:19:59 WOOOW ! 15:20:03 thanks a lot ! 15:20:05 love how they call it museum 15:21:24 i recently merged my distfiles 15:21:58 found it being 30g 15:22:19 regular server stuff, little x 15:23:45 Hecate: did you check it first? 15:23:54 before reinstall 15:26:25 i wish there's better method here 15:27:08 apart from manual repair 15:27:23 did it import? 15:31:08 found hard drive with php7.2 and bsd 10 15:31:50 wish me luck, hope I will manage to make make things work on it 15:32:04 btw, is there a chance clone active system into vm ? 15:32:15 bsd into vm ? 15:33:48 rephrase, how can I transfer entire system into VM 15:33:58 tar | tar? 15:34:15 I had tried many years ago, doesn't seems to work 15:34:26 with some cfg changes 15:34:31 symlinks, weird files - no boot, ALOT of issues 15:34:39 not worth it 15:36:08 chflags -vv noschg /mnt/var/empty 15:36:08 tar -cf- -C /root/files/fbsd/current/dist/univ.working/ . \ 15:36:08 | tar -vxpf- --clear-nochange-fflags -C /mnt/ 15:36:18 actually from my vm script 15:37:22 woow 15:40:16 :) 15:42:31 from php museum you can also get php1, which is something i never used 15:42:53 which I never even heard of .. 15:43:34 well there was php before 3! 15:44:33 Have no vim on old machine, is it possible to do something without vim > 15:44:34 ? 15:44:51 vi ... 15:44:53 vi, ee 15:45:00 no EE -hate it 15:45:07 and nano 15:45:15 well it gets files edited 15:45:17 hate nano too .. 15:45:24 ed then 15:45:35 ee has that esc delay 15:45:36 or sed -i 15:46:04 i've see people battle with it irl 15:46:14 I know 15:46:24 i almost always use mcedit somehow 15:46:32 10 yars ago migrated to vim, and world has changed .... 15:46:33 i could do vi or ee if needed 15:47:00 20 years of mcedit eh 15:47:14 AH I have MC there 15:47:24 havent used it for 15 years ... 15:47:28 WOOW 15:47:36 i don't know vim has problems in my brain 15:47:41 how to exit mcedit :D 15:47:49 not :qw! 15:47:52 :) 15:47:53 :D 15:48:17 :qw *SHOULD* work. 15:48:29 in mce? 15:48:45 I cont even type a thing in it ! 15:48:57 if f keys don't work 15:49:01 esc 0 15:49:16 no extra i, hjkl /.... 15:49:31 ARROW nav.. 15:49:51 horrific habits 15:50:16 no this is really a thing 15:50:23 ketas: it wouldn't even start the login stuff so eh 15:50:27 well *i* had arrows on my kbd 15:50:33 in 1999 15:50:34 so 15:51:05 Hecate: eh, booted off rescue media? 15:51:07 I wish I can go back to that year ... 15:51:29 what happened then 15:52:10 there was a life, no ai, friends, not virtual one, a lot of alive back then ... 15:52:19 a lot of things ... 15:52:29 there were also arrow keys on kbds in 1995, when i was 12yo 15:52:35 so yeah 15:53:39 I don't use ee myself, but I do push it on people who need to work in terminal, but can't be arsed to learn vi(m). 15:53:46 i had less life the rh 15:54:07 theb 15:54:07 The mere fact that it always displays all the hot-keys at the top lightens the support burden a lot. 15:54:08 you dont have to learn vim, 15:54:08 n 15:54:31 vim speaks your language, and if not, you can tweak it 15:55:15 I was a vi hater for a number of years. . . but I eventually forced myself to learn and use it, and ended-up loving it. 15:55:29 https://image.slidesharecdn.com/iusevim-111217062725-phpapp01/85/i-use-vim-3-320.jpg?cb=1672898711 15:55:30 Haha !!!! 15:55:32 Love how it's designed to keep my hands at the home position. 15:55:37 you dont have to learn vim, you just have to learn how to exit it :) 15:55:52 Is that an emacs joke? 15:55:54 :) 15:56:19 CrtxReavr: yes :) - btw I am a vim person ~ 15:56:37 Everyone's first VIM session: ^C^C^X^X^X^XquitqQ!qdammit[esc]qwertyuiopasdfghjkl;:xwhat 15:56:43 no images needed 15:56:51 not mine 15:57:09 I had read a book before even start unix ... 15:57:26 so my first vi was vi, :wq 15:57:30 :D 15:57:40 bsd2.1 15:57:43 i love the what part 15:57:44 coughliescough 15:58:02 but yeah who could guess the : 15:58:36 home position is fun too 15:58:44 I took a CS-105 class Intro to Pascal. .. 15:58:51 it assumes you learned touch typing 15:58:52 that was not a guess, it was written in the text book, I had no pc at that time, so books was a thing 15:59:04 if you learned hybrid, then 15:59:10 The platform was some HP SysV variant. . . and vi was offered as the only editor choice. 15:59:22 Hated it - made no sense to me. 15:59:23 * jbo is about to flip over some tables 15:59:51 relief from destruction and screaming 16:00:17 I basically bot by with editing my code in edit.exe and doing text uploads into vi. 16:00:29 s/bot/got 16:00:33 jbo, backup the entire database before :D 16:00:52 made my day 16:01:15 :D 16:01:23 my bhyve windows vm keeps getting stuck at boot 16:01:26 they got arcade game on that btw 16:01:27 it sometimes works 16:01:30 table flippet 16:01:30 like 5% of the time 16:01:32 r 16:01:40 and once it was able to boot it stays usable for 10 hours no problems 16:02:09 afk, need some calories 16:12:20 no usb-c ps in humans yey 16:12:22 t 16:12:27 pd 16:28:47 Any chane anyone has ran freebsd-update going to 13.3-RELEASE on a t2.micro aws isntance? Just had freebsd-upgrade killed.. Wonder if it's a RAM thing 16:37:23 * skered shrugs... second time around worked. 16:37:44 I did at least stop apache this time though. 16:41:26 Has anyone had success/issues with SATA PCI card passthru with bhyve? I'm hoping not to waste money buying several cards (like I did to find a compatible USB chipset once). TIA 16:41:45 davisr, PCIe passthru works well for me 16:42:25 regarding your USB comment: shouldn't matter what controller it is. Just attach the PPT driver and then let the guest worry about it. 16:47:05 eeek... gunzip: (stdin): unexpected end of file During `freebsd-update install` post reboot... 16:47:13 jbo, thanks, but it's unfortunately an un-ideal world and some peripherals, like USB controllers and GPUs, have passthru issues. I haven't any experience with SATA card passthru, but I'm hoping to hear about a card that is known to be working. 16:47:43