00:04:48 I currently have some disks at my apartment door. I need to replace a drive in my pool! 00:05:12 magically it resurrected itself but I dont trust it 00:07:09 however, im at work and so I guess the disks are gonna be left in the cold for a while. im a bit worried about them. ill let them warm up a few hours before I attempt anything 00:14:34 why are disks at door 00:15:13 do you have problems with executing tasks like i am? 00:15:16 :p 00:15:59 reminds me i fucked up my first large zpool 00:16:08 i wanted it to have guard space 00:16:13 but forgot it 00:17:11 now i have permanently sized pool until disks age or so 00:17:13 :p 00:17:42 apparently zfs really likes to hug your data 00:17:56 once you write it, there it stays 00:18:21 i never checked, where does data go in new device? 00:18:40 if it's writtend at the start, why does it need same size 00:19:16 if your fs is also a volume manager and knows where the data is 00:21:59 btw i should swap the disks around i guess 00:22:13 wanted to age them differently 00:22:57 and each one i tested too 00:23:08 which took a week for 12t funnily 00:24:41 sadly i also managed to bust my first hdd and now i need to give it to recovery 00:25:11 no idea why that didn't have automatic backups 00:26:11 why i began with 1 disk and ended up not with a mirror but with just 0 disks and 10 yeare old backup 00:26:19 and 50% hdd copied 00:26:21 :p 00:29:57 oh meh i didn't see it 00:40:19 damn thats a lot of text bro 00:40:35 ketas: they are at my door because I had them delivered. 00:41:05 nonop disks at door on this planet are fine 00:41:54 no idea what you are saying. lost in translation 00:42:03 what 00:42:23 what!? 00:42:30 WHAT?!?? 00:42:49 :P 00:49:11 i mean hdd that's factory sealed into esd bag will not get hurt if it's kept outside at most places if not anywhere on this planet... provided that it later equalizes back into approximately t-shirt temperature, accompanied with reasonable humidity, which is non-condensing 00:49:57 I've never heard t-shirt temperature before. I like it. 00:54:01 >25c <40c 00:55:05 I disagree. >3c <20c :-) 00:55:44 for hdd? 00:55:46 or you? 00:55:48 :p 00:56:14 the HDDs are not factory sealed, they are refurbished :( 00:56:21 I am not rich lok 00:56:23 lol 00:56:27 For me, T-shirt wearing temp. Although I'd like to see disk-sized T-shirts. 00:57:59 repaired hdd's eh 00:58:03 A little freebsd t-shirt for hard drives, we could get rich off that idea! 00:58:07 never tried 00:58:46 used hdd's, tho... 00:59:08 its what I can afford 00:59:30 brand new hdd has failed for me, while hdd's half of my age work 01:00:07 funnily it's one of most reliable mechanical device you can find 01:01:23 I got a set of 4 refurbished DC HC520 12TB drives, two for a backup system im building, and the other two are spares for my main pool 01:02:04 it's proubably we have making them for 70 years already 01:02:12 because 01:02:18 been 01:02:20 eh 01:02:25 HDDs are kinda engineering miracles indeed 01:02:58 and every one is expected to run 24/7 until it's obsoletr 01:03:00 e 01:03:02 pretty much 01:03:47 ok 24/7'ness is bit problematic maybe 01:04:16 mine run 24/7! 01:04:43 but i don't think it's physically harder to make hdd for 24/7 than 8/7 01:04:47 hmm 01:06:49 can't be arsed to look for bearing and arm lifetimes 01:07:00 bearings do magnets now i see 01:09:46 the fact that hdd works at all is fun 01:14:10 https://www.reddit.com/r/DataHoarder/comments/18sinux/what_is_the_physical_length_of_a_gigabyte_on_a/ 01:14:14 eh 01:39:55 wavelength of visible light is too large to see gap under a head or regions in which make up your bits, yet you are able to read and write data there perfectly, despite surface under you moves at highway speeds and you are moving fast, and to be able to actually find the right place and time for your data, you first need to read servo markers that are on same surface 01:40:15 i think it was easier 70y ago 01:41:13 Were there even disks (as opposed to drums) ca 1955? 01:42:55 yes 01:44:34 and even they were state of art already 01:49:12 flash is even funnier 01:54:44 so you peek out of car window at 110km/h and always spot a small speck 5000 times smaller than human hair on the road 02:00:09 NYOOOM 03:40:28 so, I've been working on this for a while. I'm running FreeBSD as a guest. Everything works flawlessly except my HP laptop 2 finger schrolls fine except when scrolling up.If I'm using a browser the scrolling up of my fingers loads random websites from my history. I've tried disabling 2 finger scroll in /etc/sysctl.conf /usr/local/etc/X11/xorg.conf.d and installed imwheel. 04:15:15 hrm he left ;/ 04:31:31 maybe he comes back 04:32:08 sometimes setting the mouse to a usb tablet gives better experieince 04:32:19 seller has not posted t460 yet 04:32:21 he said guest so I expect VM 04:32:26 i think it sucks 04:32:30 but anyway 04:32:34 i think ketas sucks 04:32:46 one sata ssd only 04:32:52 touchscreen tho 04:32:55 meh 04:33:00 whatever 04:33:58 usb tablet? 04:34:00 you only need one 04:34:24 well i had idea if mirror storage 04:34:26 of 04:34:34 make a z2 04:34:54 why? 04:34:59 cause why not 04:35:09 i mean could do copies=3 04:35:11 buy all the ssd's you can beforer AI takes them all 04:35:17 and a 128GB ssd is 500 04:35:37 i wonder what the ssd lifetime is 04:35:44 never tried one yet 04:35:47 varies i guess 04:36:18 damn first ssd's were hell of a ride i read 04:36:50 power goes, and data is gone 04:37:20 same with ketas 04:37:37 so yeah 07:58:57 good morning! I have some problem with urxvt and copy/paste between terminal and other applicatioons.. I have this in my .XResources https://dpaste.com/563UVY2KD but when mark some text and hit ctrl+shift+c /ctrl+shift+v it doesnt work. And also ctrl+Shift seems to be some builtin stuff I got a "yellow" popup with ISO-mode insert /and KEYCAP Picture insert mode when releasing shift button.. 08:11:53 ctrl-shift on urxvt should enable iso 14755 input (unicode stuff). you might need to disable it for crtl+shift+[key] shortcuts. 08:14:35 i never needed clipboard for terminal copy/paste. middle-click/shift+insert was enough for me. there might be some problems with the perl-extension enabling the clipboard 08:17:00 psycorama: shift+insert works when in terminal but that doesnt get paste/copy from browser /other application 08:19:20 nwe, I never use the clipboard with X and Unix windows. I only ever use the primary selection. Shift+Insert pastes the primary selection not the clipboard. 08:21:37 In the beginning X has always used the primary selection for highlight-copy with the left mouse button 1 drag. That is pasted with the mouse middle button 2. Mouse right button 3 extends the highlight and copy. 08:22:26 Then along came IBM+Microsoft and they use Control-X and Control-C for copy and Control-V for paste. But Control-C by default sends an interrupt signal to the foreground process group. So it is terrible for terminals. 08:23:00 I pretty much ignore the clipboard. But inside web browsers it is still active in the location bar and in text input areas. But I still mostly ignore it. 08:23:36 The xclip utility is available in ports and can be used to access either the primary selection or the clipboard or to transfer from one to the other. 08:24:18 rwp: I would lovly just use shift+insert if it worked for me from terminal -> X browser/application) :P 08:24:19 "xclip -selection clipboard | xclip" for example will print from the clipboard into the primary selection. And other similar transfers. Useful for when the web browser only supports the clipboard but I need it in the primary selection. 08:24:58 As far as I know the web browsers do not support Shift+Insert for anything. 08:25:20 But they do support paste from the mouse using the middle button the same as other programs. 08:25:35 Copy with the left button. Paste with the middle button. 08:27:09 Which means one could highlight-copy with the left. Then "xclip -o | xclip -selection clipboard" and then go to the web browser and Control-V to paste from the clipboard. 08:27:22 ah now when I using shift+ push in the middle of touchpad it´s working 08:27:57 It shouldn't need the shift. Just the middle button should paste. 08:28:46 rwp: yupp you had right =) 08:28:49 Touchpads can be a little tricky. 08:29:00 maybe then I dont need the clipboar stuff :D 08:29:06 I never use it. 08:29:41 After you have copy with the left going then try double clicking with the left, which will highlight words at a time. That makes it easier to copy text because it snaps across words. 08:29:54 And triple click will highlight whole lines. 08:30:38 And then after getting a feel for that then highlight something and then use the right button to extend the region, or shrink it, but mostly extending it. 08:31:17 I think I must be more careful when copy from touchpad :P it easy to right-click instead if middle click :D 08:31:50 In the terminal Shift+Insert should work okay for pasting text. But the web browser ignores it. :-( 08:32:40 rwp: thanks for the help :) 08:33:16 Happy to help! Good luck! 08:57:04 has someone here got pritunl-client up and running, or should I run this in a virtual-machine and use some bgp routing so my "main-host" can reach vpn-networks via vpn-machine? 10:42:20 Hi 10:42:44 Finally got back my sim to work 10:44:07 But as I can see the attacks is still going.. Cloudflare captcha blocked 56,000 request, but passed other 87,000 in just less than 24 hour 10:54:40 What site is this Retrofan 11:05:44 puffi: look like you weren't here yesterday 11:12:04 can someone recomend a compact usb blueetooth-dongle that works on freebsd, ? 12:00:26 unsure 12:01:01 i got 3 randoms recently, all were detected but never worked 12:01:09 i was suprised :p 12:14:32 ketas: :P 12:19:57 same with some wifis 12:20:09 i mean i have to debug 12:20:27 maybe needs hack, maybe needs fw 12:20:56 and then i have some modeswitching usb eth 12:21:19 that test only costed few euros tho 12:21:54 the working usb bt some asshole stole 12:22:49 i also have working wifis and eth luckily :p 12:24:18 bts actually follow some standard 12:24:25 when they do 13:12:44 https://forums.freebsd.org/threads/freebsd-bluetooth-virgin.88689/ 13:12:49 look a virgin 13:14:44 and then there's https://forums.freebsd.org/threads/bluetooth-audio-how-to-connect-and-use-bluetooth-headphones-on-freebsd.82671/ 13:14:58 i could try those things 13:49:39 Now I have that song stuck in my head 15:00:43 can't wait til christmas break. i am so gonna install freebsd on my thinkpad e440 this time. 15:01:36 last time i installed ghostbsd, so i could more quickly confirm that freebd & its derivatives would support the wifi card, sound etc. and it did! 16:03:43 Hi 16:04:14 They sent me this "Thank you for reaching out in this ticket as well. I was able to confirm the restrictions have been removed. No additional action is required on your part regarding this matter. If you have additional questions don't hesitate in reaching out to us. Otherwise this ticket will close in 96 hours." 16:05:29 I think this is enough to know what is happening here "No additional action is required on your part regarding this matter." 16:06:41 The same day of hack, they said their server running my service is completely broken.. so they moved my service to other one 16:10:27 Also when I reached them and I said indirectly they are may be hacked, they replied "The security of a customers server is solely the responsibility of the customer, we send these terms of service tickets to any customers who's server is in violation. We do not run checks on other customer servers as we do not have access to customer server to do so. We monitor patterns and receive reports... 16:10:28 ...from outside sources who identify a Linode server conducting malicious activities." and So on.. and India data centers is still out of service for 90% hosting providers 16:12:41 Be careful.. some of providers now mabye partly got owned (Linode is one of them).. as I said before those attacks is from September AI bots, and look like that they succeeded 16:21:38 Anyone tried the MySQL Shell? I think our DBA are going to ask for that. https://dev.mysql.com/doc/mysql-shell/8.0/en/mysql-shell-install.html 16:22:11 dvl: It's a mixed bag. 16:22:13 Retrofan, this isn't always that busy of a channel, so it's not that big of a deal right this minute, but in general, you should not paste more than one or two lines in IRC - use a paste site and share the link. 16:22:34 Sorry 16:22:47 dvl, I*'ve used it on an occasional basis. . . it got me by. 16:23:15 I am just want people to be careful, my phone number and private info in my account is leaked :( 16:23:16 CrtxReavr: installed how? 16:23:17 Was that a paste? 16:23:23 I think of it as more of a diagnostic tool, or something to work out query syntax, to plug into a python script, or whatever. 16:23:36 thumbs: how did you install? 16:23:56 mason: Yeah from the support ticket 16:24:12 Ah, kk. Pastepins are good. I like bpaste.net for the purpose. 16:24:20 but all of them refuse to say that theirs a hack on their side 16:24:37 It's a big attack 16:25:04 Might also be good for #freebsd-social. 16:25:40 mason: if you checked your server logs, specially ssh and http.. you will find tons of trash 16:25:58 from bots 16:26:03 Retrofan: That's always the case. Except for ssh, as people can't reach it. 16:26:27 but now India got owned 16:26:53 mason: I know but the last months they went crazy 16:26:55 #freebsd-social 16:26:59 dvl, prolly on linux, installed from a package. 16:27:22 CrtxReavr: I've now had a request to port it. :/ 16:28:21 dvl, you're prolly the best man for the job. >=] 16:28:39 CrtxReavr: There are much better porters than me. 16:29:13 what is a good porter? 16:32:20 paulf, porters are (still) mostly whos not whats. 16:34:43 dvl, maybe their tarball has a good Makefile and it will JustWork(tm)? 16:35:06 CrtxReavr: ... 16:35:18 dvl: ports, IIRC, but it's been a long time. 16:39:23 what is the definition of a good porter if you must? 16:39:51 thumbs: I ask because I don't see it in there now. 16:55:10 Hm. I'm trying to buildkernel for armv6, release/14.3.0-p5 ; is that still a supported target for 14.3? 16:55:27 I'm asking because buildkernel is failing with a bunch of errors ... OTOH, I see armv6-related files still in tree. 16:56:21 (linker errors, mostly `hid_*`-related, e.g. `│ld: error: undefined symbol: hid_get_device_info 16:56:22 ` 16:56:26 ) 17:12:15 Does anyone know how big is the git src repo ? I'd like to get an idea before I attempt to clone it. Thank you. 17:16:37 codin: I have no idea, but it should be huge. Cloning it with `--depth 1` can work for you? in combination with `-b` to select just an specific branch. That way it will not download all the history 17:18:16 Okay, thank you. I'll see how far I'll get. 17:19:04 oh, this says the size is about 3GB https://api.github.com/repos/freebsd/freebsd-src 17:21:47 depth 1 is a good idea in this scenario if you dont need to traverse commit history at all 17:26:11 I think there may be a regression in the 14 release for the 'em' driver, I have an Intel 82583V NIC. I'm using a little box with 4 ports and stress test the system between 2 ports at full speed ~1Gbps. The throughput is less stable starging with the 14.2 and the following releases. In 14.1 the throughput is ~stable at ~940Mbps for both RX and TX. 17:27:12 Aside from trying to bisect the code bewteen these releases, I don't know how to go about debugging this further. 17:28:53 `netstat -w1 -Iem1` shows a ~constant stream of input errors which lead to lower throughput. 17:34:56 codin, The FreeBSD git source repository is about 3.3 GB in size. 17:37:05 I believe I only need branch `stable/14` since the observed regression should be in this narrow range of commits. 17:40:46 You can probably reduce the size with a shallow clone. That's the size of the full depth git clone here. 19:30:39 How can I resume a kernel build ? I ran into an issue with a prior build, I fixed the error and would like continue the build process without deleting already compiled sources. 19:40:17 it will not work this way, unless you are building WITH_META_MODE=YES 19:40:22 https://wiki.freebsd.org/MetaMode 19:40:35 codin: so you have to delete or make clean 19:41:22 but no worries, building kernel is not much time consuming if you are not buidlign toolchain 19:42:11 Okay, thank you, that's advanced. Clean, rinse and repeat :( 19:42:16 codin: why are you bulding kernel ? 19:42:57 why not :p 19:43:01 perhaps you don't need it 19:43:11 I'm trying to track an observed regression with the em driver for one of my NICs. 19:43:43 but building kernel just to learn hot to build kernel is completely OK 19:44:42 im curious about codin problem. what chipset is your nic 19:44:44 codin: how does this regression manifest itself? 19:46:21 I have an Intel 82583V NIC. Under network load, the throughput drops and I get a ~constant stream of input errors. 19:46:45 is it the only one you got? have you ruled out the hardware being borked? 19:47:06 Yes, I ruled out hw problems. 19:47:39 The system is stable with previous FreeBSD versions, eg, 14.1.0, but regresses in 14.2.0. 19:48:47 Actually 14.1.0 sees errors at times too, but it manages to recover, whereas 14.2.0 does not. 19:49:41 might be worth sending an email in the mailing list if you cant track it down 19:49:44 codin: it could be change in TCP stack, not driver itself 19:50:40 specialbomb, Sorry, I'm new to FreeBSD which mailing list are you suggesting to post on ? 19:51:07 codin: IIRC you have submitted a PR 19:51:07 mzar, I'm fairly sure that's not the case for my issue though. 19:52:03 The box is routing packets between 2 interfaces, it does not consume them. 19:53:00 So the TCP stack should not be involved here. 19:54:21 ha.. yes, but routing stack yes 19:55:49 codin: im fairly new to freebsd as well, so im not 100% sure which list you should post to, but it should be safe to ask where exactly to go for the issue in the questions mailing list. 19:55:59 I have seen complains about 82583V on either mailing list or Bugzilla, could you point to this codin ? 19:56:29 codin: have you tried 15.0-BETA ? 19:56:47 do check the archives first 19:57:08 I tried 15.0-BETA, but it has the same issue for me. 19:57:27 I looked at: https://bugs.freebsd.org/bugzilla/buglist.cgi?quicksearch=em(4)%20Component%3Akern for the recent issues with this driver. 19:57:56 I think the issues reported that have affected the same NIC have been resolved. 19:58:13 I have had such NIC 19:58:57 making me wanna check netstat when I get home :( 19:59:29 With 14.1.0 I can iperf through the box at ~940Mbps full-duplex with a steady state. 20:00:12 Right, `netstat -w1 -Iem1` shows input errors ~constantly. 20:00:52 The router/box is stable otherwise, the network flows are not... 20:01:26 I have somethig similiar running on a single server that hasn't been decommissioned -> 82571EB/82571GB Gigabit Ethernet Controller D0/D1 (copper applications) 20:02:09 not it's still able to saturate 1Gb link 20:02:20 under 15.0-STABLE 20:02:38 *it's still able to saturate 1Gb link 20:04:35 RX and TX ? 20:05:16 RX and TX, ie, full-duplex, at the same time ? 20:05:39 codin: so if you want to bisection, then using WITH_META_MODE=YES will speed things up 20:06:01 Okay, I'll look into that. 20:06:16 codin: sure, it's backup server, every night this NIC is receiving data 20:07:08 Your NIC has a slightly different number than mine though. 20:08:07 I run iperf3 --bidir between two laptops connected through the FreeBSD box/router. 20:08:11 yep, it's not the same 20:08:32 codin: what about updating nic firmware? is it possible ? 20:09:37 I have not looked into that, it's an old machine. However, I know and verified that it can sustain the 1Gbps throughput but not with the latest version. 20:09:56 So this looks like a software regression to me. 20:22:26 codin: here's the record https://cgit.freebsd.org/src/log/sys/dev/e1000?h=stable/14 20:23:44 FreeBSD 14.1-RELEASE was released on June 4, 2024, so look at commits past this date 20:24:15 https://cgit.freebsd.org/src/commit/sys/dev/e1000?h=stable/14&id=49f12d5b38f6bbc66a81438b9bc7ed21af3be39c -- I suspect this commit, but have to verify it. 20:24:32 Thank you, I looked at the commit history for the e1000 driver too. 20:26:13 I checked out a version prior to that commit and I'm waiting for the build now. 20:27:02 i hope it works out for you 20:27:58 I think the last netstat digest I saw in my mail, I saw about 100 errors. I bond two em devices together in a lagg 20:35:56 codin: can you post somewhere output from sysctl dev.em.0.iflib ? 20:38:54 https://pastebin.com/5Rg8x9A7 20:39:25 Please note that the system is not under stress now though. 20:42:25 why it has only 2k buffer size ? 20:43:38 I assume that is a default ? as I did not change that... 20:45:16 probably yes, is it a quad device ? 20:45:53 what do you mean by "quad" ? 20:46:13 4-port nic ? 20:46:23 Ah! yes, it is . 20:52:28 codin: have you tried any tuning, for example overrating TX and RX descriptors ? 20:53:45 I did not, I ran the default configs that come with the live images/releases. 22:05:08 the 14.3-p5 crash i was having is /probably/ not freebsd but a cpu that went bad 22:05:22 if anyone was following my udp rust server app OS crash issue 22:20:21 hah, I could believe ut 22:20:24 *it 22:25:43 anyone know why firstboot-freebsd-update isn't working? it's installed during bsd scripted install, and i place /usr/local/etc/rc.conf.d/firstboot_freebsd_update file there 22:26:01 scripted bsd install finishes, system reboots into first boot, nothing 22:26:28 the file has firstboot_freebsd_update_enable="YES" in it 22:26:49 is that even used dir 22:27:05 hmm, if you have some IRC logging I swear someone else was having this issue 22:27:07 why not? i put all the conf files for pkg in there 22:27:25 mmm maybe 22:27:31 what even uses iy 22:27:33 t 22:27:36 i only use the /etc/ one for base stuff 22:27:41 i could try 22:28:40 specialbomb: Yeah, that was kerneldove_ :) 22:28:42 if you wouldn't mind 22:28:45 lol 22:28:52 haha 22:29:32 i guess i'll just skip this pkg and add a note to my first login setup notes. pos pkg 22:29:34 yeah it does use it 22:29:38 but 22:29:42 we must keep a kerneldove daily issue counter hehe 22:29:50 days since... 22:29:54 fuck knows really? 22:30:05 debug a rc 22:30:14 ya i'll just take out the pkg 22:31:50 does it work elsewhere? 22:31:58 first time trying it 22:32:26 if you don't really need file per package, you can use more than one rc.conf 22:33:02 meh i like that style so i'll just not use the pkg 22:33:09 ty tho 22:33:43 hopefully nothing installs /usr/local/etc/rc.conf 22:33:55 but it's a option 22:34:07 nothing reads it, have to say that 22:36:03 it's also all, like, open source, so you could add new features you wish, you could even submit them for review for potentional inclusion : 22:36:07 p 22:36:36 if i get around to testing and figuring out why it isn't working i'll do that 22:39:46 1/1 22:44:06 rc_conf_files="/etc/rc.conf /etc/rc.conf.local /usr/local/etc/rc.conf" 22:44:28 rc_conf_files="$rc_conf_files /usr/local/etc/rc.conf" 22:44:30 :p 22:44:44 is that from the firstboot pkg? 22:46:15 rc_conf_files="$rc_conf_files $_localbase/etc/rc.conf" 22:46:25 what? 22:46:39 no it came from my braib 22:46:42 brain 22:48:00 could put into rc.conf rc.conf.local, even /etc/defaults/vendor.conf 22:48:25 why your one didn't work, no idea 22:48:32 nah if it doesn't use standard rc behaviors i won't use the pkg 22:48:37 correct name i hope? 22:48:58 well i listed all the details, confirm in a vm? 22:49:02 there is no standard eh 22:49:12 ofc there is 22:49:32 ALL of the sys config i do, including with pkgs, works within this framework. and it's documented to work in the man pages 22:49:48 it's not like i just put files in /foo/bar/rc.shit/ and expected it to work 22:49:57 OMG: ">>> Kernel(s) MYKERNEL built in 8891 seconds, ncpu: 4" :-) 22:50:17 that's a big kernel 22:50:28 or it's a 4 core 486 22:50:30 :p 22:51:01 Well... 1. it is a slow/older machine, yes, and 2. I built the GENERIC kernel with everything in it. 22:51:06 or you built 10 of them at once 22:51:42 Also I should have started the build with -j4 maybe... 22:51:57 Unless the parallel build is the default. 22:52:08 unsure what's default 22:52:26 parallel isn't always better funnily 22:52:49 because you need memory for all, and machine also needs to use cpu for fs 22:53:29 tmpfs could do ncpu*2? 22:53:34 still... 22:53:43 Yeah, the build time may no be linear with # of cpus, but I would expect a shorter then 8k seconds build time :) 22:53:49 something needs to run the ram 22:54:06 *than 22:55:16 Anyway, my first FreeBSD kernel build ! time to stree test the network on this box. 22:55:47 why was it built anyway 22:56:31 I want to test for what I suspect to be a regression introduced around 14.2.0 22:56:34 i mean eh 2h28m11s 22:56:50 unsure how much kernel took here 22:57:56 the hell is .0 22:58:21 /remembers actually compiling the kernel on a 486 22:58:27 well you could view it as one 22:58:41 how long was it? 22:58:45 think it was to get a specific ISA NIC working, took almost 24 hours from memory lol 22:58:55 just kernel? 22:58:58 yes 22:59:01 :p 22:59:16 yeah i only ran fbsd on 486 once 22:59:22 diskless netboot 22:59:30 probably only had 4MB RAM and a PIO HDD 22:59:48 everything else was at minimum p1 23:01:11 worked great, was a fun experiment 23:02:20 i still have the 3Com® EtherLink® III i had there 23:02:45 had to take bracket away 23:02:56 and hack the rj45 23:03:09 so it would fit into case 23:03:12 :p 23:03:19 haha, if it works it works 23:03:59 i kept and still have my first real serial (non-winmodem) modem for some reason 23:04:14 iirc had to plug cable in before inserting a card 23:04:22 lol 23:04:43 frame was blocking it 23:05:13 those old steel pc cases were no joke either 23:05:52 designed carry the weight of CRT sitting on them when laid on side 23:07:27 I buy sliger chassis, all of their stuff is plate steel. I was using one as a stool the other day :D 23:07:36 tuff stuff 23:08:47 some place used old something as a stool 23:08:55 vax sun whatever 23:08:58 square box 23:10:04 rj45 hackery reminds me a plan to hack a low profile cat5 for even lower profile plug 23:10:44 remove part of tab, remove plug end and outer shell of cable 23:11:06 end result is something you could use with laptop if you want eth 23:13:19 I disagree with this, fundamentally. do it though 23:13:53