rwp | I don't have pacct installed on my XFCE system. Probably a Recomends pulled it in? | 00:21 |
---|---|---|
rwp | The Refracta live XFCE image is 1.3GB in size. Boot it and see what it does. https://get.refracta.org/files/stable/ | 00:22 |
rwp | Among the very large packages are LibreOffice, and TeXlive, and others for sure. But I remember those as being very large. | 00:23 |
brocashelm | yeah, i already installed refracta | 00:26 |
brocashelm | i removed acct and that deleted /var/log/account | 00:36 |
fluffywolf | for a mini install, don't tasksel anything. start with just the bare base system and only install things as you need them. 60GB is absurd - it should be about a sixtith of that.... | 00:43 |
brocashelm | yeah, i figured out it was due to pacct and obsolete wine folders (removed both) | 00:44 |
brocashelm | that helped bring down the size | 00:44 |
fluffywolf | this box, with tons of crap installed over four versions of devuan, a kernel build, etc, is only 26GB total. | 00:45 |
fluffywolf | most of which is /var, which probably means I should clean things up in there. lol | 00:46 |
fluffywolf | 13GB of it is apt's cache. heh. | 00:46 |
fluffywolf | and 1.3GB of it is wine-staging | 00:47 |
brocashelm | daemon.log and syslog are also bloat-happy files | 00:48 |
fluffywolf | why are you trying to shrink down an existing install instead of doing a fresh install? | 00:49 |
brocashelm | because i applied a lot of my own configs and it would take forever to restore them | 00:51 |
brocashelm | and my original question was the curiosity of getting packages down to 1,000 on an xfce system | 00:51 |
brocashelm | since arch systems that use xfce can pull that off | 00:51 |
fluffywolf | ok, so curiosity. I thought you were trying to move an installed system onto a smaller machine or something... | 00:53 |
rwp | I really meant to "boot" it and look. It's a live boot image. Doesn't need to be installed in order to look to see what it does. | 00:55 |
brocashelm | oh, no, that wasn't my intention at all (moving to smaller machine) | 00:55 |
brocashelm | just to clean up my devuan systems as much as i can, since i like to tidy things up every once in a while | 00:55 |
brocashelm | yeah, refracta is exclusively made with refractatools and basically an out-of-the-box experience. that's the one i settled for | 00:56 |
rwp | I have really gotten to like 'ncdu'. Running "ncdu /var" and "ncdu -x /" could be informative. | 00:57 |
brocashelm | never heard of ncdu | 00:57 |
rwp | It will add 41302 bytes to your system. :-) | 00:58 |
brocashelm | bloated! | 00:58 |
rwp | According to "apt-cache show ncdu | grep Size:". Oh, that's package size. Installed says 92 which I assume is KB. | 00:58 |
brocashelm | pretty nice, though | 00:58 |
brocashelm | might switch to this | 00:59 |
fluffywolf | I've always use plain ol' du. | 00:59 |
rwp | ncdu is pretty nice. I mostly use it now. Previously I would have used "du | xdu" which is a venerable old X program from phil@arl.army.mil notice the army.mil address? | 01:01 |
rwp | (arl is the Army Research Laboratory. And xdu was one of the programs that came out of there.) | 01:02 |
fluffywolf | I ususally use du, du -s, du -sc, or du | sort -n, depending on what I'm trying to do... | 01:03 |
rwp | Those are all good too. And I still do the say, such as when trying to capture a disk event. So might be running du -xml repeatedly in order to capture a usage spike. | 01:04 |
bsd4me | I just found out my uefi systems (including devuan) are on a msdos partition table hdd. Will I need to delete everything and create uefi partition table on hdd? | 02:01 |
Xenguy | rwp, ncdu is great (it could be enhanced, but pretty damn useful) | 02:22 |
bsd4me | I for one didn't know about those :( | 02:28 |
WoC | afaik, you can use mbr partition tables for uefi as well | 02:32 |
WoC | mbr is good for up to 2 TB | 02:33 |
WoC | Actually 2 TiB | 02:33 |
bsd4me | WoC, thanks! | 02:37 |
WoC | np, i have some weird setups... confuses most installers | 02:38 |
bsd4me | This is first time I did this. Well, devuan and redcorelinux seem to be booting ok, so guess I'll keep using them as is. | 02:40 |
WoC | Just a note; PARTUUID is much shorter when using mbr vs gpt | 02:41 |
bsd4me | Xenguy, I downloaded ncdu. Though you meant it for rwp, it's very interesting app! | 02:41 |
bsd4me | WoC, thanks again :) | 02:41 |
WoC | np | 02:42 |
onefang | I tried it to, looks useful. | 02:43 |
WoC | Anyone know if it's possible to mount a FreeBSD ufs filesystem somehow ? | 02:44 |
rwp | Regarding ncdu... I am glad to make more people aware of it! :-) | 02:46 |
WoC | really handy; ncdu | 02:47 |
rwp | Regarding UEFI versus Legacy BIOS boot... That depends upon your motherboard firmware. If it is set to boot Legacy BIOS then you will be booting Legacy BIOS and you don't need UEFI partitions. | 02:47 |
rwp | On the other hand if your motherboard firmware is set to boot UEFI then you already have an ESP (EFI Service Partition) or it wouldn't be already booting. | 02:47 |
rwp | You only need to re-partition to create an ESP if you are switching from Legacy BIOS boot to UEFI boot. Which is independent of Devuan. | 02:48 |
Xenguy | bsd4me, Yeah, ncdu is a keeper | 02:48 |
rwp | If you are moving a Legacy boot system disk from a motherboard that boots Legacy BIOS to another motherboard that only boots UEFI that's when you are screwed. | 02:48 |
rwp | Because then you will need to resize and repartition things to create an ESP that did not exist on that disk. | 02:49 |
rwp | I didn't mention DOS MBR versus GPT because AFAIK that isn't part of the UEFI v. Legacy equation either. It's just that GPT is better so always use it if possible. | 02:50 |
rwp | Oh, on the other hand if you are moving a disk from a system that boots/booted UEFI with an ESP to an older motherboard that only boots Legacy BIOS then... | 02:52 |
rwp | one must remove grub-efi and switch to grub-pc to boot. Probably by booting a rescue disk and doing that switch. | 02:52 |
WoC | It's also possible to do a hybrid with both legacy and uefi | 02:52 |
bsd4me | my ESP is one ssd that came with laptop. I installed the hdd and then installed distros and grub. | 02:52 |
rwp | Yes. That will work too. Using multiple storage devices. | 02:53 |
WoC | even on one device | 02:53 |
WoC | just takes more time when updating the kernel | 02:53 |
bsd4me | Glad to know that. I was pretty upset thinking I was going to have to start all over! | 02:53 |
rwp | bsd4me, You had two independent statements though. a) Your UEFI systems used DOS MBR partitioning. b) Contemplating needing to create UEFI ESP on hard drive. | 02:55 |
rwp | The first did not control the second. But the second may or may not be needed depending upon the other factors. | 02:56 |
rwp | Also UEFI booting firmware often can be switched to boot Legacy mode. Usually. So that is another way out. | 02:57 |
bsd4me | uefi systems are installed on mbr partition table hdd. Actually my question was if I needed to delete systems, change hdd to uefi, and re-install systems (distros) :) | 02:58 |
rwp | AFAIK UEFI is perfectly fine on DOS MBR partition tables. It just needs a UEFI ESP (EFI Service Partition) of about 1MB as a FAT32 to hold the UEFI files. | 02:59 |
bsd4me | interesting, didn't know that about switching to Legacy mode. Thanks for all the info! | 02:59 |
rwp | Regarding switching modes, that all depends upon the firmware. I say "most" allow it. But some do not. And clearly in the future all will be UEFI most likely. | 03:00 |
rwp | I guess I should word that as, in the future it is more likely that firmware will be only UEFI. | 03:01 |
bsd4me | my ESP partition is 100MB. Came that way on laptop. Plenty of room for my needs | 03:02 |
rwp | That should be more than sufficient! :-) | 03:02 |
bsd4me | :D | 03:02 |
bsd4me | I've actually read where some people say it should be 500MB! I discussed that with them, but they kept their opinion. Said they read it somewhere. | 03:03 |
fsmithred | rodsbooks.com | 03:04 |
fsmithred | some motherboards require that | 03:04 |
fsmithred | I haven't run into one yet. | 03:04 |
fsmithred | but I also haven't run into two motherboards that treat uefi the same way | 03:04 |
bsd4me | Interesting! I sure didn't know that. Have to read up on it. | 03:04 |
rwp | I think those are systems where at boot time you have a full graphical OS running with mouse, windows, graphics, and everything, in order to configure the system firmware. | 03:05 |
rwp | That can eat up a lot of space! Talk about bloat... | 03:06 |
rwp | I just looked at the base Devuan installer creates a 512MB ESP /dev/sda1 at install time. I am currently using 148KB of it. | 03:07 |
rwp | On my system with "df -h" I see: "/dev/sda1 511M 148K 511M 1% /boot/efi", and with "parted /dev/sda print" I see " 1 1049kB 538MB 537MB fat32 boot, esp" | 03:09 |
bsd4me | I'm using 60MB of mine, but it has a bunch of junk in it when testing distros that needs to be removed. | 03:09 |
rwp | For me the easiest way to tell on a running system how it booted is "dpkg -l |grep grub" and if I see grub-efi then its UEFI and if grub-pc then it is Legacy. | 03:11 |
rwp | I also said "GPT is better so always use it if possible" which isn't possible on older BIOS systems pre-dating GPT. Older Legacy BIOS won't understand GPT so that forces MBR. | 03:14 |
fsmithred | rwp, you would be fooled by a devuan live-iso | 03:14 |
fsmithred | to see if you booted in uefi, look for /sys/firmware/efi | 03:15 |
rwp | fsmithred, Excellent! I will use that moving forward. That's definitely better. | 03:16 |
fsmithred | the amd64 iso has grub-efi-amd64 installed. If you boot legacy and install the system, the installer will install grub-pc for the bootloader. | 03:17 |
bsd4me | fsmithred, thanks! checked mine and is efi | 03:17 |
fsmithred | the deb package is in the iso | 03:17 |
fsmithred | another way to tell is to run efibootmgr | 03:17 |
* bsd4me is rebooting to check something | 03:22 | |
bsd4me | When I saw my problem, I deleted swap partition and created swapfile. After rebooting, I see that my Devuan swapfile exists. | 03:28 |
rwp | bsd4me, What was your problem with swap? | 03:32 |
bsd4me | rwp, I thought I'd need a partition for my distro testing and since hdd was mbr, I moved swap to a swapfile :) | 03:54 |
bsd4me | On this new laptop, I like testing different distros basically for fun :) | 03:55 |
bsd4me | old laptop was lenovo ideapad. This new one is acer aspire 5 | 03:56 |
rwp | Using swapfiles are viable these days. | 04:26 |
bsd4me | First time I've created one. | 04:31 |
bsd4me | Doubt it will be used. This machine has 16G of ram. I don't do much that requires a lot of memory usage. | 04:32 |
onefang | That's why I use swapspace, it creates and manages swapfiles for you, when you need it. | 04:37 |
bsd4me | onefang, lol, I never heard of that one. Will have to download it and check it out. Thanks! | 04:54 |
rwp | A little bit of swap is always good because it allows the kernel to take memory blocks that are never accessed again and free up that ram for file system buffer cache. | 05:11 |
joerg | onefang: that sounds interesting. I patched the suspend-to-disk scripts on my system to add sufficient _free_ swap just immediately before suspend | 05:12 |
fluffywolf | a little bit of swap is always good if you plan on running firefox with less than 64GB of ram. :P | 05:12 |
rwp | Firefox and Chromium are both definitely pigs. | 05:13 |
joerg | ulimit | 05:13 |
rwp | Real swap is needed for suspend-resume too. | 05:13 |
joerg | honestly, the only way to deal with this memhog-critters | 05:14 |
joerg | rwp: yep, all RAM gets saved to swap then system shuts down | 05:15 |
joerg | so you need more FREE swap than RAM for suspending | 05:15 |
rwp | The problem is that half the memory hog is the program but the other 3/4 of the problem are piggy web sites that must get displayed. (I learned that from Yogi Berra.) | 05:15 |
rkta | I want to kill my gpg-agent when I suspend my laptop. Turns out, I have no idea what actually happens when I close my lid. Where is the point to hook in my command? | 11:55 |
cyteen | @rkta: What actually happens when you close your lid is found in /var/log/pm-suspend.log, add a hook in /usr/lib/pm-utils/sleep.d to add suspend resume behaviour for pm-action events. | 12:39 |
rkta | cyteen: I don't have pm-utils installed | 12:41 |
fsmithred | maybe something in /etc/acpi | 12:43 |
ham5urg | Has anyone used serial tty via bluetooth to login into a machine? | 12:54 |
rrq | rkta: maybe suspend is handled by elogind? (/etc/elogind/logind.conf) | 12:56 |
rrq | if so, "man loginctl" is supposed to be a help | 12:58 |
debdog | rkta: on beowulf it started at /etc/acpi/events/lidbtn and script is /etc/acpi/lid.sh to add stuff. on chimaera I think it was /etc/elogind/logind.conf. | 12:59 |
debdog | logind.conf: HandleLidSwitch=ignore but IDK what else is allowed there and how to hook a script or an action there | 13:04 |
rkta | seems to be elogind. man 1 loginctl points to some 'Hook directories', that could work. | 13:12 |
rkta | this elogind stuff looks a bit like bloat | 13:12 |
debdog | the acpi solution I found and altered within 15 minutes and without internet access. the elogind 'solution' took several days with many conversations here. I hate it | 13:14 |
dokma | I still can't figure out why my initramfs won't assemble my raid array on boot automatically... | 13:39 |
dokma | I've added a script to init-premount to assemble the array manually and now I can boot | 13:39 |
dokma | Is there any way to see what exactly initramfs is doing during boot? | 13:40 |
dokma | This started when I upgraded to Chimaera and my kernel went from from 3 up to 5 | 13:42 |
gnarface | maybe you need to run update-initramfs | 13:50 |
dokma | I ran that a few hundred times while trying to fix | 13:50 |
gnarface | is it a custom kernel? | 13:51 |
dokma | stock | 13:51 |
gnarface | mostly the initramfs just loads modules | 13:51 |
dokma | linux-image-5.10.0-11-686 | 13:51 |
gnarface | there's some config variables in /etc/initramfs-tools/ | 13:51 |
dokma | gnarface: it also has to mount root file system or no boot | 13:51 |
dokma | and for that it needs to assemble the raid array if root is there | 13:52 |
gnarface | yea, it's not gonna find the initramfs though if it can't mount anything | 13:52 |
onefang | Might need RAID support in your bootloader. | 13:52 |
dokma | In grub that should be load the raid module no? | 13:53 |
dokma | I do have insmod mdraid1x in grub.cfg | 13:55 |
gnarface | i think there's more to it than that | 13:56 |
dokma | gnarface: what sould I look for? | 13:56 |
rrq | dokma: as I remember /usr/share/initramfs-tools/scripts/local-block/mdadm has an unfortunate * on line 53, which causes assembly to be abandoned unless there is some /dev/mdX device already | 13:56 |
onefang | /etc/mdadm/mdadm.conf has the proper ARRAY lines? | 13:57 |
dokma | rrq: this? mdadm -q --run /dev/md?* | 13:57 |
dokma | onefang: yes | 13:57 |
rrq | yes | 13:57 |
dokma | rrq: that's the script that should assemble the raid array prior to booting? | 13:58 |
rrq | I think so... maybe an "|| true" is helpful | 13:58 |
dokma | rrq: any way to trace what this thing is doing during boot? like set -x | 13:58 |
dokma | rrq: because the next line would have enabled my array: mdadm -q --assemble --scan --run | 13:59 |
dokma | it either doesn't go into this if or line 53 breaks it somehow... | 14:00 |
dokma | rrq: I've seen some reference to raid arrays not assembling due to some delays | 14:00 |
dokma | why if [ "$COUNT" = "$MAX" ] ??? | 14:00 |
dokma | that's line 50 and I don't get what they were trying to do with that | 14:01 |
rrq | I think it's trying 30 times | 14:01 |
rrq | (I played with raid once some while ago.. no clear memory) | 14:02 |
dokma | I don't have /run/count.mdadm.initrd in the booted system... | 14:02 |
dokma | Would set -x show up in the boot messages ?? | 14:03 |
rrq | it's in the initramfs .. all happens pre-pivot | 14:03 |
dokma | what is pre-pivot ? | 14:04 |
rrq | the boot consists of two phases: pre-pivot and post-pivot | 14:04 |
rrq | pivot is when the disk file system becomes root | 14:04 |
dokma | Is that related to set -x in this script? | 14:04 |
rrq | initramfs is run by an init script that originate as /usr/share/initramfs-tools/init | 14:06 |
rrq | it becoms /init in hte initramfs | 14:06 |
dokma | That makes sense. I still don't understand how is that related to set -x | 14:06 |
dokma | I thought I might be able to get some output from the mdadm script into boot messages to see what's going on | 14:07 |
rrq | the "debug=y" boot parameter might be helpful | 14:10 |
furrymcgee | line 50 seems to be related to bug #830770 mentioned earlier in the script | 14:10 |
dokma | rrq: will check debug=y | 14:15 |
dokma | damn... there is even break=premount which drops you into an initramfs shell | 14:19 |
dokma | but my initramfs shell does not accept keyboard input | 14:19 |
dokma | neither via KVM nor from a local USB keyboard... | 14:19 |
rrq | there is a "break=.." parameter ... try "break=top,premount", it should load keyboard modules | 14:26 |
rrq | (that will stop twice) | 14:28 |
dokma | rrq: I'm just reading this: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=615831 | 14:35 |
dokma | So keyboard is not available in break=top | 14:35 |
dokma | Possibly not even in break=premount | 14:35 |
rrq | are all the modules available in the initramfs? | 14:36 |
dokma | Actually, since keyboard is not available when I get dropped into (initramfs) when /dev/md0 is not found it shouldn't be available if I break at any earlier point | 14:36 |
dokma | I haven't checked that actually | 14:36 |
dokma | Let me unpack it and take a look | 14:36 |
dokma | ehci-hcd, uhci-hcd, ochi-hcd and usbhid are present | 14:39 |
dokma | They probably just aren't loaded | 14:40 |
dokma | is there any way to tell initramfs to load them before anything else? | 14:40 |
rrq | the maybe_break function loads the if break=top ... see scripts/functions | 14:41 |
dokma | tx | 14:41 |
furrymcgee | is there a way to drop initramfs completely? | 14:44 |
rrq | dokma: you might want to use "netconsole=.." instead of any "break" and capture the whole debug output to file on another machine | 14:45 |
rrq | furrymcgee: should be possible though maybe not terribly easy with grub boot. | 14:48 |
dokma | rrq: will check it out | 14:48 |
fling | lxd 4.0.11 is in debian testing | 17:12 |
fling | are we going to have lxd soon? :> | 17:13 |
* fluffywolf wonders what lxd is | 17:14 | |
fluffywolf | oh god, more fucking container nonsense. | 17:41 |
fluffywolf | bbl | 17:42 |
furrymcgee | it manages the overhead introduced by linuxcontainers | 17:46 |
fsmithred | how do you know that lxd is in debian? | 17:49 |
fsmithred | fling, what is the actual package name? | 17:50 |
rwp | I don't think lxd is actually in Debian yet. But it appears closer now. https://bugs.debian.org/768073 | 18:24 |
fling | fsmithred: seen it in #lxc channel | 18:33 |
fling | fsmithred: it is called lxd | 18:33 |
rwp | fling, By design as soon as the package appears in Debian Unstable and Testing then it will automatically appear in Devuan Unstable and Testing. | 18:35 |
fling | thanks | 18:39 |
fling | rwp: which devuan version to try if I want to use lxd? | 18:39 |
rwp | Once it appears then it will appear in Unstable "Ceres" which is always the top of trunk and sometimes/always/never broken. | 18:40 |
buZz | lxc/oldstable 1:3.1.0+really3.0.3-8 amd64 | 18:40 |
buZz | Linux Containers userspace tools | 18:40 |
buZz | keep in mind that lxd isnt really needed for lxc usage :P | 18:41 |
rwp | After ten days if there are no release critical bugs posted against the package then it rolls into Testing. | 18:41 |
rwp | So in order to get lxd in a distro suite before the next release you would need to install Testing. | 18:41 |
rwp | Testing is better than Unstable. But sometimes has problems too. | 18:41 |
rwp | It's always good to have people running Unstable and Testing in order to have them hit bugs and then report those bugs. | 18:43 |
rwp | But as the old saying goes, if you have to ask then maybe it isn't the best suite. When it works, it works great. When it is broken, it needs expert debug and handling. | 18:44 |
fling | rwp: I'm using devuan in lxd so I can have both testing and unstale | 18:50 |
fling | buZz: sure lxc works best in lxd but I've not tried it for years and I'm not going to | 18:50 |
fling | rwp: I just don't know current version numbers and names for testing and unstable :> | 18:51 |
buZz | fling: cool, i dont know anyone that uses lxd for lxc | 18:52 |
buZz | but have at it | 18:52 |
fsmithred | I don't see lxd in sid, looking with apt-cache search or on packages.debian.org | 19:01 |
fling | ok maybe they lied, going to ask | 19:01 |
rwp | I am sure https://bugs.debian.org/768073 will be updated when it is uploaded to Unstable. | 19:10 |
dokma | Any maintainers come here? | 20:53 |
dokma | Devuan maintainers | 20:53 |
rkta | dokma: There are maintainers here, just ask. | 20:54 |
dokma | rkta: just asking if I decide to fix my initramfs issue properly to know to come here and report | 20:57 |
dokma | long live Devuan, death to Poetterware ! | 20:58 |
SymbioticFemale | is the desktop.iso (4gb) a livecd or no? | 21:26 |
fsmithred | SymbioticFemale, not a live-iso | 21:28 |
fsmithred | it has the devuan-installer (aka debian-installer) | 21:29 |
SymbioticFemale | i need to boot to a livecd to install because i need touch screen support during the install process | 21:29 |
fsmithred | it's for offline installs. | 21:29 |
SymbioticFemale | the livecd version won't boot because i don't have network access configured yet | 21:30 |
fsmithred | uh, I don't know if the live-isos have touch screen support. | 21:30 |
SymbioticFemale | mmmm i see | 21:30 |
fsmithred | ...and I'm the one who made those live isos | 21:30 |
fsmithred | xfce4 desktop | 21:30 |
fsmithred | whatever comes with that. | 21:30 |
fsmithred | do you happen to know what is needed for touch? | 21:30 |
SymbioticFemale | well, i can't test whether it has or not because i just get "Certificate Vericication Failed!!! Press any key to return" when attempting to boot from UEFI | 21:31 |
fsmithred | why do you need network for livecd? | 21:31 |
fsmithred | it should boot without network and you can install the running system | 21:31 |
SymbioticFemale | this "certificate verification failed" , i presume that is network related? | 21:31 |
SymbioticFemale | weird | 21:31 |
fsmithred | where did you get that error? | 21:32 |
SymbioticFemale | hmm maybe it is from my bios | 21:32 |
SymbioticFemale | let me play with my bios settings and try this again. be back in a few mins. | 21:32 |
fsmithred | this was during boot? | 21:32 |
SymbioticFemale | i think my bios settings might have got reset. this might be from secureboot | 21:32 |
SymbioticFemale | give me some mins | 21:33 |
fsmithred | ew | 21:33 |
fsmithred | kill it | 21:33 |
SymbioticFemale | ouput gets stuck at "fb0: switching to inteldrmfb from EFI VGA" | 21:54 |
fsmithred | what's your hardware? | 21:57 |
fsmithred | drm doesn't sound good | 21:57 |
SymbioticFemale | xiaomi mipad 2 / x86_64 | 21:57 |
SymbioticFemale | i did expect it to be a stretch to install duvuan on it. was going for debian with non-free firmware but I don't have a USB hub (yet) so i can't connect a keyboard while having my install media attached | 21:59 |
fsmithred | you're not the first one to get that error message: https://github.com/sebanc/brunch/issues/1039 | 22:00 |
fsmithred | how are you planning to do an install without a keyboard? | 22:00 |
SymbioticFemale | well, if i can boot a livecd and install from within said livecd, then i can presumably use the touchscreen | 22:01 |
SymbioticFemale | otherwise i need to setup pxe or other methods that could get me a ssh connection | 22:02 |
fsmithred | oh yeah, touchscreen. | 22:03 |
fsmithred | what does that need? wacom? | 22:03 |
SymbioticFemale | don't even know yet | 22:03 |
fsmithred | damn, my Refracta isos usually have openssh-server running and accessible, but I forgot to enable password authentication in the latest set. | 22:04 |
fsmithred | you could use a beowulf-based iso, boot it, ssh in and do a debootstrap install of devuan chimaera | 22:05 |
fsmithred | (refracta is just a devuan re-spin) | 22:05 |
SymbioticFemale | yeah but i need to configure the wifi too before i can ssh in lol | 22:05 |
fsmithred | oops | 22:05 |
fsmithred | mine does not have non-free firmware installed, it just has the deb packages. So catch-22. no keyboard. | 22:06 |
fsmithred | oh wait... | 22:06 |
SymbioticFemale | i'm going to have to go back to trying to preseed the debian firmware-*.iso i think | 22:06 |
fsmithred | https://get.refracta.org/files/experimental/refracta_test_oblx_5.14_bpo-20211121_1804.iso | 22:07 |
fsmithred | that one has some wireless firmware installed and a backports kernel | 22:07 |
fsmithred | I'm booting that one now to make sure ssh works | 22:11 |
SymbioticFemale | alright, i'm downloading and a few other things concurrently | 22:12 |
fsmithred | don't bother with the backports iso. Can't use password for ssh. | 22:12 |
SymbioticFemale | don't worry about my issue here. i'll resolve it through other means and worst case i'll just wait a few days for a USB hub | 22:13 |
fsmithred | I hate being beaten by the machine. | 22:15 |
SymbioticFemale | lol | 22:15 |
fsmithred | I'm now in the process of updating the backports iso. | 22:20 |
fsmithred | newer kernel, allow ssh access, newer policykit-1 | 22:20 |
fsmithred | and whatever else | 22:20 |
fsmithred | at the rate we're going, I'll probably get it all ready and it'll turn out that you have one of the broadcom chips that require internet access to get the wireless driver. | 22:21 |
SymbioticFemale | yeah i don't know that it is going to go that far to be honest | 22:22 |
SymbioticFemale | currently i'm taking it apart because it got stuck at the manufacturer logo on restart | 22:22 |
SymbioticFemale | lol | 22:22 |
SymbioticFemale | gotta disconnect the battery now | 22:22 |
SymbioticFemale | i really stress that it is not worth your time here | 22:23 |
fsmithred | libts0 for touchscreen | 22:26 |
fsmithred | that's ok. I now know that for only 300kb I can make people's touchscreens work. | 22:28 |
SymbioticFemale | ubuntu seems to be installing with its "safe graphics" option. it'll do for testing until i get a hub or figure out how I'm going to install something more permanent. | 23:09 |
SymbioticFemale | the touch screen in fact works in their installer and they even came up with an on-screen keyboard which works for some things (but not passwords).... however copy+paste worked to fill in the password slot | 23:10 |
SymbioticFemale | /5/ | 23:20 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!