00:18:03 *sigh* et tu, freebsd 00:47:22 ehh> 00:47:24 ? 01:21:22 ws=\w 04:02:21 Oh, no. 04:04:58 anth: ? 04:06:45 I went looking for what Ltning might’ve been talking about, and made a guess. Lots if talk in the past day anout seeing “AI” “contributions” in freebsd (although possibly notnin main yet? I haven’t been able to validate yet). 04:07:41 FreeBSD was my first unix, way back in the late 1900s; I run it personally and for clients. 04:08:32 I don’t know how to communicate to the foundation or core how assuridly this will have me moving to another BSD and urging everyone else to. 04:09:34 anth: Do you count AI assisted as AI generated? 04:16:39 the whole thing is pernicious, because absent extremely careful review, most "generated" code looks superficially acceptable (by its nature) but then has egregious issues when actually thought about 04:17:03 the difference between "assisted" and "generated" is quite subjective, and most of the aggresive AI booster types will call the latter as the former 04:17:59 it's usually pretty obvious when it's being looked for, but imo anything more than "accepted a few lines as a completion" is straying into the dangerous / problematic domain, especially for core SW that is supposed to be reliable 04:18:23 like I wouldn't be too unhappy about slop packages or whatever in the ports tree, but the core stuff should be held to a higher standard 04:21:40 xales: right,I’m not religious about packages or ports. But core needs to be held to the highest standards. 04:21:50 Bah, I’ve got to. 04:21:58 Go to bed. 04:59:01 Yeah, "AI" in FreeBSD is not great and would be enough for me to start looking to others after using FreeBSD since version 5. 04:59:43 And I hope they ignore teh age verification BS too. 05:07:02 luser: Meanwhile... https://www.youtube.com/watch?v=6h_i8ObqA8o 05:30:53 Isfand: yeah I'm aware they wrote a quick sqlite DB app. I wouldn't move to MidnightBSD, I would likley look towards OpenBSD or NetBSD. 05:32:09 And whoever that is talking has an irritating voice. 05:32:36 I need ZFS, so all my other options are bad 06:43:03 "… it's very easy to pick a side, and then be outraged that the Project hasn't taken *your* side." – 06:44:49 It's not about picking a side, it's about what I'm willing to tolerate on my computers/network. It may end up being a storm in a tea-cup. 08:25:57 For a port Im currently looking into WRKSRC_group but cant figure out how to specify the extraction directory. The file doesnt seem to be extracted at all, the group is also set in DISTFILES and MASTER_SITES 08:27:48 I could ofc manually extract it in post-extract but I dont think that a good way 09:51:04 Is there a seperate Administrator user kind/account in BSD (like in MacOS) or is the only User & Root like in Linux ? 09:53:34 Maccampus: usually `root` is used for superuser 09:54:31 what macos has is the root user, regular users which are administrators and can su root, and then non-admin uses which cannot su root 09:54:35 So is there a separation betwween Superuser & Admin ? 09:54:55 so i think you could recreate that easy enough by creating two user groups, and giving one sudo access and one not 09:55:02 you mean like the wheel group? 09:55:34 or the operator group? 09:56:44 I guess, it's in BSD like in Linux then where there is no admin User ? 09:57:19 Can i ask how this was in original Unix & nowdays in Solaris ? 09:59:49 And maybe more important how is Apple Intel Hardware support in BSD, mainly Wireless 'wifi' and nvidea/intel dual gpu on older Macbooks '2012-2013' 10:02:22 dunno how "original" you want to get, but Edition 7 had "root" and "everything else". sudo did not exist 10:04:58 Anything else was non root, none admin you mean ? 10:05:28 You need to log in as root to do admin/root tasks 10:06:03 Correcction: need=> neded and ? at the end 10:06:21 at some point the wheel group was added to allow su to root and probably other stuff 10:06:58 And you say whenver my Mac gui ask for my admin pasword it is actually doing sudo without telling me ? 10:16:08 I ask s it seems weird to gui login as root 10:16:16 s = as 10:18:52 this was before GUIs 10:19:33 Also i seem to recall in early MacOSX there was the ned to activate root so Admin was not equal to root 10:20:49 no idea about stuff from apple, but afaik you can't become root there anymore 10:24:24 exactly, so whereto does Administrator elevates then, or did they move needed rights to Admin kind of accounts to make root unneeded. (this is one of the reasons i am investigatin macOS replacement candidate) 10:25:18 i bet if you look at the sudoers file on your mac, it will answer this 10:25:19 FreeBSD seems closest relative to MacOS 10:28:59 I don't think mac uses sudo for that, maybe look into https://en.wikipedia.org/wiki/Mandatory_access_control for the general concept 10:29:29 nimaje: an Apple user here. macos does use sudo 10:29:50 Maccampus: macos is actually the Darwin distribution 10:30:23 I think technically it's a derivative of freebsd plus some code from the original bsd 10:30:32 macos x came from next step, and next step was based on freebsd wasn’t it? 10:30:40 and plus some code they got from their own 10:30:42 or was it more of a freebsd/nextstep merger into macosx 10:31:09 Darwin is actually dead or atlest on lifesupp as Apple crippled in in 2017 10:31:09 its been so long i don’t remember the exact details anymore 10:31:11 GoSox: I think it's got some code from the $1000 tape source code back then 10:31:25 scottpedia: it does use sudo for those gui password dialoges when the admin needs to confirm an action? 10:32:10 i’m here because i’m still running a macos x web server which is insanity in 2026. And theres no sane way to replace my mac web server with a newer mac web server. And freebsd is likely the easiest platform to move to 10:32:13 nimaje: I can confirm some of them are like that. like when you try to flash an SD card with a distro image for example 10:32:14 GoSox, Next was 4.3Lite & switched to FreeBSD at version 4 aka OpenStep 10:33:03 some of the settings that require password are those made by apple. for example changing ip addresses in the CLI doesn't take root privileges 10:33:23 And Aple added OpenBSD to the mixure, but main source of darwin came from NeXT/FreeBSD 10:33:24 but in System Preferences you have to at least "unlock" for the change 10:33:58 Maccampus: did you look that up? 10:34:13 on wiki 10:34:26 not a 100% trust source 10:34:42 should be accurate 10:34:47 thx for the info 10:35:30 just still wondering was the Admin type added by NeXt or by Apple when converting NS to MacOSX 10:35:52 wat Admin type? Maccampus 10:36:02 Admin user type 10:36:54 should be the GUI's way of saying "root" 10:37:28 when you use smb sharing for example you can adjust the access rights of each account and there is an "Administrator" account I think 10:38:25 looking at the sudoers file on the Mac i’m on now, it looks like admin users are just users in teh group ‘wheel’ 10:38:29 like thats what makes them admins 10:38:29 seems more like a way to say "in the wheel group" in this context from what I understand 10:38:44 for the sharing (smb) administration, not the system wide Administration 10:38:54 ? 10:39:16 sorry it's technically called "System Administrator" (I just checked) 10:39:41 which i thought was standard and thats how i set up my freebsd server. just add myself to wheel and give group wheel sudo access 11:05:54 History of FreeBSD and macOS 14:05:59 https://discuss.linuxcontainers.org/t/incus-6-23-has-been-released/26447#p-89931-freebsd-vm-support-6 14:06:11 this could help convince more people to use freebsd 14:06:17 however I am a little confused by this: 15:15:17 sorry forgot to paste it 15:15:30 > Daily builds of FreeBSD 14 and 15 are availabe on our image server. 15:15:55 FreeBSD base only changes (ABI wise) like once a year 15:16:04 and security patches are released as needed 15:16:16 why the fuck would you compile the image every night, this is not linux where packages change on a daily basis 15:16:24 freebsd base remains the exact same unless there is a security patch 15:18:32 :D 15:24:01 it can be much easier to just build every day than to handle the details of how to do it more efficiently 15:46:01 polarian: it is a demonstration of taking a concept in one domain and thinking it applies to another domain.. it would be fair to say that -CURRENT is updated quite frequently and one could even take it one step further and say that one could build a custom base kernel, with optimizations.. and demonstrate that capability. You can build a daily baseline version, with anything you want. 15:47:18 the concept of building and re-building is quite common for a kernel + a suite of applications. It provides a feature/tool for a problem that may or may not exist in a different operating system eco-system.. but you could do it, if you wanted to. 15:47:22 hodapp: not when the compute costs you hundreds of quid a month 15:47:53 and open source projects never were good with money imop 15:47:55 imo* 15:48:10 I see the financial statements of some of these open source projects, and they run up hundreds of dollars in servers they realistically didnt need 15:51:52 financial discipline is a different issue 15:52:24 throw money at a problem or think through it and determine.. should i try money at the problem or re-engineer the solution to the problem (that may or may not exist) 16:11:03 polarian: cost of "spike of compute once a day" tends to be so close to "spike of compute at some lesser interval" as to make hardly any difference here 16:11:24 and I seriously doubt it's requiring some higher tier of anything to accomplish the former 16:15:18 hodapp: you are forgetting the norm is serverless 16:15:23 the less you need to compile the cheaper it will be 16:15:41 and in any case, say they do have a dedi for all the compiling, and do not do pipelines 16:16:05 you would rather that dedi be compiling linux images like arch linux more than stable releases like freebsd/debian etc 16:39:31 does their image server definitely run serverless? 16:45:58 if they're running on a dedi or a VPS, chances are the bill is the same regardless of what they do on it, barring cases of excessive bandwidth use over what's metered 16:57:33 hello fellow freebsders 16:57:39 hi 16:59:19 so here's what i already have. i have a box with an IP address running freebsd 15, and on it I have installed and configured ZNC. 16:59:39 nice 17:00:02 here's what i'm thinking: I want ZNC to only connect to the internet via a wireguard vpn tunnel 17:00:39 but the machine should also be open to the public internet so i can connect to it via an IP address.. so i'm thinking i could maybe put ZNC in a jail 17:02:25 and then after many hours of effort im planning on giving up and just installing templeos... lol jk 17:03:05 i'm not asking to hold my hand, i'm mostly just asking is this even reasonable to want? 17:04:38 agent314: sure, it's not clear what you're missing that made it that difficult 17:05:21 it's probably overkill but I don't think it'd be all that difficult 17:05:22 if you only want it to have wireguard access you could even use a simpler setup: vnet jail, have the host create a wg interface and give it to the vnet jail, then the jail can configure the interface/peer and you're office to the races 17:05:33 s/office/off/ 17:05:43 you don't even need epairs that way 17:05:44 to step back from the details and sorta expound on my motivation: i like having a ZNC bouncer so i can collect every single message where somebody calls me stupid in #gentoo 17:05:55 o_O 17:06:12 but i dont like how the bouncer is like tied to my VPS IP 17:06:29 yeah i believe vnets were discussed 17:06:47 why aren't you connecting with a cloak? 17:07:06 also FIBs came up throughout my research 17:07:21 hodapp: tbh i'm not entirely sure what a cloak is 17:08:27 also whatever it is, my znc will still be attached to my vps ip, even if that ip would only be visible to the IRC server? 17:09:42 what im having difficulty is understanding how exactly will znc be able to bind onto my VPS_IP and yet also route all the networking through a wireguard 17:10:12 h/o ive made some progress, i was following some article let me see what im currently on just a sec 17:11:38 unless i'm interrupting another important discussion that was happening before, in which case lmk and i'll come back a bit later 17:16:47 agent314: you -> wireguard -> znc = the ircd sees the IP of the VPS. you -> znc -> wireguard = the IP sees the IP of the wireguard node 17:17:04 the ircd sees the IP of the wireguard node 17:17:22 oh another thing same server is running mosh/ssh thingy on port 5188 and obviously id need to keep that going, but also a regular web-server on 80/443, and apparently i'm also running a 1080 socks server which i dont even remember running, hold on im gonna run me nmap -p- on this server what other things does it have going on lol 17:17:59 ialokin: yea thats the plan 17:19:07 but i want to be able to connect to the VPS_IP:6667 17:20:19 so znc would have to bind to VPS_IP on port 6667 and i suppose that connection will all happen without wireguard cloak? but then it turns around and sends messages to say Libera and that i want to go through wireguard 17:20:52 i think it'll be easier if i draw a diagram (mainly for me) 17:28:19 https://docs.freebsd.org/en/books/handbook/jails/#creating-vnet-jail 17:33:09 agent314: it's a per-network thing but e.g. on Libera you can ask for a cloak that hides your IP address 17:33:14 it's just i noticed i discussed this with llms quite a bit but i havent had any breathing, living, eating homo sapiens tell me what they thought of it 17:35:08 hodapp: i definitely know of servers where cloaking is much more of a big deal like some anon-related ones (mostly extinct now but yeah) 17:35:49 hodapp: also who does it hide the IP address from? 17:37:11 agent314: e.g. from /whois agent314 17:37:18 which right now shows your IP 17:44:08 hodapp: does it? :) 17:44:32 hodapp: can you tell me what it shows please? 17:44:59 hodapp: idk 17:45:11 but theres still no excuse to waste resources to recompile the same bloody release every night 17:45:20 * agent314 is not going through a bouncer right now 17:50:32 https://it-notes.dragas.net/2026/03/16/why-i-love-freebsd/ oh man 17:52:30 oh there was something similar to this, possibly multi-part article that's also like "why freebsd blablabla" 17:54:10 im paying $3/month for a Raindrop account and also i have a shaanti server that is also quite active, and neither of them has that article lol 17:54:22 *have 18:15:32 * slow realization that you just accidentally flooded the room with your stupid texts and now you look like an asshole * 18:15:53 * agent314 self-imposes 30 minute cool-down period on #freebsd 18:37:03 * ivy wonders why the FreeBSD-lld package exists 18:37:14 should this not just be in -toolchain the rest of the toolchain? 18:38:15 It looks like I may be stuck on FreeBSD 13.4 forever. 14.4 doesn't seem to have support for a Perc 4e/Si (H430) anymore. Unless there is a mernel module I can enable? 18:38:36 Mar 29 14:29:46 saturn kernel: amr0: mem 0xd80f0000-0xd80fffff,0xdfde0000-0xdfdfffff irq 46 at device 14.0 on pci2 18:38:36 Mar 29 14:29:46 saturn kernel: amr0: Using 64-bit DMA 18:38:36 Mar 29 14:29:46 saturn kernel: amr0: delete logical drives supported by controller 18:38:36 Mar 29 14:29:46 saturn kernel: WARNING: Device "amr" is Giant locked and may be deleted before FreeBSD 15.0. 18:38:36 Mar 29 14:29:46 saturn kernel: amr0: Firmware 522A, BIOS H430, 256MB RAM 18:38:36 Mar 29 14:29:46 saturn kernel: amr0: Deprecated code (to be removed in FreeBSD 14): amr(4) driver 18:42:28 tuaris: some LSI cards can be flashed to be mpt(4) compatible using an "IT" firmware (which disables most hardware RAID functionality), if that's an option. but i don't know if that's possible on that specific card, and you'd need to backup and reinstall if you're currently using RAID 18:43:43 Can I cancel a freebsd upgrade. I was able to boot the old kernel. 18:48:06 tuaris: from/to what version 18:48:23 I was upgrading from 13.5 to 14.4 19:04:37 tuaris: what was your upgrade method? What issues are you having? As root, what is the output of "freebsd-version -kru"? 19:06:16 I was attempting to upgrade(freebsd-update upgrade) from 13.5 to 14.4, and was unpleasently surprised that the "amr0: " was removed in 14.0 (by Netflix :( ). I did manage to boot back into the old kernel 19:07:15 freebsd-version -kru 19:07:15 13.5-RELEASE-p10 19:07:15 13.5-RELEASE-p10 19:07:15 13.5-RELEASE-p11 19:07:15 I copied my old kernel back to the default and renamed the newone to 'broken' 19:07:17 Ah, try the MPR driver. That's what I use 19:07:43 I need to load that manuallyy, right? 19:07:59 It looks like you never updated the world. 19:08:05 luser: do you use that on a PERC 4e? i thought that card was a bit old to be supported by mpr 19:08:22 I have an old Intel RAID controller 19:08:45 Most of wwhat I found online says it's the end of the line for this controller. Unless someone adds it back 19:09:06 that's a shame, the system itself is still very capable for the workload I have it doing 19:09:30 man mpr 19:09:38 tuaris: you could replace the card with an old mpt(4) off eBay -- i can't say how long mpt will stay around, but those cards are still pretty popular 19:09:41 Will provide details on what to load. 19:10:13 But I'd need a PCI-X/Express Slot right? (those are currently occupied with NVMe adpaters) 19:11:16 Oh I use "mrsas" 19:11:50 PCI-X is something completely else. 19:11:57 "mrsas - LSI MegaRAID 6Gb/s and 12Gb/s SAS+SATA RAID controller driver" 19:12:31 Yeah, min is a parallele SCISI 19:13:09 That driver support sother PERC systems 19:14:45 I used to use "amr" from memory and had to switch to "mrsas". 19:15:01 That was a while ago though. 19:15:23 tuaris: oh, it's not an add-in card? i didn't know they did on-board percs (i don't know much about dell though, clearly :-) 19:16:10 No, it's an on board one. It's replaceable, but using a proprietary slot. (Dell Power Edge 1850) 19:17:02 It's almost 22 years old, had it new since day 1 and has been a great system. 19:17:16 mpt supports Dell PowerEdge 1750 through 2850 19:18:26 You could try booting from an installer shell and loading the driver(s) to see what is detected 19:19:24 mpt only works for cards with the "IT" firmware, which is why i suggested attempting to reflash it - you may break the card, but if you'd have to get rid of the system anyway... 19:20:30 What's the "IT" firmware? 19:21:32 luser: LSI cards (at least older ones, i'm not sure if they still do this) can support two different firmwares, one for a basic HBA, and one with RAID functionality. some MegaRAID cards can be converted into basic HBAs by flashing them with the IT firmware instead of the MegaRAID firmware 19:21:59 which means they're now supported by mpt(4)/mpr(4)/mps(4) rather than the MegaRAID drivers 19:22:03 I may just need to cancel the upgrade for now. Is copying the old kernel back from /boot/kernel.old to /boot/kernel the olny thing I need to do? 19:23:01 tuaris. Copying the old kernel is fine, but you are out of sync as per "freebsd-version -kru" 19:23:23 kevans: do you know if reviews that aren't the responsibility of any particular person can be assigned to #srcmgr now? i recall some discussion about this but i don't know if it was just a proposal 19:23:32 You can just recompile/buildworld to 13.5-p11 19:23:51 tuaris: if you create a new ZFS boot environment before starting the upgrade, you should be able to just boot the old BE in the loader menu 19:23:55 s/create/created 19:25:00 The boot volume is UFS 19:25:36 tuaris: just buildworld from source at 13.5-p11 19:26:26 That's interesting. why would it be out of sync? I had run a freebsd-update fetch/install (the a reboot) before attempting the upgrade 19:26:51 tuaris: what exactly did you upgrade? freebsd-upgrade upgraded only the kernel first, then rebooted, right? it never upgraded the rest of the system? 19:27:13 tuaris: in that case you should be able to just put a 13.5 kernel in /boot/kernel (but make sure to run freebsd-update again to get current security patches) 19:27:44 correct. After I updated to the latest 13.5 and rebooted. I proceeded with freebsd-udate upgrade -r 14.4. Then ran freebsd-update install when prompted, rebooted, and stopped 19:28:14 tuaris - your systems didn't upgrade to 14, so just finish the 13.5 update 19:28:39 root@saturn:~ # freebsd-update fetch 19:28:39 Looking up update.FreeBSD.org mirrors... 3 mirrors found. 19:28:39 Fetching public key from update2.freebsd.org... done. 19:28:39 Fetching metadata signature for 13.5-RELEASE from update2.freebsd.org... done. 19:28:39 Fetching metadata index... done. 19:28:39 Fetching 2 metadata files... done. 19:28:40 Inspecting system... done. 19:28:40 Preparing to download files... done. 19:28:41 No updates needed to update system to 13.5-RELEASE-p11. 19:28:41 WARNING: FreeBSD 13.5-RELEASE-p10 is approaching its End-of-Life date. 19:28:42 It is strongly recommended that you upgrade to a newer 19:28:42 release within the next 1 month. 19:29:34 Use -F 19:29:45 freebsd-update -F fetch 19:29:54 You have an unfinished update 19:30:17 I see, okay 19:30:56 Still says "No updates needed to update system to 13.5-RELEASE-p11.". Maybe p11 isn't a kernel update? 19:32:20 tuaris: p11 is a patch version. 19:32:22 No, you have an out of sync world as per freebsd-version. 19:32:48 What happens if you run "freebsd-update install" again? 19:33:23 No updates are available to install. 19:33:38 tuaris: And what does "freebsd-version -kru" show? 19:33:42 Did you use -F for that command also? 19:33:52 i'm fairly sure this is expected for 13.5-p11, for 13.x specifically /boot/kernel/kernel wasn't updated 19:34:30 (the tcp and nvmf issues only apply to 14/15) 19:34:31 ivy: That's what' I'm thinking. If "freebsd-version -kru" shows everything in sync, that'd be my first thought. 19:35:26 It's not in sync, but I am thinking there is no new kernel beyond p10: 19:35:26 # freebsd-version -kru 19:35:26 13.5-RELEASE-p10 19:35:26 13.5-RELEASE-p10 19:35:26 13.5-RELEASE-p11 19:35:47 that's the same as my 13.5 test vm which i upgraded the other day 19:35:59 Yes. That's in sync, actually. 19:37:46 It's a shame I have to retire this system. Is there no chance of the hardware support being added back in at all? 19:38:02 It just seems so wasteful 19:38:04 I have never had a system that shows different patch levels after update. Maybe it's different when compiling from source. 19:40:18 tuaris: You could try reflashing the firmware as ivy suggested, or booting from a 14/15 USB installer, and from the shell load mpt/mpr/mrsas drivers and see what is detected without modifying your system 19:40:19 luser: yes, this is specific to freebsd-update 19:40:54 ivy: gotcha. 19:42:01 Okay, let me try booting off a 14.4-bootonly ISO 19:42:14 ivy: I checked man freebsd-version and it states what you said under "implementation notes". 19:42:42 pkgbase will fix this since it always updates the entire kernel package (which i consider a good thing as many people are confused by this) 19:43:02 Yep. Another benefit of pkgbase. 19:44:52 So I'm stuck between a server that's too old for freebsd 14 or risk buying a server that's too new for freebsd 15. :( 19:45:53 tuaris: What about just updating the RAID/HBA? Is that an option? 19:46:46 It's a PowerEdge 1850, the hardware flexiabilty is limited 19:48:49 tuaris: if you don't want to mess with the firmware, it's very likely Dell sold some compatible PERC card that uses the firmware you want - you could look on places like servethehome where people repurpose this old hardware and might know the answer 19:49:56 with Sun hardware of that vintage, you had to pick RAID or non-RAID when purchasing the and non-RAID card was an mpt(4)-compatible card, so Dell may have been similar... it's been long enough i don't remember, even though i think we did have a few PEs back then 19:50:01 Yep. It also looks like the PERC 6e would work as a replacement. Same PCI configuration and everything. 19:50:12 Interesting, hmm 19:52:32 (that said, i don't know how long mpt(4) will be around, but i think it supports hardware a lot newer than amr(4) and there would be quite an outcry if someone tried to remove it) 19:58:26 Removal of mpt would be pretty devastating. I already know of 100's of storage systems that are still using it at my old ${JOB}. 19:58:45 Well, I'm assuming they're still being used. No reason to get rid of them. Just chugging along. 20:04:19 right, i don't think there's any imminent risk of it being removed, but we do seem to be quite aggressive at removing old drivers recently 20:06:36 Oh, I've noticed. :) 20:19:11 I booted into a 14.4 live session and all 4 of those drivers are already loaded mpt/mps/mpr/mrsas. None of them show the Perc4 controller. 20:20:38 Goging to try 15, but I'm not expecting anything 20:33:45 I'm not sure how badly I need to upgrade to 14.4. I can make do with what I have, but how difficult would it be to compile this as a standalone .ko? https://cgit.freebsd.org/src/commit/?id=60de2867c9fc 20:54:16 polarian: yeah, the excuse is "computation is cheap and they pay the same regardless" 20:54:45 polarian: and if you're going to complain in the same breath about how projects spend money, there is *definitely* no excuse to pay for resources and then just have them sit idle 21:04:09 tuaris: Likely, not very difficult at all. The code will still compile (might need a few tweaks in the coming future), but should be rather ready to go. They're just removing the code for support of old hardware reasons. Not because the code doesn't work. 21:06:02 Maybe this is a good first experiance into driver development (I know nothing), but I do know my way around C. I guess I can start by grabbing the files and placing them in a single directory. 21:07:37 Maybe I may throw Claude at it and see. I've had some luck in the past, but I've never attempted it with something like this. 21:08:07 tuaris: Sure. It's easy enough. Grab the code, compile wherever you want (On a FBSD system) and then just take the .ko and load it. 21:47:29 hodapp: well if you colocate you pay for electricity 21:47:31 but why waste money 21:47:39 even if you arent the one paying for it 21:47:49 its like deploying a bloated electron app and saying "well I aint paying to run it" 21:55:00 polarian: and are they colocating? 21:55:09 hodapp: idfk 21:55:12 does it matter? 22:01:55 polarian: if their being supposedly wasteful matters, but if it's only wasteful if they're colocating..... then yes, clearly it does? 22:03:52 hodapp: say you pay £5 for a vps 22:03:58 cool 22:04:16 do you then max out the cpu, memory, and storage as its "wasted" otherwise? 22:04:40 sure you wont pay more, but the company will, which will either affect their profit margins or cost them money, the outcome would be the same, they would increase their fees. 22:05:45 they're more likely to either: not care because they already put effective quotas in place, or just kick you off 22:06:28 or in some cases, just kinda quietly throttle you 22:08:35 a nightly build isn't likely to push things to that point, even on garbage-tier providers 22:09:07 in that regard: the difference between doing a build once a day, versus once a week, is completely negligible