05:34:44 ox1eef_: got your message, thanks for the info, I'm going to install it and see :) 07:14:16 If I were to start a BSD user group in Seattle, would anyone be interested? (Or... are there any pre-existing groups I didn't find?) 07:35:55 xtile, the website looks dead: seabug.org/ 07:36:34 Yep, I noticed that. The web archive has a snapshot of it in 2017 but it doesn't seem very up to date at that time either. 08:42:48 Little bit further up "Hamilton BSD Group" ( https://www.hambug.ca/ ) meets up rather regularly 08:42:49 Title: Hamilton BSD User Group 08:43:03 s/BSD/& User/ 08:47:20 Pretty far away, though their meets are online, and Jitsi is pretty fine. 11:48:57 is there any event like hardcore DIY driver-writing workshop? since my hardwares are supported i lost the motivation 11:49:27 write-your-own-driver workshop 12:22:23 micttyl: what Hardware is it that you have there? 12:25:15 just graphics and wireless 12:50:59 it's more like admins here i think. such questions seem to be answered well. 13:14:38 anbody up for some ipv6 help? 13:15:21 I swapped back to my freebsd router again, and the (AFAICT) working config from May last year doesn't do the ipv6 things again 13:16:00 internet <> ISP-inflicted pass-through router <> FreeBSD gadget <> pppoe 13:17:22 pppoe being md5 13:17:26 *mpd5 13:19:03 mpd5? pppoe? hell, i used that for "VPN" 13:22:21 yeah its all I can use here 13:31:18 dch: that's a lot of hops, and lots of opportunities for things to go wrong 13:31:36 meena: its no different to anybody else 13:31:56 pppoe is the way the freebsd box does a pass-through connect to the internet 13:32:13 some people have a nicer connection option, sadly not here 13:32:50 micttyl: #freebsd-dev is where (some) devs hang out. #bsddev on EFnet is (also) where (some (other?)) devs hang out 13:33:37 micttyl: that said, those are two incredibly complex devices, and writing drivers for them will not be easy 13:34:56 micttyl: wifi is on #freebsd-wifi and graphics #freebsd-xorg both on EFnet 13:35:05 active but not across all timezones 13:37:02 micttyl: I guess you already know about https://docs.freebsd.org/en/books/arch-handbook/driverbasics/ & the book `FreeBSD Device Drivers: A Guide for the Intrepid` too 13:37:04 Title: Chapter 9. Writing FreeBSD Device Drivers | FreeBSD Documentation Portal 13:44:35 hey all if you had the following disks: https://dpaste.org/xX8uF and you was setting up a development work station which is a mixture of one bhyve container (running linux+docker) and some local development on freebsd (over ssh); how would you configure such a selection of disks 13:45:34 my present implementation is: zfs/RAID1(3 magnetic disks) with log and cache on the nvme, swap is also on the nvme, the SSD is where freebsd is installed 13:45:48 the VM sits on a zpool that is on that raid1 13:46:28 starting to think I should have done for freebsd to the RAID1 of magnetics and used the ssd for the VM, the nvme for cache,logs,swap etc 13:52:02 daemon: thats kinda what I ends up with 13:52:29 2x2 hdd in striped mirror 13:53:39 with a SLOG on the tail end of the NVMe 13:54:10 also, for a large part of my stuff, I turned sync=disabled on because its not very valuable data 13:54:17 at least, on the hdd 13:54:31 anyway its all about the backups guarantees 13:55:00 what I have on the nvme is a small set of stuff thats tarsnapped, that I really rely on, and a large chunk of stuff that is sent to rsync.net with restic 13:55:24 ah I use the magnetics as its 3 disks in raid1 for important stuff 13:55:32 I use the nvme for low latency fast read/writes 13:55:54 with the SSD kind of as the OS disk 13:55:58 yeah so I can afford the risk of a few hours of lost work, most stuff is git pushed elsewhere 13:56:05 same here 13:56:10 if you have an nvme do you really need an os disk at all 13:56:45 yeah the bios does not support booting from it, its not motherboard native I shoved a nvme card into an unused pciex16 13:56:57 thats a good rason 13:57:00 *reason 13:57:42 its such a weird box, its actually my work box (given to me by work a few years ago) 13:57:53 but you get 150 GBP 'desktop budget' per year to do what you want 13:57:56 so I got it the nvme 13:58:15 the hard drives come about because I was dealing with a huge database and needed the space 13:58:23 but its almost entirely headless 13:58:41 do not think its ever even had X installed (except the first 2 weeks I was there, where linux mint was pre-installed) 13:58:49 so its a kind of odd amalgamation 13:59:56 if I owned it I would upgrade its ram, when I am in the office on wednesday I will pop the panel have a look see what it supports 14:00:22 according to dmidecode, its a Gigabyte Technology Co., Ltd. H81M-DS2V 14:01:11 bah only supports 16G of ram 14:41:19 how to get the total writen bytes of an SSD ? 14:51:38 daemon: There is a good chance that max RAM statement is wrong. Many boards' specs are not updated as larger RAM modules become available. 15:13:54 occ look at smartctl in ports, and consider reviewing vendor-specific tools 15:14:15 ok, I got ipv6 working ... on the route 15:14:32 tomorrow, extend it to my PC so I can work instead of shaving yaks 15:18:45 bald yaks , bald yaks everywhere 15:18:47 dch: I used smartctl ,but it tells the TBW from system online,not the TBW of life. 15:18:55 s/ ,/,/ 18:43:55 rokit go zoom 18:44:08 whoops, wrong window. 20:38:20 Hello friends! Do any of you use tigervnc? I was wondering why there doesn't seem to be a service in any of the `rc.d` directories 20:39:21 I have used that, only as a client 20:46:08 Hmmm 20:46:34 I've used it extensively in Linux. They recommend you only start it from a service 20:56:22 "They" who, Linux system or "tigervnc"? What was the context? I had seen "tigervnc" suggest to use "systemd" on Rocky Linux for otherwise life is damn too hard (because of "systemd"). Sure same might have been, or could be, done for FreeBSD by the interested parties 21:02:47 I have used *vnc routinely off and on as needed. I have always run it as myself rather than as root. 21:02:59 I can't imagine the use model for starting vnc as a root system service. But I have a hard time fathoming many systemd decisions. 21:35:54 rwp: it's not always necessary to start services as root, you know that, right? 21:50:20 parv: "They" being tigervnc. The recommendations are the same regardless of Linux or BSD. When I try to start it without a service, it complains about "deamonizing failed," so I'll have to figure that out later 21:56:32 meena, Even if "root" passes control to "rwp" that seems like a silly place to do it. Better for me to simply start vnc up as rwp. 22:07:58 rwp: yeah, but if you start a service as you for you, it's still a service 22:18:31 I think parv had the right idea as to it though because systemd wants to start everything for a strict parent child relationship where systemd is the parent of every process and has a concept of user services. 22:18:51 It's very patriarchal. 22:19:27 But systems not running systemd have no such need for that strict requirement of being a child of systemd. 22:20:47 I am not of that body but this describes it in some detail https://wiki.archlinux.org/title/Systemd/User 22:20:48 Title: systemd/User - ArchWiki 22:24:12 OT Rant: On Linux/systemd type systems such as Ubuntu just doing a file transfer with sftp will launch for that user a lot of processes such as the audio server such as the tracker-miner-fs indexer and gvfs services and more. WAT?! 23:00:41 where do i find changelogs for minor release updates? 23:04:43 specifically i'm trying to figure out if the kernel changes to linuxkpi needed for drm-kmod PRs 163 and 208 (updating to 5.11 and 5.12 respectively) have been integrated into a release yet. 23:04:46 phryk: reading over the commit log generally gives me a good enough idea what changed 23:06:34 phryk: https://freshbsd.org/freebsd/src/branch/releng/13.1 but also, the UPDATING file should inform you about breaking changes 23:06:36 Title: FreeBSD / src - FreshBSD 23:16:24 meena: oh duh, good call, now i feel dumb for not thinking of that. :'D i assume -CURRENT is the main branch? :) 23:20:50 okay, was able to make some educated guesses about which commits in main correspond to what i need and can see in cgit that these are indeed not merged into releng/13.1 23:21:16 thanks for the pointers^^ 23:25:29 phryk: when you are interested to changes into releases, consider also reading release notes such as https://www.freebsd.org/releases/13.1R/relnotes/ . Of course, they are not as complete as the commit logs however 23:25:30 Title: FreeBSD 13.1-RELEASE Release Notes | The FreeBSD Project 23:31:39 salvadore: those are only for the initial release tho, aren't they? i'm looking for stuff happening *after* the initial release. 23:32:56 phryk: the order is CURRENT, STABLE, RELEASE. A CURRENT commit marked as MFC, means merge from CURRENT. 23:34:03 These are so marked if the developer has the foresight that it'll be good to have in STABLE branches and in a RELEASE 23:34:38 sometimes, commits are merged without MFC, and sometimes MFC commits are forgotten ;) 23:35:33 phryk: you can find post-release stuff int the errata pages, e.g. https://www.freebsd.org/releases/13.1R/errata/ 23:35:34 Title: FreeBSD 13.1-RELEASE Errata | The FreeBSD Project 23:35:55 meena: wait, stuff is merged into STABLE *before* RELEASE? 23:36:58 salvadore: i thought only security- and bugfixes are in there? i'm more interested in features. 23:37:03 phryk: CURRENT is for development, STABLE is for testing, and RELEASE is for running in production 23:37:57 meena: ah, i always assumed that STABLE would see even less merges than RELEASE, so I just avoided it^^ 23:38:25 normally no new features are added to a RELEASE after it is released. If you want more features, then you have to move to STABLE 23:38:31 nah, STABLE is for… stabilising 23:39:07 mhh, bit of a misnomer then, but i think changing it now would just add to the confusion^^ 23:41:49 mhh, the changes already seem to be in stable/13. but with what salvadore said, i assume this means they'll be integrated into the upcoming 13.2 but not the already released 13.1? 23:43:33 phryk: depends on what changes, but chances are: correct 23:44:06 yes, indeed 23:44:55 not 100% sure bit it sounds like it only adds functionality needed for newer drivers to linuxkpi. that might still mean somebody could install a too-new drm-kmod version and drivers on a kernel that can't handle them… tho i think pkg can detect kernel abi requirements? 23:45:08 s/bit/but/ 23:47:32 okay, now i'm even more confused. i just found the package for my new gpus firmware in the pkg repo… 23:48:11 i'll see if pkg installing it will raise an error about kernel abi, that would clear some of the confusion^ 23:48:32 nope, it installed just fine. :'D 23:52:14 pkg can indeed detect kernel abi requirements: if you do pkg info you will see a field called FreeBSD_version that is used for that 23:52:47 more information about FreeBSD_version can be found here: https://cgit.freebsd.org/src/tree/sys/sys/param.h#n63 23:52:48 Title: param.h « sys « sys - src - FreeBSD source tree 23:53:59 and a useful list of changes corresponding to the FreeBSD_version values can be found here: https://docs.freebsd.org/en/books/porters-handbook/versions/ 23:54:01 Title: Chapter 18. __FreeBSD_version Values | FreeBSD Documentation Portal