bleb | hey pals | 05:10 |
---|---|---|
bleb | is there a beowulf netinst iso with proprietary intel wifi firmware? | 05:10 |
xrogaan | oh boy | 05:28 |
golinux | bleb: Firmware should be available if needed. | 05:45 |
Centurion_Dan | o/ | 06:41 |
Centurion_Dan | preparing for lockdown in NZ in 2 days.... | 06:42 |
Centurion_Dan | We have what is essentially martial law. | 06:42 |
xrogaan | Good. | 06:45 |
xrogaan | btw, if you want more info about that disease, please join ##covid-19 | 06:46 |
xrogaan | Ideally, people should have prepared over the span of several weeks, buying a bit more than necessary. | 07:02 |
xrogaan | gamemode doesn't have a init.d script, only systemd stuff. | 08:18 |
xrogaan | Should work without systemd in the more recent version, as they fell back to dbus stuff. | 08:19 |
xrogaan | ref: https://github.com/FeralInteractive/gamemode/releases/tag/1.4 | 08:20 |
xrogaan | should be able to work with elogind, but it needs some change. | 08:40 |
bleb | 2020-03-23 00:45 <golinux> bleb: Firmware should be available if needed. | 15:08 |
bleb | golinux: it could not be found on the netinst media | 15:08 |
bleb | i ended up installing with a wired connection | 15:09 |
bleb | but now i need to install the firmware somehow | 15:09 |
debdog | just in case someone has a quick answer, it's not really important (to me): regarding su vs. su -. during the update from ascii to beowulf there was a message "If you want to restore behaviour more similar to the previous one you can add 'ALWAYS_SET_PATH yes' in /etc/login.defs." | 15:12 |
debdog | but if I add that then I get a message on login "configuration error - unknown item 'ALWAYS_SET_PATH' (notify administrator) | 15:12 |
fsmithred | debdog, the message is harmless... | 15:13 |
debdog | oh | 15:13 |
fsmithred | and instead of putting it in login.defs... | 15:13 |
fsmithred | you can put it in /etc/default/su | 15:13 |
fsmithred | (create it) | 15:13 |
fsmithred | or you can put 'export PATH=blah' in /root/.bashrc | 15:14 |
fsmithred | If you leave it as is and use 'su' then you have to use full path to commands in any sbin | 15:15 |
bleb | so in /etc/apt/sources.list i did s,deb http://deb.devuan.org/merged beowulf main,& non-free, | 15:16 |
bleb | and the same for deb-src (not sure if this is needed) | 15:16 |
bleb | then i install firmware-iwlwifi | 15:17 |
fsmithred | only if you want to download source packages | 15:17 |
bleb | restarted | 15:17 |
bleb | and wicd still says "no wireless networks found" | 15:17 |
fsmithred | bleb, were you talking about b43 the other day? | 15:17 |
bleb | fsmithred: no | 15:17 |
fsmithred | oh, right - you jst said iwlwifi | 15:18 |
fsmithred | lsmod | grep iwlwifi | 15:18 |
fsmithred | to make sure the module is loaded | 15:18 |
bleb | yup it is | 15:18 |
fsmithred | is it very new hardware? | 15:18 |
bleb | it's a thinkpad x201 | 15:18 |
fsmithred | no physical on/off switch for the wireless? | 15:19 |
fsmithred | use rfkill to check it or turn it on/off | 15:20 |
bleb | there is a switch, i turned it on and off and theres no change | 15:21 |
bleb | what would i do with rfkill? | 15:21 |
fsmithred | it'll give you the status of wireless and bluetooth | 15:21 |
fsmithred | and it lets you turn them on or off | 15:21 |
fsmithred | is your wireless led on or off? | 15:22 |
bleb | hm it says command not found even though i have the man page | 15:22 |
bleb | fsmithred: the led is off | 15:22 |
debdog | thanks fsmithred. going with your first solution now. but I've made a note on your second one, too, in case I find that more useful in the future | 15:22 |
fsmithred | you might need to be root | 15:22 |
bleb | i am | 15:22 |
bleb | apt says rfkill is installed | 15:23 |
fsmithred | did you su or 'su -'? | 15:23 |
bleb | su | 15:23 |
fsmithred | then you'll need to use the full path to the command | 15:23 |
bleb | what | 15:23 |
debdog | *WAT | 15:23 |
fsmithred | see my current conversation with debdog about changes in su | 15:23 |
fsmithred | it's /usr/sbin/rfkill | 15:24 |
bleb | ok | 15:25 |
bleb | it says 0 wlan phy0 unblocked unblocked | 15:26 |
fsmithred | that means it should be working | 15:26 |
fsmithred | do you have another system you could boot to see if it works? | 15:27 |
fsmithred | coincidentally, the wireless on my T420 just stopped working about a week ago | 15:27 |
bleb | i could put the hard drive in another computer i guess | 15:27 |
fsmithred | that won't test the hardware | 15:28 |
bleb | oh you mean boot another os | 15:28 |
fsmithred | yeah | 15:28 |
fsmithred | maybe live-usb? | 15:29 |
fsmithred | rfkill gives me the same result on my dead wifi | 15:29 |
bleb | it doesnt work on alpine either | 15:31 |
fsmithred | when was the last time you know it worked? | 15:32 |
bleb | it worked on void about a week ago | 15:33 |
GyrosGeier | there are a few laptops with a weird bug that makes them interpret the lid switch as an rfkill | 15:33 |
mason | fsmithred: Did we determine that our T420 wireless is different? | 15:33 |
fsmithred | mason, I don't recall comparing | 15:34 |
GyrosGeier | but that would show up in rfkill command output | 15:34 |
mason | Booting, will check. | 15:34 |
fsmithred | Centrino Advanced-N 6205 (Taylor Peak) (rev 04) | 15:35 |
fsmithred | I don't know how to test it further | 15:36 |
mason | Hm, I don't see the rev 04 here, I see: 03:00.0 Network controller: Intel Corporation Centrino Advanced-N 6205 [Taylor Peak] (rev 34) | 15:36 |
mason | from lspci | 15:36 |
mason | And from dmesg: [ 18.521179] iwlwifi 0000:03:00.0: Detected Intel(R) Centrino(R) Advanced-N 6205 AGN, REV=0xB0 | 15:36 |
bleb | not booting without the ethernet cable, it is stalling at "Configuring networkg interfaces...ifup: waiting for lock on /run/network/ifstate.eth0" | 15:36 |
bleb | oh nvm it's continuing now | 15:37 |
mason | bleb: That'll eventually time out. If you want it to work, don't have an "auto" clause for it. | 15:37 |
fsmithred | mason, yeah rev 34 not 04 | 15:37 |
fsmithred | newegg.com has replacements for $5-10 | 15:38 |
mason | bleb: Also, FWIW, I'm most recently using Network-Manager to make selecting random wifi networks more trivial. Not that I see new wifi networks lately. :P | 15:38 |
GyrosGeier | I dimly remember that "allow-hotplug" instead of "auto" would allow configuring in the background | 15:38 |
mason | fsmithred: You have the firmware package, etc? | 15:38 |
fsmithred | yes | 15:38 |
fsmithred | it was working fine a few weeks ago | 15:38 |
mason | kk | 15:38 |
fsmithred | I'm usually on a wire, but occasionally go to wireless | 15:38 |
mason | Maybe the hardware is kaput... This is a relatively fresh install of Devuan Beowulf. | 15:39 |
mason | fsmithred: You didn't accidentally hit the fairly-well-hidden hardware wifi disable switch, did you? | 15:39 |
mason | fsmithred: It's a subtle thing and not impossible, and hard to debug if you did without realizing it. | 15:39 |
bleb | i'm going to see if i can boot the void linux hard drive that worked a couple days ago | 15:40 |
fsmithred | um, no I don't have the firmware installed. WTF??? | 15:40 |
mason | fsmithred: Check between your optical bay and eSATA port. | 15:40 |
fsmithred | yeah, I found that switch. | 15:40 |
mason | and the firmware: firmware-iwlwifi | 15:40 |
fsmithred | I don't know where that went. | 15:40 |
mason | fsmithred: Easy to miss and if it's toggled, it's a nightmare to debug. | 15:40 |
fsmithred | looks like I purged that package on 2/29 | 15:41 |
bleb | ah fuck i forgot i overwrote this hard drive | 15:43 |
bleb | well | 15:43 |
bleb | the wifi led is blinking on netbsd | 15:44 |
bleb | so it must be a software problem causing it not to blink on devuan | 15:44 |
mason | bleb: Blinking is generally an indication that it's having trouble. | 15:44 |
mason | bleb: It ought to be solid if it's working. | 15:45 |
bleb | the wifi led on my x61 is blinking right now as i type this over ssh | 15:45 |
mason | odd, guess it's different from the T420 - same thing, using it right now, no blinking | 15:46 |
bleb | point is, the wifi led did not go on at all when i was booted to beowulf | 15:46 |
mason | bleb: Sounds like either missing firmware, rfkill, or similar. | 15:46 |
bleb | mason: i installed firmware-iwlwifi | 15:46 |
bleb | modprobe iwlwifi indicated that it was loaded | 15:47 |
fsmithred | ok, try unloading and reloading the module. That just got mine to scan. | 15:47 |
mason | kk | 15:47 |
bleb | rfkill indicated that it was unblocked | 15:47 |
mason | bleb: Is wpa-supplicant running? | 15:47 |
bleb | on netbsd? | 15:47 |
mason | sorry, wpa_supplicant | 15:47 |
mason | Oh, thought were were still talking Devuan. | 15:47 |
mason | But... Doesn't NetBSD use wpa_supplicant as well? | 15:47 |
bleb | im swapping out the hard drive now | 15:47 |
bleb | so i will boot beowulf again momentarily | 15:48 |
bleb | but i thought devuan used wicd | 15:48 |
mason | Maybe with some arbitrary desktop environment, and I'd still expect that to drive wpa_supplicant. | 15:48 |
mason | Nice thing about sticking with wpa_supplicant: your configs are largely the same regardless of OS. I use the same stuff between Devuan and FreeBSD here. | 15:49 |
fsmithred | I take it back - mine is still down | 15:49 |
fsmithred | I take it back again - mine is working. Wicd was set for a usb wireless dongle. | 15:51 |
mason | Ah, no wicd here. | 15:51 |
mason | ifupdown or NM only | 15:51 |
fsmithred | you wouldn't like doing that with the usb interface | 15:53 |
fsmithred | wlx947557885157157 | 15:53 |
mmlj4 | halp! I wanna install wireguard, but the docs now say to enable backports if your debian is older than bullseye (testing)... the old instructions apparently evaporated. So is wireguard in a devuan repo somewhere? | 15:53 |
fsmithred | beowulf-backports and ceres have wireguard | 15:54 |
fsmithred | beowulf=buster | 15:54 |
mason | fsmithred: tab complete works. :P | 15:55 |
fsmithred | oh, that's good. Can't do that in wicd. | 15:55 |
fsmithred | (but I can copy/paste) | 15:55 |
bleb | ok booted back into beowulf | 15:57 |
bleb | no wpa_supplicant, but wicd is running | 15:57 |
bleb | fsmithred: you got your wireless working on your t420? | 15:58 |
mmlj4 | fsmithred: so I'd run beowulf-backports with ascii? | 15:58 |
bleb | ok wireless is working | 16:01 |
bleb | i had to enter "wlan0" under "wireless interface" in the preferences for wicd network manager | 16:02 |
fsmithred | bleb, yeah, I got wireless working | 16:12 |
fsmithred | cool | 16:12 |
fsmithred | mmlj4, you would need to be running beowulf to use beowulf-backports | 16:12 |
bleb | i wonder if any of the beowulf isos have firmware built and and would configure that automatically | 16:12 |
bleb | <bleb> not booting without the ethernet cable, it is stalling at "Configuring networkg | 16:12 |
bleb | interfaces...ifup: waiting for lock on /run/network/ifstate.eth0" | 16:12 |
bleb | mason: how would i do that? | 16:13 |
bleb | <bleb> oh nvm it's continuing now | 16:13 |
bleb | <mason> bleb: That'll eventually time out. If you want it to work, don't have an "auto" clause for it. | 16:13 |
fsmithred | bleb, the live isos have firmware pre-installed, and the installer isos will install it if you need it. To avoid installing it, you need to use Expert install | 16:13 |
fsmithred | exception is some of the broadcom chips that require you to have a network connection to get the software that allows you to have a network connection | 16:14 |
bleb | fsmithred: i did a basic install and firmware was not pre-installed | 16:14 |
bleb | or rather, the netinst did not install it | 16:15 |
fsmithred | that's odd | 16:15 |
fsmithred | oh | 16:15 |
fsmithred | is there a /firmware directory on the iso? | 16:15 |
bleb | it asked if i should try to lead it from the installation media, i tried to, and it couldnt find it | 16:15 |
bleb | s/lead/load/ | 16:15 |
fsmithred | that was with latest netinstall iso? (beowulf beta) | 16:15 |
bleb | devuan_beowulf_3.0.0_beta_amd64_netinstall.iso | 16:16 |
mason | bleb: Sorry, meeting, I'll be able to read/reply after. | 16:18 |
fsmithred | bleb, that question is about loading the files from some other external media | 16:19 |
fsmithred | which you shouldn't need to do except for those few broadcom chips | 16:19 |
bleb | so it shouldn't be asking me about it if i have intel wifi | 16:20 |
fsmithred | it asks everyone | 16:20 |
bleb | oh ok | 16:20 |
bleb | so maybe i could've installed using wifi after all | 16:21 |
bleb | lets see | 16:22 |
bleb | so i get the screen which says "Some of your hardware needs non-free firmware files to operate." | 16:22 |
bleb | ... "The missing firmware files are: iwlwifi-6000-4.ucode" | 16:22 |
bleb | "If you have such media available now, insert it, and continue." | 16:23 |
bleb | "Load missing firmware from removable media?" | 16:23 |
fsmithred | yeah, I always say no to that | 16:23 |
bleb | Isn't iwlwifi about intel wifi not broadcom? | 16:23 |
fsmithred | yes | 16:24 |
fsmithred | any firmware can be loaded at that point | 16:24 |
bleb | so it doesn't actually need it? | 16:24 |
bleb | yet it thinks it needs it? | 16:24 |
fsmithred | you don't need to add it then | 16:24 |
fsmithred | it doesn't really think, it just acts | 16:24 |
fsmithred | and it is recalcitrant to changes | 16:24 |
bleb | so you're saying if i select "No", i should still be able to do a netinst over wifi? | 16:25 |
fsmithred | I'm gonna take a short break | 16:25 |
fsmithred | no, you need to add the firmware before you can get wireless connection | 16:25 |
bleb | right that's what i thought | 16:25 |
fsmithred | or use one of the live isos | 16:25 |
fsmithred | brb | 16:26 |
bleb | <fsmithred> bleb, the live isos have firmware pre-installed, and the installer isos will install it | 16:26 |
bleb | if you need it. To avoid installing it, you need to use Expert install | 16:26 |
bleb | fsmithred: ok so the netinst iso doesn't have firmware built in | 16:27 |
bleb | but it _should_ install it automatically? | 16:27 |
bleb | like i said i went through the basic install and it wasn't installed | 16:27 |
bleb | trying to figure out if this is a bug i've found | 16:28 |
fsmithred | pretty sure it is | 16:30 |
fsmithred | we talked about this at last dev meeting, but I don't remember exactly what was said | 16:30 |
fsmithred | netinstall iso is probably going to get some significant changes | 16:31 |
fsmithred | you did regular or expert install? | 16:32 |
fsmithred | bleb ^^^ | 16:35 |
fsmithred | nm | 16:36 |
fsmithred | the firmware is on the iso. I don't know why the installer can't find it. | 16:36 |
bleb | ok | 16:40 |
bleb | well it seems to be working now | 16:40 |
bleb | just wondering, should i have something about wlan0 in /etc/network/interfaces? | 16:40 |
enyc | hrrm, are there particular things devuan beowulf like tested? | 16:45 |
enyc | I have ascii upgraded to beowulf since ages system | 16:46 |
enyc | will be back at from tomorrow | 16:46 |
enyc | Iiirc i sltill have the udev race conditions at startup with devices / encrypted disk | 16:46 |
enyc | which then leads to problems with automounting | 16:47 |
enyc | but doesn't fail on every boot | 16:47 |
enyc | I'm sure there were some other main issues, but I forget! =) | 16:47 |
enyc | ofcoz a 'reinstalled beuwolf' may be differeent | 16:47 |
fsmithred | bleb, don't configure in /etc/network/interfaces if you're using wicd or other network manager | 16:49 |
fsmithred | enyc, people are still complaining about intermittent boot delays | 16:49 |
fsmithred | with or without encryption | 16:50 |
bleb | ok | 16:51 |
bleb | i'm guessing if the installer were aware that i use wifi, it wouldn't have put that there? | 16:52 |
bleb | because that's what's causing a delay when booting, if the ethernet cable is not plugged in | 16:52 |
mason | Alright, now I'm back. Reading scrollback. | 16:53 |
fsmithred | bleb, if your talking about the installed system hanging while it waits for eth0... | 16:53 |
fsmithred | change 'allow-hotplug eth0' to 'auto eth0' in /etc/network/interfaces | 16:54 |
fsmithred | or even 'manual eth0' if you want to bring it up with ifup | 16:54 |
fsmithred | wait... | 16:54 |
fsmithred | not sure if that's right | 16:54 |
fsmithred | iface eth0 inet manual | 16:55 |
fsmithred | don't need 'auto' or 'allow-hotplug' with that | 16:55 |
bleb | in the event that wifi doesnt work, i would like ethernet to work by just plugging in a cable | 17:00 |
bleb | for that should i use auto? | 17:00 |
mason | bleb: So, if you're using wicd, I don't know. NM has a notion of managed interfaces, and can ignore interfaces in /etc/network/interfaces easily. Not sure how wicd does it. For interfaces controlled by ifupdown, which is to say, /etc/network/interfaces, you just don't say "auto" and it won't come up. No need for allow-hotplug, which I tend to see as equivalent to auto. | 17:00 |
mason | bleb: With NM, you can script it to prefer wired when it's available, disabling wifi in that case. There's no good logical switching of this sort with bare ifupdown. | 17:01 |
mason | And again, no clue about wicd. | 17:01 |
mason | The one thing I'd recommend considering is driving your wifi interfaces directly through wpa_supplicant. NM and wicd and ifupdown all smooth it over and provide dynamic wpa_supplicant config, but the actual native wpa_supplicant configuration is pretty straightforward and very portable. | 17:02 |
bleb | that might be good | 17:04 |
bleb | but the default setup seems to work fine | 17:04 |
bleb | well, after uncommenting the stuff in /etc/network/interfaces | 17:05 |
fsmithred | just use wicd | 17:06 |
fsmithred | and you can select the wired connection if you want | 17:07 |
fsmithred | to have it come up when you plug in the cable, you need allow-hotplug | 17:07 |
fsmithred | and now that I think of it, there's a patch for that to eliminate the delay | 17:07 |
bleb | why have the delay to begin with | 17:09 |
fsmithred | I don't recall the reason it happens | 17:13 |
bleb | so su doesn't give you a root path, but su - moves you to /root | 17:14 |
bleb | is there any command which acts the way su used to | 17:14 |
bleb | i.e. give you a proper root path but leave you in the same directory | 17:15 |
fsmithred | https://dev1galaxy.org/viewtopic.php?pid=15493#p15493 | 17:15 |
fsmithred | there's the patch and somewhat of an explanation | 17:15 |
fsmithred | for su, add to /etc/default/su the following: | 17:15 |
fsmithred | ALWAYS_SET_PATH yes | 17:16 |
bleb | ok thanks | 17:16 |
bleb | why was su changed? | 17:16 |
fsmithred | it got moved from the shadow source package to util-linux | 17:16 |
fsmithred | not that that's an explanation | 17:17 |
fsmithred | I don't know why they changed the behavior | 17:17 |
bleb | so that would affect all linux distros? | 17:17 |
bleb | doesn't seem to affect void fwiw | 17:17 |
fsmithred | not sure if the change was in debian or upstream | 17:18 |
jiefk[p] | Hello Devuan users! I encounter an error with eudev package : | 17:32 |
jiefk[p] | (Can anyone point me to a "pastebin"-like site ? | 17:33 |
jiefk[p] | Setting up eudev (3.2.7-6) ... | 17:35 |
jiefk[p] | dpkg: error processing package eudev (--configure): | 17:35 |
jiefk[p] | installed eudev package post-installation script subprocess returned error exit status 1 | 17:35 |
jiefk[p] | Errors were encountered while processing: | 17:35 |
jiefk[p] | eudev | 17:35 |
jiefk[p] | https://termbin.com/vjgd | 17:37 |
xrogaan | invoke-rc.d: initscript eudev, action restart failed. | 18:04 |
xrogaan | jiefk[p]: check the init script | 18:05 |
xrogaan | ? | 18:05 |
xrogaan | and logs | 18:05 |
xrogaan | invoke-rc.d is used to do actions with /etc/init.d scripts | 18:07 |
xrogaan | like start, stop, reload etc | 18:07 |
xrogaan | so the daemon couldn't be restarted for some reason, which should be in syslog. | 18:07 |
jiefk[p] | xrogaan: Thanks for the tip, Will check the log asap | 18:10 |
jiefk[p] | /var/log/syslog has nothing worst it :( | 18:12 |
jiefk[p] | I managed to do something before reading your messages, not sure about it to be the best thing to do : https://termbin.com/d0tl | 18:13 |
jiefk[p] | but now, apt wants to downgrade eudev and libeudev to a past version : https://termbin.com/1sp2o | 18:15 |
jiefk[p] | I have this : root@devuan:~# service eudev stop ; service eudev start | 18:19 |
jiefk[p] | [ ok ] Stopping hot-plug events dispatcher: udevd. | 18:19 |
jiefk[p] | [....] Starting hot-plug events dispatcher: udevdmkdir: cannot create directory ‘/run/sendsigs.omit.d/udevd’: File exists | 18:19 |
jiefk[p] | I'm still on eudev_3.2.7-6, and now apt won't upgrade to 3.2.9-4 (both versions are on which is on the /var/lib/apt/archives) | 18:23 |
jiefk[p] | sorry for the nonsense above :p | 18:25 |
jiefk[p] | sorry for the noise, eudev for Beowulf IS 3.2.7-6. | 18:28 |
jiefk[p] | https://pkginfo.devuan.org/cgi-bin/d1pkgweb-query?search=eudev&release=beowulf | 18:29 |
bleb | looks like most other distros use the util-linux su, but they also have /sbin in the default path, or link it to something that is | 18:37 |
bleb | so now debian is unique in that you can't run "su" followed by e.g. "rfkill" | 18:38 |
bleb | and the whole reason for moving to util-linux was to be more like other distros in the first place | 18:38 |
bleb | but it had the opposite effect in terms of the user interface | 18:39 |
mason | bleb: There are lots of little ironies like that. | 18:41 |
mason | bleb: Examples, UsrMerge is intended to make Linux more like other Unix and Unixlike systems, at the same time that systemd utterly rejects the idea of portability or cross-platform compatibility. | 18:42 |
bleb | actually it sounds like buster is merged-user by default? | 18:42 |
mason | bleb: #debianfork is a great place to rant about this, so come join us. | 18:42 |
mason | bleb: Don't think it is yet, but it's an option some folks test. | 18:42 |
mason | https://wiki.debian.org/UsrMerge | 18:43 |
bleb | the devs in OFTC/#debian-devel say it is | 18:43 |
mason | bleb: If you have that package installed, sure. | 18:43 |
xrogaan | jiefk[p]: I never got eudev 3.2.9 | 18:45 |
xrogaan | jiefk[p]: https://bugs.devuan.org/cgi/bugreport.cgi?bug=407 | 18:50 |
bleb | closing my x201 lid starts xscreensaver | 18:56 |
bleb | can i make it go to sleep instead? | 18:57 |
xrogaan | jiefk[p]: anyhow, good luck, i'm going to sleep. You might want to ping some devuan people dev team if it's a critical issue of yours. | 18:57 |
fsmithred | bleb, check power manager settings | 19:01 |
fsmithred | you might need lightdm and light-locker for that | 19:02 |
bleb | yeah it was set to lock the screen on lid close | 19:04 |
bleb | i changed it to suspend | 19:04 |
fsmithred | check to make sure it really works | 19:04 |
bleb | it doesn't | 19:05 |
bleb | it turns of the display but doesn't actually suspend | 19:05 |
fsmithred | ok | 19:05 |
fsmithred | there have been discussions about this on the forum | 19:05 |
fsmithred | I have not been able to get it to work, but I also didn't try for long. | 19:06 |
mason | bleb: Oh, weird, I've had that work via the traditional acpi tools. Does suspend work for you? pm-suspend? | 19:06 |
mason | I know it's not helpful to say "works for me" but it at least indicates that it's possible. | 19:07 |
mason | In fact... Someone came into this channel not long ago and I sorted them. Checking scrollback. | 19:07 |
mason | bleb: 14:33 < mason> tse: I bet you didn't edit /etc/default/acpi-support, so go in there and uncomment lid support, then service acpid restart | 19:08 |
bleb | hm i don't have that file | 19:08 |
bleb | nor do i have pm-suspend | 19:08 |
mason | half a sec - you want both | 19:09 |
fsmithred | pm-utils | 19:09 |
mason | (I seem not to have apt-file installed - addressing that.) | 19:09 |
fsmithred | and probably acpid | 19:09 |
bleb | ok pm-suspend works | 19:10 |
fsmithred | there's also acpi-support and acpi-support-base | 19:10 |
mason | fsmithred: There are competing packages. Half a sec and I'll have a definitive answer. | 19:10 |
mason | Or... at 92k/sec from deb.devuan.org, maybe a minute or two. | 19:10 |
mason | bleb: install acpi-support and it'll pull in the rest you need, then look at /etc/default/acpi-support and restart acpid | 19:13 |
bleb | no luck | 19:17 |
fsmithred | doesn't work here, either | 19:17 |
fsmithred | just turns off the screen | 19:18 |
mason | Weird. Well. =cough= "Works for me." | 19:18 |
fsmithred | you've tried to ssh in and failed? Or some other test? | 19:18 |
fsmithred | oh yeah, there should be a moon | 19:19 |
mason | bleb: You uncommented LID_SLEEP=true and restarted acpid? | 19:19 |
bleb | mason: yes | 19:20 |
bleb | i'll try restarting | 19:20 |
mason | bleb: And pm-suspend works? That's very odd. | 19:20 |
bleb | yes | 19:20 |
mason | bleb: Are you using some kind of desktop environment? Maybe there's some sort of DE-specific power management suite inhibiting the normal tools. | 19:24 |
bleb | xfce | 19:24 |
mason | Maybe zap xfce4-power-manager if it's installed. Just guessing. | 19:24 |
bleb | it was running | 19:26 |
bleb | but killing it and restarting acpid didnt help | 19:26 |
bleb | oh wait | 19:27 |
bleb | actually it just worked | 19:27 |
mason | \o/ | 19:27 |
mason | Strip off the layers of cruft and gunk and the older tools work great. | 19:28 |
bleb | next thing is, can i get my audio volume buttons working? | 19:29 |
mason | Hrm, XFCE ought to just handle that. | 19:29 |
bleb | if i open pulseaudio volume control, it is stalled at "Establishing connection to PulseAudio. Please wait..." | 19:30 |
bleb | changing the volume in alsamixer works fine | 19:31 |
mason | Oh, PulseAudio. So, that one. | 19:31 |
mason | There's a PulseAudio file you want to edit. Will look at scrollback. | 19:31 |
mason | bleb: /etc/pulse/client.conf.d/00-disable-autospawn.conf | 19:31 |
mason | bleb: But note that you don't need PulseAudio for much, potentially. Firefox will work with bare ALSA as Debian packages it. | 19:32 |
bleb | i would love to eradicate pulseaudio | 19:32 |
bleb | i need zoom to work though | 19:32 |
mason | bleb: I don't have it on my living room media center. Very easy to live without. | 19:32 |
mason | Oh, does Zoom require it? | 19:32 |
bleb | idk | 19:33 |
mason | Try it without - I don't see pulse hard linked in. | 19:33 |
mason | I don't have Zoom on a Devuan system to test. | 19:33 |
bleb | apt purge pulseaudio? | 19:33 |
mason | there's more than that | 19:33 |
mason | dpkg -l | grep pulse | 19:34 |
mason | some bits of it you might have to keep, but the daemon and utilities can go | 19:34 |
mason | XFCE might object. More reason to not rely on DEs. | 19:34 |
bleb | "if you are not using systemd, comment out this line" | 19:34 |
bleb | well well well | 19:34 |
furrywolf | I hughly recommend bleb's suggestion of purging pulse. heh. | 19:35 |
furrywolf | highly | 19:35 |
bleb | commenting out the autospawn=no worked | 19:36 |
bleb | then as long as i have the pulseaudio plugin in my xfce panel, volume keys work | 19:36 |
mason | There you go then. | 19:36 |
bleb | it does feel like if devuan is ripping out networkmanager, we should rip out pulseaudio too | 19:37 |
bleb | but idk if there is any loss of actual functionality if you just use alsa | 19:37 |
furrywolf | you do lose some functionality... for example, the parts where you get random stutters or difficult-to-troubleshoot silence all stop working. | 19:38 |
bleb | lol | 19:38 |
furrywolf | alsa doesn't have these basic features of pulse. :P | 19:39 |
bleb | two more things for any xfce users | 19:39 |
bleb | i'm trying to map caps lock to control | 19:39 |
bleb | in the xfce keyboard settings, there is a "change layout option" field | 19:39 |
bleb | it has an entry "Caps Lock" but wtf does that mean | 19:40 |
mason | bleb: NetworkManager is actually a different group of people, unrelated to systemd/pulseaudio/avahi/etc. FWIW. | 19:41 |
bleb | i see | 19:41 |
bleb | can we patch whisker menu to not have the new search animation | 19:53 |
rowbee | bleb: shouldn't be too hard to apt source xfce4-goodies and then apply the patch | 20:02 |
bleb | is it possible to just install an earlier version | 20:10 |
bleb | because there isn't actually a patch | 20:10 |
bleb | i added 'deb http://deb.devuan.org/merge ascii main' to /etc/apt/sources.list | 20:22 |
bleb | then did apt update | 20:23 |
bleb | then apt install xfce4-whiskermenu-plugin=1.6.2-1 | 20:23 |
bleb | worked :) | 20:23 |
bleb | i hope addding that to sources.list is safe and won't break anything else | 20:23 |
nemo | does anyone here happen to know whether acroread is packaged for debian (or just ubuntu) and which package would be best for manual installation on Beowulf? | 20:23 |
nemo | ('cause it does not appear to be in any repos, even non-free, for devuan) | 20:24 |
nemo | ftp://ftp.adobe.com/pub/adobe/reader/unix/9.x/9.5.5/enu/ I'm gonna guess this .deb is all there is | 20:25 |
nemo | adobe seems to have abandoned linux | 20:25 |
nemo | and regrettably sometimes their PoS is still necessary due to all the garbage they tossed into PDF over the years | 20:25 |
fsmithred | acroread still exists?? | 20:27 |
mason | fsmithred: It's old and unmaintained, 32-bit only, but it lets you interact with some DRM'd PDFs. | 20:27 |
fsmithred | I think I had one of those, once | 20:28 |
mason | 50 Years of X-Men suffers from it. | 20:28 |
nemo | fsmithred: it's regrettably usually a necessity if someone provides you a "form pdf" | 20:40 |
nemo | fsmithred: the ones that require things like filling it out and "signing" it (image considered just as good as a cert usually) and saving it to send back | 20:41 |
fsmithred | yeah, I resort to printer and scanner for those | 20:42 |
nemo | fsmithred: I've done that before yeah, or GIMP | 20:42 |
nemo | depends. sometimes it works well sometimes not | 20:42 |
fsmithred | haven't seen one in a long time | 20:42 |
nemo | fsmithred: this particular one has irritating linked dependent sections | 20:42 |
fsmithred | oy | 20:42 |
nemo | fsmithred: so based on stuff I fill out, pages and parts of pages appear/disappear | 20:42 |
nemo | really gotta see if we can rework this form as a tiny app - we've done that with some of their other HR stuff | 20:43 |
nemo | couple of nice friendly web pages with 90% of it pre-filled from the employee database | 20:43 |
fsmithred | that sounds good | 20:43 |
nemo | fsmithred: wish I could share you the before/after for the one I hooked up last month, but it's all internal :/ | 20:43 |
nemo | fsmithred: but basically pages of PDF and walking from boss to boss replaced with some purty checkboxes the bosses are autonotified of ☺ | 20:44 |
fsmithred | cool | 20:44 |
fsmithred | my bank just redesigned their website, and I wish they'd do it again. Current layout sucks. | 20:44 |
fsmithred | lots of glitz that takes up most of the space when I'm trying to look at a list of transactions | 20:45 |
enyc | Date last logged on : 15:03:53 - Fri, February 28 (Haven time) | 20:45 |
nemo | fsmithred: that's one thing that frustrates me with the bank interface redesigns | 20:55 |
nemo | fsmithred: suddenly I have to whitelist half a dozen servers for cookies when it just used to be one | 20:56 |
nemo | fsmithred: ditto JS | 20:56 |
nemo | fsmithred: suddenly the browser rendering requires far more resources.. | 20:56 |
fsmithred | yeah, js for facebook and youtube are not appropriate (but they fucking exist on the page.) | 20:56 |
nemo | hm | 21:25 |
nemo | adobereader-enu:i386 depends on libgtk2.0-0 (>= 2.4). | 21:25 |
nemo | weird. I thought I had gtk2 on here already | 21:25 |
nemo | I do... | 21:26 |
nemo | 2.24.32-3 | 21:26 |
nemo | oh. too new eh | 21:26 |
nemo | wait no. | 21:26 |
nemo | guess I need to unpack the package to see what it's checking for exactly | 21:26 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!