17:14:09 I already cloned for a previous port, but did it obviously wrong and am trying to correct it now. 17:16:22 Hurray, 13.2-RC2 is here! 17:16:25 how many commits are diverged? but it reads like it is easier to create your commits again manually instead of letting git doing that automatically with rebase 17:19:11 elgrandel: It depends on the choice of storage and interface; what's the latency distribution according to `zpool iostat -w` 17:25:51 elgrande: ⬆️ 17:30:27 https://bsd.to/BLDA 17:30:29 Title: dpaste/BLDA (Plain Text) 17:32:48 nimaje: only .gitignore is left diverging 17:36:49 https://bsd.to/HSZb ^^ 17:36:50 Title: dpaste/HSZb (Plain Text) 17:38:03 I'm trying to load a perl plugin for Hexchat, however I'm missing the perl plugin for Hexchat and I can't find a package for it. Does anyone know if there is an equivalent of hexchat-perl for FreeBSD? 17:41:53 Not bad for a laptop huh? 17:43:37 Hi 17:44:46 I want to use freebsd as desktop (personal use) and server. I found it tough to do the basic thing of sound, network, login screen etc. I come from linux. Its different. I heard there are premade distors around freebsd like nomad or something. Which one do you guys recommend? 17:46:12 i recommend freebsd; you don't need any crufty deskdop 'distros' 17:47:30 I reckon kde might be one of the better supported desktops 17:48:51 I'd say it depends on whether you can set up Linux with a desktop environment. 17:49:29 If you can do that, it's not gonna be as much work to learn FreeBSD. Otherwise, having to learn both to setup a DE as well as FreeBSD might be a lot, depending on the type of person you are. 17:50:21 And I don't mean install a distro that has a DE pre-configured, I mean starting from a non-GUI shell, installing and configuring everything, and ending up with a working environment. 17:50:58 It's not impossible, and the documentation (the handbook, specifically) assumes you don't know how to set up a desktop environment on Linux. 17:51:07 Valeria22: perl is already on in the default build: https://www.freshports.org/irc/hexchat/ 17:51:08 Title: FreshPorts -- irc/hexchat: IRC chat program with GTK and Text Frontend 17:51:20 But all the same, some concepts are going to carry over. 17:53:24 I know meena, however when I try to load a perl plugin I get this error: Unknown file type /home/valeria/.config/hexchat/addons/channel_highlight-0.3.pl. Maybe you need to install the Perl or Python plugin? 17:54:31 * meena doesn't actually hexchat 17:57:40 Valeria22: maybe ask in Hexchat support channel? 17:57:56 rtprio debdrup I see. I would love to but the time consumption is the only concern 17:58:03 might not be very FreeBSD specific 17:58:37 I'm considering that, but from what I've seen with Hexchat on some Linux Distros they need a package called hexchat-perl to use perl scripts. 17:58:56 I also see there are many features it misses? long configs of network, sound, user, no encryption at install time 17:58:57 etc 17:59:58 Sircle, "Encryption at install time" of what, the OS or the user $HOME? 18:00:15 Encryption at install time can be done Sircle, though AFAIK it's the whole disk. 18:00:30 parvXirc the OS (and home if needed) 18:00:44 Valeria22: we don't usually do that kind of thing. i don't why 18:00:45 Valeria22 don't think its an option in fbsd 18:00:51 install time I mean 18:01:07 Sircle, As Valeria22 mentioned that is (whole disk) absolutely offered during install time 18:01:20 one of my disks are in btrfs, does freebsd supports btrfs? 18:01:26 probably because the ports framework doesn't have a good mechanic to produce several packages from the same source 18:01:28 Ha, no 18:01:31 parvXirc ok. maybe I missed 18:01:34 Valeria22: do you have perl installed? 18:01:38 parvXirc oh no support for that? 18:01:42 Yes I do debdrup 18:01:58 parvXirc oh no support for that? (btrfs) 18:02:07 Sircle, yes 18:02:14 oh.. 18:02:43 There might be something in Ports (I do not know 18:02:46 Valeria22: does ldd report that hexchat is linked against perls libraries? 18:02:51 ok 18:03:13 Wait, w ould it be? Since perl is interpreted. 18:03:32 Nevermind, that might be a dumb question. 18:04:23 Sorry, I'm used to irssi where perl support works. 18:06:18 I assume the invocation for building hexchat from scratch is similar to what's in the Makefile? https://cgit.freebsd.org/ports/tree/irc/hexchat/Makefile#n66 18:06:20 Title: Makefile « hexchat « irc - ports - FreeBSD ports tree 18:07:45 Give me a moment, I'm trying to figure out how to use ldd. 18:07:55 ldd `which hexchat` 18:10:15 https://pkg-status.freebsd.org/beefy18/data/main-amd64-default/p6a11326d040c_s44b31d37ad/logs/hexchat-2.16.1_3.log here's the buildlog for hexchat (though it's for the latest packages for 13.1 on amd64, that shouldn't matter) and it confirms perl is enabled 18:10:41 interestingly, it says that "Perl legacy API" is false though 18:10:53 maybe that's why? 18:12:55 Maybe? The plugin was last updated in 2010. 18:13:14 domlaut[m]: that's right, I forgot to mention this when we were talking about package builds the other day; the poudriere instances themselves are accessible via ipv4; just put the instance name into pkg-status.freebsd.org/instancename/ 18:14:26 Valeria22: You'll have to ask in #hexchat whether the output from that log where it says "Perl Legacy API: false" is what it should be 18:15:05 also, that's for 14-CURRENT, not FreeBSD 13.1-RELEASE.. but again, it shouldn't matter 18:15:23 Got it, thanks for your time debdrup. 18:15:29 No worries. 18:22:16 debdrup: cool! thanks 18:23:15 still waiting for a committer to merge my patch so not much to look at until then :-) 18:28:25 domlaut: did you email the maintainer? 18:29:28 yeah. I email to double check whether it'd be okay if I asked for maintainer-approval (since I missed to do that, and he already gave maintainer-feedback+) and if I needed to bump PORTREVISION 18:29:53 so I submitted a new patch that also bumps PORTREVISION, asked and got maintainer-approval+ 18:30:49 oh I also asked whether he's in contact with any committers around the port or not, he said they'll handle it when they see maintainer-approval+ 18:35:25 alrighty then 19:56:47 I <3 syslogd 20:02:42 Hi. I am a linux user and I am serious considering moving to freebsd. I have a question. How is it to run the linux emulation layer to run few packages from linux that don't run on freebsd like the brave browser and to watch netflix or the tor browser bundle in freebsd? 20:05:49 for tor browser there is https://www.freshports.org/www/tor-browser/ not sure how well that drm stuff likes running under linuxemu, but I think it should work 20:05:51 Title: FreshPorts -- www/tor-browser: Tor Browser for FreeBSD 20:08:17 I don't need the tor browser to run DRM. As long it just works. As for DRM I would prefer brave, as that's my preferred clearnet browser 20:10:03 What is freshports btw? 20:10:24 The best ever website anyone has ever seen. 20:10:28 Information and build statuses on packages. 20:11:27 Ah ok. I thought it was an information about existing ports in freebsd.. or a Archlinux AUR or Gentoo GURU / Overlay situation where freebsd has a "third party community ports" thingy 20:12:00 does tor browser even support that drm stuff? that comment was about brave, as using native tor browser wouldn't have drm support as that needs some properitery binary that isn't build for freebsd 20:12:20 I know this is controversial question but related to the above. Is it possible to run flatpak in FreeBSD with the linux compatbility layer? Does it work? That could solve the hurdles I have for few specific packages 20:12:36 flatpak is a linuxism 20:12:52 debdrup, and does the linuxism work with the linux emulation is my question? 20:13:06 I got packages and ports swapped, my bad. But the site does have an FAQ icarious: https://www.freshports.org/faq.php 20:13:07 Title: FreshPorts -- FAQ (Frequently Asked Questions) 20:13:33 Thanks Valeria22 20:13:37 and lol @ VimDisel 20:14:41 icarious: if you can ask that, it means you've more to learn about the Linuxulator ;) You'll quickly learn that Linuxulator is a syscall translation layer that simply translates a subsection of all syscalls in Linux into their equivalents in FreeBSD. 20:15:37 :S .. Ok I will rtfm, But thanks for pointing me towards urmm Linuxulator 20:17:08 It's somewhat ad-hoc what gets added to Linuxulator, because what ultimately controls it is what's needed to make Linux binaries run - but that doesn't mean that support for flatpak is likely to exist or be added, I think. 20:17:42 Flatpak is, more or less, a way of getting around the issues of dynamic libraries and such - and FreeBSD has a completely different solution to that. 20:19:05 In FreeBSD, packages are _ALWAYS_ built in a way that ensures that dependencies of a leaf are rebuilt if the leaf is bumped, so things like dynamic library incompatibilities aren't really an issue (there may be exceptions, of course - there usually is) 20:20:12 Flatpak is reminding me of PC-BSD. 20:20:21 pretty much 20:20:37 I never used that, so I have no idea what that means. :P 20:21:09 they did something similar to flatpak called a PBI 20:21:20 prior to pkg-ng 20:21:23 That was iXSystems' desktop distro 20:21:46 after pkg-ng it didn't really make much sense anymore 20:22:37 Anyone have ideas on what might make "nice" not work in a jail? 20:22:40 Mar 13 16:20:05 php8 /usr/sbin/cron[94422]: (root) CMD (/usr/bin/nice -n 5 /usr/local/bin/php -q /usr/local/www/froxlor/bin/froxlor-cli froxlor:cron 'letsencrypt' -q 1> /dev/null) 20:22:40 Mar 13 16:20:05 php8 cron[94422]: setpriority 'root' (daemon): Permission denied 20:23:06 Googling just brings me to a bug I commented on back when 9.x was current. 20:24:00 And what about the argument that Flatpak does solve "albeit a decade old" problem of isolation that Android has much better over some of the Linuxes and BSD's. What is I want to run a specific package all isolated in userspace instead of being all root 20:25:27 noting that the "setpriority" error only happens when run from cron... 20:26:34 icarious: jails are a much older solution to that problem than even Android has (and is, in fact, the first such solution) 20:27:21 debdrup, I am curious. Can I run apps in jails as regular user? Does it isolate from even the user's home dir 20:27:23 ? 20:27:38 If it does it, then idc about flatpaks. 20:27:51 icarious: a jail is way of confining the root user (and all other users) 20:28:01 Ok i need to test it. But thanks 20:28:25 I always found flatpaks just a tad bit better (although a decades older to what Android does) in terms of app isolation.... 20:28:33 Seems like FreeBSD might be my opsec go to now 20:28:46 FreeBSD jails are about as old as Google is. 20:29:59 The MAC feature of FreeBSD. Is it workable like SELinux is in redhad based systems with their targeted policies? 20:31:07 There was a fork of FreeBSD called SEBSD, which implemented SELinux notions using the MAC framework - but it never really caught on, so "all" you have is the regular MAC framework (which is still incredibly powerful). 20:31:26 Ah ok 20:36:03 The SEBSD thing reminds me. Why is it that BSD "distros" don't survice under BSD projects as they do with Linux? For instance recently TrueOS died too 20:36:43 Time and/or money. 20:37:20 Hmm. Its sad. BSD projects specially FreeBSD can do with a few custom distros 20:37:43 See, I think that's a mistake. 20:37:47 same 20:37:56 fragmentation for no gain/purpose 20:38:41 they're not improving the base OS, they're just changing some userland, so why not just stick with FreeBSD and do your userland customizations after install like everyone else 20:38:51 Think of all the work that's been put into making projects based on FreeBSD; all of that work could've been put into making FreeBSD better, but either someone wanted to be the BDFL of their own project or they didn't succeed in finding the right people to mentor them into FreeBSD. 20:39:01 Actually the linux ecosystem is definitely fragmented. Too many distros doing nothing. But For instance a project like FreeBSD can have a let's say workstation / desktop or security focused distro ecosystem 20:39:04 If not too much 20:39:45 dkeav, another way to avoid the fragmentation would be if freebsd allowed official customisations / spins then. 20:39:52 which would ammount to basically a meta-pkg 20:40:08 don't even need a "spin" 20:40:29 There's some exceptions to this, of course - iX has been fairly good about upstreaming what could be upstreamed, but they still started out with the notion that what they were making should be something produced by them. What they could've done would be to build everything into the ports system and used the same tools as used by the release engineering team to produce FreeBSD RELEASEs. 20:41:47 icarious: the license two clauses: Don't claim you wrote code you didn't write, and if it breaks you get to keep all the pieces. 20:42:05 the word 'has' got lost between my fingers and keyboard, apparently 20:42:36 icarious: everyone can make a fork of FreeBSD and customize it to their heart (in fact, many have, that's why we have all these things that're now defunct) 20:42:46 ya I guess. there is no incentive. The system, is perfect in its default form. just rtfm and build you own system. I get it 20:43:01 *your 20:43:03 Nah, there's always room for improvement. 20:43:17 Thing is, FreeBSD as an OS is a different kind of an OS than other OS'. 20:43:28 with spins you mean changed ports options? set up some build cluster and provide your pkg repo with your changed options to everyone if you want 20:43:30 but improve the project instead of trying to re-invent the wheel entirely 20:44:08 Oh, I guess there's also one other thing you can't do: You can't call your project FreeBSD. 20:44:20 Actually the spotlight is always on FreeBSD as a BSD project cause its the most multipurpose universal one. The flagship of the BSDs that anyone can use 20:44:31 The foundation would, quite literally, be forced to stop you - because they hold the trademark. 20:45:42 Well, that's not necessarily a bad thing. But I guess a Fedora like approach for "Spins" Might be a more modern approach to prevent fragmentations. 20:46:44 Honestly, I am not ranting. I do still think the project is near perfect as it exists. To the point that it just keeps on reminding Linux what it should have been. But a bit of modernity and flexibility isn't that bad 20:47:13 That reminds me of the thing i never finished 20:47:24 ..writing about 20:48:06 FreeBSD as an OS is more of a relatively small toolbox, with an extensive set of tools available down the line in the form of the ports/packages. 20:48:35 By itself, it's only got the compiler used to build it, a few editors, and a fairly decent set of device drivers for the most essential peripheral devices. 20:48:54 That's what an OS used to be, a very long time ago. 20:49:06 Onething I do want FreeBSD to succeeed beside linux would be device / driver support. But its not the project's fault. We need those bloody corporate that Linux enjoys to lend some hand 20:49:17 That has strengths and weaknesses, of course - but so does the other end of the scale. 20:49:29 Linux's too big to fail isn't a good thing 20:49:46 icarious: thing is, Linux peripheral device support doesn't come from Linux being technically superior, it comes from companies using Linux as part of a product. 20:50:29 And it doesn't guarentee high-quality code. Have you looked at the code Realtek is shoving into the kernel? It reads like something that's machine generated from internal documentation. 20:50:36 debdrup, I know man. Its not a strength of linux. that's why i said i don't blame freebsd as a project. those companies should lend a hand now 20:51:02 If they did, which they won't, their code still has to conform to FreeBSDs style(9). 20:51:06 Oh god. Linux's (as in the kernel and gnu c library) code quality is horrible. trust me i know. one of the reasons I am looking to move 20:51:12 the attack surface just keeps on increasing 20:51:28 The code would also have to be copyfree-licensed. 20:51:34 liks a simple program like "ls" in gnu coreutils has 5000 lines of code whereas in freebsd its just 1000 if i am correct 20:53:27 I'm not convinced that lines of code is a good argument, because it's a very essentialist approach that quickly becomes reductionist to the point that a developer says "you don't need X" where X can be anything from features in a program to small things like being able to supply the program with a config file, so that you're instead forced to recompile the program if you want it to function differently. 20:54:22 There's a careful balance between lines of code vs functionality gained from that number of lines of code, and going too strongly in either direction doesn't seem to me to be a good solution. 20:54:26 debdrup, what about in an archaic memory unsafe language like C? I am not refuting you. But still. You don't agree it's still a thing to write clean code? 20:55:34 icarious: the advantages of replacments to C aren't as big as one might like; if you want revolutionary ideas, you'll have to start by using a CPU that doesn't behave like a PDP-11 from the 1970s. 20:56:07 CheriBSD, as far as I understand it, address all the biggest issues of C - and it's done in real hardware like the ARM Morello. 20:56:10 * rtyler is confused by the topic 20:56:22 well. that's not a bad argument. I mean I don't completely disagree 20:56:24 rtyler: I think thre's a bit of drift going on. 20:58:31 icarious: "Clean" code is a suspect term, I think; people (at least FreeBSD committers) don't introduce errors intentionally, knowingly. They do so because of mistakes, and for that there's the usual solutions of code review, static analysis, compiler warnings (unlike a lot of opensource projects, FreeBSD uses WARNS=6 to make builds quite noisy when something is potentially bad), fuzzing, sanitizers, and 20:58:37 things of that nature. 21:00:05 debdrup, i didn't know about the WARNS=6. So would you say FreeBSD followes better practices than the GNU project in general and as a system is better posed towards security? 21:00:12 *follows 21:00:38 icarious: I can't comment on that because I know almost nothing about GNU as I used Linux for all of 1 week in the year 2000 before switching to FreeBSD. 21:01:10 Ah ok 21:01:16 I'm also not a programmer, I'm a documentation person. :P 21:01:26 You are like the polar opposite of me. I am like a primarily linux user. 2003 - Present 21:01:27 lol 21:01:56 A linux user looking to move to freebsd cause i am fed up with systemd and linux's general direction 21:02:00 What I will say that it's fairly well-established that ignoring warnings from compilers hasn't been known to produce secure code. 21:02:40 https://cgit.freebsd.org/src/tree/share/mk/bsd.sys.mk#n31 this is the source for the WARNS thing in case you're curious 21:02:41 Title: bsd.sys.mk « mk « share - src - FreeBSD source tree 21:04:51 I think CheriBSD builds with even stricter warnings, and upstreams just about everything that isn't part of CHERI project. 21:06:02 CheriBSD? Goodness gracious there's like more FreeBSD based distros than I knew. Where are these even documented? 21:06:37 CheriBSD is a project run by Cambridge university in the UK. 21:06:55 It's not really a distro in the same way the others are. 21:07:21 I mean PC-BSD / GhostBSD and DragonFly are the majors ones I know of 21:07:36 And as for OpenBSD / NetBSD I don't even know if distros even exist 21:07:37 DragonFlyBSD isn't a distro, it's a real fork. 21:08:06 NetBSD is a fork from the same base FreeBSD came from (though the other end of it, but that's a long story), and OpenBSD is a a fork of NetBSD. 21:08:10 debdrup, the term "distro" needs a closer analysis. I do get it that the BSD world hates the world "distro" . But the "D" of BSD still acknowledges it 21:08:17 *word 21:08:47 NetBSD, OpenBSD, DragonFlyBSD and FreeBSD are _completely_ separate OS' where code occationally flows between them. 21:09:17 I always took Dragonfly as a FreeBSD distro 21:09:19 They're not distros, in the sense that Linux distros all use (roughly) the same Linux kernel and different libraries+userland. 21:09:22 May be I am wrong 21:09:22 It's not. 21:09:34 I'm pretty sure the DragonFlyBSD people would feel rather insulted by that. 21:10:34 If you know your Linux history, it's equivalent to saying Linux is a fork of MINIX. 21:11:02 Well, I think so, I don't really know Linux, nor MINIX. :P 21:11:03 Well. Doesn't matter what they feel. If I am not wrong they forked from FreeBSD 4.8. And the BSD guys should learn to accept what is a "project" and what is a fork / distro. The "D" itself in BSD still legitimizes the word "Distribution" 21:11:15 Unless they have started to deny their "BSD" origins 21:11:22 They can then deny the "D" 21:12:05 We're gonna have to agree to disagree here, and I'm hate to say so, but I'm pretty sure a majority of people who use one of the BSDs agree with me. :P 21:12:31 Illumos is a BSD too 21:12:40 ... 21:14:13 debdrup, What is the "D" in a BSD? 21:14:14 trolly trolly 21:14:37 when BSD was coined the concept of a distribution was the distributing of source 21:14:54 the "Distribution" in BSD has no "of", it just means some software together 21:14:58 icarious: if you want a semantics discussion, take it elsewhere. 21:16:10 I mean is it correct to call FreeBSD as the "Free Berkeley Software Distribution" without offending anyone? 21:17:21 debdrup, are you consciously asking me to not exercise my freedom of speech and opion of facts and just take it elsewhere? 21:18:22 icarious: are you familiar with the linguistic concept of a word having multiple meanings depending on context? And for the record, you don't have freedom of speech on IRC. 21:18:55 * rtyler laughs 21:19:15 If you continue trolling, I'm going to be forced to either mute or kick you. 21:21:14 debdrup, I asked you a simple question which you never answered and kept on insuinating that I am trolling you. I asked you what the "D" stood form in FreeBSD 21:21:21 debdrup, you can kick me 21:21:34 but do it under the pretext and reasoning that you did it while diagreeing with me 21:21:42 and you are a supporter of censorship 21:21:45 icarious: context-sensitive meaning is the answer. 21:21:46 and you hate healthy debates 21:22:08 That sounds like Soviet Communism to me. "I will vaporise you cause you don't agree with the party line" 21:22:18 You just gave the project a bad name 21:22:23 Hahahaha 21:22:36 Oh it's been a while since we've had a troll like that, that was almost fun. 21:22:40 Shame they gave up so quick. 21:22:51 Screed-to-quit ratio too high. 21:29:38 currently trying to figure out journaling. can i use gjournal on a ufs gmirror? my particular fs onion would look something like ufs(gmirror(geli)) 21:29:41 debdrup: so you're +o in here? 21:30:00 rtyler: I'm a FreeBSD committer. 21:30:23 phryk: ufs does softupdates+journaling by default, as of many years ago. 21:30:58 debdrup: so no manual mucking about with gjournal? just do newfs and I'm good? 21:32:35 I mean it's been the default in bsdinstall; if you're using newfs, you'll wanna use the appropriate flags for it. 21:35:11 https://cgit.freebsd.org/src/commit/?id=4a8b3e41f5d it's been the default since the start of 2018 21:35:12 Title: src - FreeBSD source tree 21:35:34 debdrup: so newfs -jU, anything else I have to watch out for? 21:36:10 I haven't used UFS since 2007, so I'm not the right person to ask for that. 21:37:03 alrighty. i assume if i do something wrong, I'll probably be able to rectify it with tunefs afterwards… so it hopefully won't be too bad. :P 21:40:31 spork_css: too much noise in the backlog to see if you had been given an answer, but setpriority(2) in a jail only lets the nice value go up it looks like 21:41:33 kevans: is that documented somewhere? 21:41:57 spork_css: it ends up here: https://cgit.freebsd.org/src/tree/sys/kern/kern_resource.c#n268 and there's no jail policy set for PRIV_SCHED_SETPRIORITY (it would be in kern_jail.c) 21:41:59 Title: kern_resource.c « kern « sys - src - FreeBSD source tree 21:42:00 no idea 21:42:18 well, I guess it kind of is in the manpage 21:42:30 "Only the super-user may lower priorities." 21:42:49 Which manual page? 21:42:54 setpriority(2) 21:43:34 So I'm still gonna feel justified in adding an item to my TODO list to add it to jail(8). 21:43:54 I would maybe add some wording to nice(1) as well, I note that renice(8) already talks about it a little bit, too 21:44:27 (last paragraph of the description talking about only being able to monotically increase their nice value) 21:45:07 kevans: for a (whole) process 21:48:29 ? 21:50:31 kevans: that's what the comment above says 21:50:50 ahh, yeah, sorry 21:51:13 * Set "nice" for a (whole) process. 21:51:41 how do you set nice for less than that? i would've thought that's like the bare minimum 21:52:44 that reminds me: is it possible to create login classes ad-hoc, on the fly, or do they need backing from login.conf? 21:53:27 cuz i'm fairly certain you can just setuid(rand()) and get away with it not having backing in passwd 21:58:34 yeah looks like you can just kinda set one and it'll get created 21:59:21 see https://cgit.freebsd.org/src/tree/sys/kern/kern_loginclass.c#n207, loginclass_find will allocate a new loginclass if it's not already in the list 21:59:22 Title: kern_loginclass.c « kern « sys - src - FreeBSD source tree 22:00:13 nice is a whole process concept but you can set priority more granular than that 22:01:47 with, e.g., rtprio_thread(2) 22:09:30                 uint16_t pad_len = min_pkt_len - m_head->m_pkthdr.len; 22:09:30                 if (!m_append(m_head, pad_len, pad)) 22:13:09 kevans: ah, right. threads… duh 22:27:39 okay, so i partitioned the drives, did geli init for everything that's supposed to be encrypted, and added gmirrors… however when i run newfs on any of those mirrors i get "newfs: Can't stat mirror/fastread: No such file or directory" after the long list of "super block backups"… should i worry? can i ignore this? 22:29:34 ah, might be a bug. only triggered when using paths relative to /dev (i.e. mirror/) – when i use /dev/mirror/ this doesn't happen… 22:34:52 I think I figured out the issue to my problem with Hexchat thanks to some help from their IRC channel. In /usr/local/lib/hexchat/plugins there are three .so files that run at startup, one for DDC's checksum, one for perl scripting, and the last for python scripting. The checksum plugin works as expected however when trying to start the perl and python scripting plugins the following errors are reported: "PL_thr_key" for perl.so and 22:34:52 "PyCapsule_Type" for python.so 22:35:46 Undefined symbol (error message) <- Forgot that, sorry. 22:46:56 Valeria22: that sounds like there's no threading enabled in your version of perl? 22:48:11 Valeria22: the python issue sounds like a version mismatch 22:49:08 so, like, maybe the plugin is expecting 2.7, but it's getting 3.9, or vice versa 22:49:21 I'm trying to use pkg query to check what versions of perl and python I'm currently using. 22:50:38 I used pkg version instead. perl is using 5.32.1_3 and python is using 3.9.16 22:50:57 Valeria22: where did you get that plugin from and did you compile it? did it come like that? 22:51:43 I believe that the plugin comes with Hexchat? I used pkg install Hexchat last night around 8:30-10:00 PM last night. 22:51:50 PM PDT* 22:52:37 pkg info -l hexchat should show you 22:53:28 Yes, Hexchat comes with those .so files for perl and python by default. 22:55:16 My version of Hexchat is currently 2.16.1_2, but I see the latest is 2.16.1_3 22:56:07 I feel like I'm really close to knowing what the answer is here, but also too tired to try any harder 23:20:29 I screwed something up 23:20:41 I accidentally unpacked base.txz in a directory 23:20:51 and now certain files won't delete, like certain libraries, and init, 23:20:59 because of some kind of Operation not Permitted 23:21:08 rm -rf won't delete them 23:21:11 what do I do please? 23:21:29 DrKK`: chflags(1) 23:21:36 ok 23:21:40 thanks 23:21:42 you're looking to remove the schg flag 23:22:06 I assume I can do this recursively 23:22:12 yes 23:22:14 looks like -R 23:22:15 thanks 23:23:31 ah yes that was very easy 23:23:37 chflags -R 0 [dirname] 23:23:38 done 23:23:45 thanks again 23:37:38 kevans: I am trying to make something "more nice" yet it fails, and oddly only when called from cron. 23:38:16 Mar 13 19:35:12 php8 /usr/sbin/cron[9511]: (root) CMD (/usr/bin/nice -n 5 /usr/local/bin/php -q /usr/local/www/froxlor/bin/froxlor-cli froxlor:cron 'tasks' -q 1> /dev/null) 23:38:16 Mar 13 19:35:12 php8 cron[4710]: setpriority 'root' (daemon): Permission denied 23:38:48 (also someone should maybe look at nuking that froxlor port, it's nowhere near functional - would need a LOT of work to be install and go) 23:39:24 run directly: 23:39:26 [root@php8 ~]# /usr/bin/nice -n 5 /usr/local/bin/php -q /usr/local/www/froxlor/bin/froxlor-cli froxlor:cron 'tasks' 23:39:26 Checking froxlor file permissions...OK 23:39:26 Running "tasks" job 23:39:28 [root@php8 ~]# 23:43:40 I keep googling and ending up in a bug report where I have a comment: https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=124248 23:43:42 Title: 124248 – [jail] [patch] add support for nice value for rc.d/jail + rc.conf 23:52:33 spork_css: to be clear, what's the nice of the cron process? 23:54:29 hmm, shouldn't matter