plasma41 | Who is the op for this channel? | 00:45 |
---|---|---|
buZz | why | 00:45 |
plasma41 | The topic could use an update | 00:46 |
buZz | to reference that offtopic channel? | 00:46 |
plasma41 | Among other things | 00:46 |
buZz | parazyd is the topicwriter :D | 00:47 |
buZz | but beowulf isnt out yet | 00:47 |
buZz | is it? | 00:47 |
fsmithred | does it still say 'post-ascii bikeshed'? | 00:48 |
buZz | it does not | 00:48 |
buZz | that i can see | 00:48 |
plasma41 | fsmithred: That's in the #devuan-dev topic still | 00:48 |
buZz | isnt that fine? :) | 00:49 |
plasma41 | parazyd: Is probably asleep right now if I have my timezones straight | 00:49 |
fsmithred | well, technically we are post-ascii release | 00:49 |
fsmithred | and will be for eternity | 00:49 |
plasma41 | s/parazyd: Is/parazyd is/ | 00:49 |
buZz | real nerds never sleep | 00:50 |
buZz | ;) | 00:50 |
onefang | I'm a real nerd, and I should sleep soon. That way I can sleep through the water being turned off for the day. ;-P | 00:50 |
onefang | I agree the topic should change, I made my suggestion a while ago, at least for the first sentence. | 00:51 |
fsmithred | that didn't work as expected | 00:51 |
fsmithred | I think this topic is ok | 00:52 |
fsmithred | it's timeless | 00:52 |
fsmithred | Join us, and celebrate in freedom and justice | 00:52 |
* timeless looks up | 00:52 | |
fsmithred | oh, sorry | 00:52 |
onefang | lol | 00:52 |
fsmithred | you may go back to sleep if you wish | 00:52 |
timeless | thanks, sounds like a good plan | 00:52 |
timeless | no celebrating for me | 00:53 |
fsmithred | sorry to hear it | 00:53 |
fsmithred | you at work? | 00:53 |
timeless | yes :-) | 00:53 |
fsmithred | Get back to sleep! | 00:53 |
timeless | actually i have to go to a (non-work) meeting | 00:53 |
fsmithred | probably should be awake for that | 00:54 |
timeless | since it involves roads and the last time i tried i was almost killed... | 00:54 |
* onefang yawns and heads to bed. | 00:54 | |
timeless | operator was proceeding in the wrong direction at a reasonable clip on a crowded road, head-on at me... | 00:55 |
fsmithred | yikes | 00:55 |
fsmithred | back in a few minutes | 01:10 |
fsmithred | my jessie is dying. :( | 02:07 |
fsmithred | can't get to the desktop. It starts to load, I see background image and panels start to come up, then the monitor tells me that it's shutting down | 02:08 |
fsmithred | I can ssh in | 02:08 |
fsmithred | I can drop to console with crtl-alt-F1 before I log in (using lightdm) but not after I log in. (or after the screen goes black.) | 02:09 |
fsmithred | magic sysrq keys don't work. | 02:09 |
buZz | fsmithred: possibly a gpu issue? | 03:31 |
fsmithred | only happens with one user | 03:35 |
fsmithred | me | 03:36 |
fsmithred | might be xrandr | 03:36 |
fsmithred | I'll disable it and try again | 03:36 |
fsmithred | that worked. | 03:41 |
fsmithred | and running xrandr again to separate my two monitors kills them again | 03:42 |
fsmithred | but it didn't do that when I rebooted into ascii/openbox, and I'm sure I didn't have mirrored monitors. | 03:43 |
buZz | curious\ | 03:43 |
buZz | maybe you configured xrandr in your xorg conf? | 03:45 |
fsmithred | no xorg.conf | 03:47 |
fsmithred | it runs from .config/autostart/lxrandr.desktop in xfce and from .config/openbox/autostart in openbox | 03:48 |
fsmithred | same settings in both systems | 03:48 |
gnarface | maybe a tiny bit of bad video ram? | 03:50 |
fsmithred | maybe | 03:50 |
gnarface | mirrored mode uses less video ram i think | 03:50 |
fsmithred | psu fan is kinda noisy. | 03:50 |
fsmithred | the noise comes and goes in a regular cycle | 03:51 |
fsmithred | this thing has wonky graphics anyway. It does not like a grub background image. It posterizes them. | 03:57 |
fsmithred | I think its code name is skylake | 03:58 |
golinux | Skylake is known to be Linux unfriendly | 03:59 |
gnarface | yea that's another one that needed a newer kernel | 03:59 |
gnarface | you said that was jessie? i didn't even know that kernel would boot it | 03:59 |
golinux | There is an early post on the form that talks about it. All the *lakes have similar issues. | 04:00 |
fsmithred | when it was new, only xubuntu-14 worked | 04:00 |
fsmithred | seems ok with ascii. | 04:01 |
golinux | OK | 04:02 |
fsmithred | I can put the drives back in the other box, which is more appropriate for winter. | 04:02 |
fsmithred | sleepy time. | 04:03 |
fsmithred | probably should sign off, too. | 04:03 |
fsmithred | see y'all tomorrow or so | 04:04 |
chomwitt | hi from greece. | 08:26 |
chomwitt | i am a virtual private server nomad... bytemark , hosteurope now a greek company that wont allow me to install on my own a new os on my cloud server, an on top of all that i d like to run devuan, so are there any server providers suggested? | 08:28 |
gnarface | someone definitely has an answer to that | 08:37 |
golinux | chomwitt: Maybe https://devuanhosting.com/ | 08:39 |
onefang | I use and recommend Yisp.nl. The let me install my own OS, which is Devuan. They where looking at providing official Devuan support. | 08:40 |
chomwitt | thanks. | 08:41 |
chomwitt | i also check https://devuanhosting.com/ | 08:42 |
chomwitt | but the currency may be an issue | 08:42 |
MinceR | seems to be automatically converted upon paying with debit/credit card | 08:43 |
onefang | Last I checked Yisp accepts Euros (what I pay them with), US$, and Bitcoin (I think). | 08:44 |
onefang | And yes, credit / debit cards and PayPal usually automatically convert currencies, for a small fee. | 08:44 |
MinceR | what sort of technology is yisp.nl built on? | 08:47 |
yeti | https://yisp.nl/ <<< high performance? page took >15 sec to load... | 08:48 |
onefang | I've been with them for a long time. Started on a VPS, now I'm on a Supermicro Xeon server. Things might have changed. | 08:48 |
onefang | When I switched, I ended up with twice the RAM, better performance, and cheaper rent. B-) | 08:49 |
MinceR | well, at least with a dedicated server, you know your stuff doesn't have systemd underneath | 08:50 |
onefang | https://sledjhamr.org is my server, one of the Devuan mirrors. | 08:51 |
chomwitt | bitcoin is cool in case of problems in euroland, in the greek crisis i couldnt pay bytemark with a bank transfer!! | 08:54 |
chomwitt | onefang: does Yisp offers a web interface to select from a list of OS images to do a 'format' anytime i want or i have to ask it ? | 09:00 |
onefang | I'm not sure. My last OS "format" I asked for a temporary disk to be installed, and did a debootstrap Devuan ASCII install on it, then swapped over, coz that's what I wanted to do. | 09:01 |
onefang | Ah they have a "(Re)install" button, I'm not gonna click on it to see what it does. I'd guess it'll give you a choice of operating systems to install or re-install. | 09:06 |
onefang | I had written a debootstrap based script that I extensively tested at home before I unleashed it on my server, to install exactly the Devuan setup I wanted. | 09:09 |
xinomilo | if you don't mind asking for format each time, there's a small greek coop that offers vps with debian/devuan. | 09:33 |
chomwitt | xinomilo: interested, since i also value low latency and pingtime on a server for greek visitors | 09:39 |
chomwitt | in my current cloud server i have debian 9. could i transform it to devuan without asking the hostcompany | 09:40 |
chomwitt | ? | 09:40 |
onefang | Debian Stretch to Devuan ASCII upgrades are well supported by Devuan. | 09:41 |
xinomilo | it's on germany based data center, unmanaged. remember this a small coop not big cloud provider... | 09:43 |
xinomilo | if you're interessted message me, i don't like ads | 09:44 |
xinomilo | :) | 09:44 |
avbox | Just for your information, I made a new version of AVMultimedia, the full desktop (around 6 GB uncompressed) starts now in 10 seconds, have a look here: https://youtu.be/sbDpZUgil1w | 11:12 |
avbox | Unfortunatelly obs does not capture the first splash screen, but yes, it is realtime and I get this startng speed to on a normal notebook, on very old ones it is about 30 seconds, but not mor. | 11:13 |
zoobab | I guess everybody here saw SystemD eating /home | 15:40 |
onefang | Then went off to #debianfork to discuss it. I hope. B-) | 15:41 |
zoobab | https://streaming.media.ccc.de/asg2019/relive/164 | 15:41 |
zoobab | https://www.phoronix.com/forums/forum/linux-graphics-x-org-drivers/opengl-vulkan-mesa-gallium3d/1127452-systemd-homed-systemd-now-working-to-improve-home-directory-handling | 15:42 |
onefang | The story has been posted here a couple of times. | 15:42 |
onefang | #debianfork is the better place to discuss it, this is a support channel. | 15:43 |
nemo | gnarface: apt-get -t ascii-backports install linux-image-4.19 ? | 15:56 |
nemo | onefang: used to be a general gossip channel so understandable people get confused | 15:57 |
nemo | onefang: and... occasionally used to discuss distro planning, so how things like debian or systemd could impact its maintenance is relevant, but I guess clearly we are mostly looking to gossip | 15:57 |
onefang | Devuan planning stuff has it's own channels now. Which also got that story posted there before. | 15:59 |
nemo | in any case, even if this channel still was into gossip he's late to the story 😝 | 16:00 |
nemo | so | 16:00 |
nemo | having trouble selecting a backports linux kernel | 16:00 |
nemo | anyone know how one does that in debian? | 16:00 |
nemo | was reading https://wiki.debian.org/HowToUpgradeKernel | 16:00 |
nemo | so I tried the apt-get line above | 16:01 |
nemo | resulted in: | 16:01 |
nemo | linux-image-4.19.0-0.bpo.5-amd64 : Conflicts: linux-image-4.19.0-0.bpo.5-amd64-unsigned but 4.19.37-5+deb10u2~bpo9+1 is to be installed | 16:01 |
nemo | ah. looks like: | 16:02 |
nemo | apt-get -t ascii-backports install linux-image-amd64 | 16:02 |
nemo | is the correct name | 16:03 |
nemo | but... | 16:03 |
nemo | linux-image-amd64 : Depends: linux-image-4.19.0-0.bpo.6-amd64 but it is not installable | 16:03 |
nemo | bleh | 16:03 |
nemo | hm. wonder if the devuan backports are simply incomplete | 16:05 |
xinomilo | apt install linux-image-amd64/ascii-backports ? | 16:10 |
xinomilo | sometimes "-t" flag isn't working well | 16:10 |
xinomilo | should be the same | 16:10 |
nemo | oh | 16:15 |
nemo | no change | 16:16 |
nemo | hum | 16:16 |
nemo | welp. lemme see if linux-image-4.19.0-0.bpo.6-amd64 exists in the devuan package search | 16:16 |
nemo | https://pkginfo.devuan.org/stage/ascii/ascii-backports/linux-image-4.19.0-0.bpo.6-amd64-unsigned_4.19.67-2~bpo9+1.html | 16:17 |
nemo | is that "unsigned" a problem? | 16:17 |
xinomilo | not a problem that's just the name https://packages.debian.org/stretch-backports/linux-image-4.19.0-0.bpo.6-amd64-unsigned | 16:22 |
nemo | ok... | 16:22 |
nemo | https://pkginfo.devuan.org/stage/ascii/ascii-backports/linux-image-4.19.0-0.bpo.5-amd64_4.19.37-5+deb10u2~bpo9+1.html | 16:22 |
nemo | does not have unsigned | 16:22 |
nemo | and, presumably it does indicate something about the package 'cause synaptic notes it as (unsigned) in the packagel ist | 16:22 |
* nemo shrugs. anyway. can't seem to install it, apt-get install -f isn't indicating any breakage.. | 16:23 | |
nemo | I guess I could just try downloading the .debs and installing 'em that way | 16:24 |
xinomilo | suppose you run `apt update` ? | 16:24 |
nemo | yes | 16:24 |
nemo | several times | 16:24 |
xinomilo | hmm | 16:24 |
nemo | wish it would give me a bit more detail as to *why* linux-image-4.19.0-0.bpo.6-amd64 is not installable | 16:25 |
nemo | oh. hm. | 16:25 |
nemo | E: Unable to correct problems, you have held broken packages. | 16:25 |
nemo | I fail at reading 😃 | 16:25 |
nemo | but... but... I held a package? whaaat? maybe back when I manually installed to get this thing working | 16:25 |
nemo | welp. let's see if I can find out what packages have hold flags | 16:25 |
nemo | apt-mark showhold returns nothing | 16:26 |
nemo | nothing in dpk -l either | 16:27 |
xinomilo | ok, watching it here as well | 16:27 |
xinomilo | apt install linux-image-4.19.0-0.bpo.6-amd64-unsigned | 16:28 |
xinomilo | skip linux metapackage | 16:28 |
xinomilo | probably trouble there | 16:28 |
nemo | translation? | 16:28 |
xinomilo | linux-image-amd64 is a metapackage depending on the actual kernel deb | 16:29 |
xinomilo | which is currently in ascii-backports : linux-image-4.19.0-0.bpo.6-amd64-unsigned | 16:30 |
nemo | ok.. | 16:31 |
nemo | xinomilo: that part is implied by what apt is doing, I was wondering what "probably trouble there" meant - did you find somethin? | 16:34 |
xinomilo | this is the correct package : https://pkginfo.devuan.org/stage/ascii/ascii-backports/linux-image-4.19.0-0.bpo.6-amd64-unsigned_4.19.67-2~bpo9+1.html | 16:35 |
nemo | oh. so the metapackage is pointing to a bad name string | 16:35 |
nemo | so... how do I install it cleanly. extract every package listed in the meta and install them one by one? | 16:36 |
xinomilo | meta tries to install kernel deb without "-unsigned" in the end. and it fails | 16:39 |
xinomilo | Depends: | 16:39 |
xinomilo | linux-image-4.19.0-0.bpo.6-amd64 | 16:39 |
xinomilo | same in debian, probably bug there | 16:40 |
xinomilo | https://packages.debian.org/stretch-backports/linux-image-amd64 | 16:40 |
nemo | why is bpo.5 signed and bpo.6 unsigned? | 16:40 |
nemo | always a bit worrying to me to see a key package have a word like "unsigned" in it ☺ | 16:41 |
nemo | even if I have no idea what it means in practice for the kernel | 16:41 |
xinomilo | probably because of this : https://release.debian.org/transitions/html/auto-linux-signed-amd64.html | 16:41 |
xinomilo | don't really know | 16:42 |
xinomilo | there it is : https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=941042 | 16:43 |
nemo | xinomilo: oh. good. I guess I just have bad timing | 17:23 |
nemo | xinomilo: been waiting for this kernel for a year and I pick the day they break the backport ) | 17:23 |
xinomilo | you can still install the unsigned package without trouble | 17:25 |
nemo | xinomilo: just have to do it all manually | 17:25 |
nemo | xinomilo: future upgrades will proceed normally? | 17:25 |
nemo | there's no other packages in the meta? | 17:25 |
xinomilo | no future upgrades need to be done by hand, untill meta is fixed | 17:27 |
james1138 | General question about Devuan in its beginnings. Not a complaint - just wondering. Why the decision to make LXQT as one of the options for a desktop on Devuan instead of LXDE?? | 17:33 |
FatPhil | any ideas why my raspi refuses to enact the changes that ntpd tells it to make? | 18:16 |
FatPhil | It keeps saying stuff like "adjusting local clock by -149.744902s" in syslog, as it just drifts further and further from reality. | 18:16 |
FatPhil | some time in the past a restart of openntpd forced it to fix itself, but even that didn't work today. | 18:17 |
r3boot | is this xntpd or openntpd? And is bursting enabled? | 18:18 |
FatPhil | openntpd, and I don't know what bursting is. | 18:18 |
FatPhil | absolutely stock .conf | 18:19 |
r3boot | if your time is N seconds off from ntp time, the ntp daemon will step towards the wanted time in small increments. xntpd knows burst/iburst, which allows you to set the time right w/o needing to step | 18:19 |
r3boot | Sounds like openntpd does bursting @ startup tho... Dont have a lot of experience with it tho, avid xntpd user here | 18:20 |
FatPhil | I don't particularly care how it gets to the correct time, whether gradually or in one step, as long as it actually does it. the daemon wakes up often enough for less than a jiffy's new drift. | 18:21 |
FatPhil | yeah, with no battery/rtc on the pi, it has to burst at boot time. I'm curious why it didn't do anything on a daemon restart. | 18:21 |
r3boot | maybe bc the clock is drifting harder then ntpd can correct for | 18:22 |
r3boot | have seen that on quite a lot of esoteric hardware myself | 18:22 |
r3boot | See if you can adjust the step size | 18:22 |
r3boot | and/or the polling intervals | 18:23 |
golinux | There is a post on the forum about this: https://dev1galaxy.org/viewtopic.php?id=3076 | 18:24 |
FatPhil | It's only drifted .4 s during the day today, so that's not much to keep up with. | 18:27 |
r3boot | -149.744902s is a lot more then .4s ;) | 18:28 |
FatPhil | Yes, because it last actually changed its time about 50 days ago. | 18:29 |
FatPhil | It's now logging "adjusting local clock by -150.127659s" | 18:29 |
FatPhil | it's almost certainly due to cranky h/w on the pi | 18:30 |
r3boot | yeah, drift issues are usually caused by crappy hardware | 18:30 |
r3boot | and ^^ is reason enough to *NOT* use a rpi as a time server :) | 18:30 |
FatPhil | I notice xntpd isn't an available package for me presently (still on jessie), so I'm a bit stuck with openntpd. | 18:34 |
FatPhil | Ouch! "When ntpd receives a SIGUSR1 signal, it will write ... to syslog" didn't seem to work - it killed the daemon! | 18:41 |
FatPhil | trying rdate... | 18:48 |
_abc_ | Reading news: Poettering wants to modify / update the home dir... | 18:48 |
FatPhil | rdate -n worked fine | 18:51 |
golinux | _abc_: Old news. Please take it to #debianfork | 18:55 |
FatPhil | I don't want to criticise the little thing, it normally behaves fine, but this ntpd thing is beginning to be a drag. I really ought to dist-upgrade, and see if things magically fix themselves. | 18:56 |
_abc_ | Does it impact devuan in any way? | 18:56 |
_abc_ | ^ golinux | 18:56 |
golinux | When it does we'll let everyone know. | 18:57 |
_abc_ | Is there a step by step upgrade guide from ascii to err what's the next one called? I see the handy table with versions vs debian versions has moved from the front page of devuan.org? | 19:27 |
_abc_ | This is the one to follow, right? https://devuan.org/os/documentation/dev1fanboy/ | 19:29 |
_abc_ | What was the next step called ? Beowulf? No | 19:30 |
yeti | beowulf. | 19:30 |
_abc_ | I see only jessie and ascii in mirrors. Where is Beo? | 19:31 |
yeti | images arent out yet... maybe except in some pseudo secret places... | 19:31 |
_abc_ | I find ascii has some really ancient packages in it, it's upstream's fault, but I really need the SSL enabled irssi for irc @freenode | 19:32 |
yeti | http://deb.devuan.org/devuan/dists/ | 19:32 |
_abc_ | thanks | 19:32 |
_abc_ | Is that linked from the top index of devuan.org? | 19:32 |
yeti | no idea... | 19:33 |
_abc_ | Is it okay to mention one is on devuan on #debian and ask for info / advice on packages there? | 19:34 |
_abc_ | I.e. what is the common practice for this? | 19:34 |
Jjp137 | btw, there seems to be a more recent version of irssi in ascii-backports if you want to go that route first | 19:35 |
_abc_ | "more recent" than what. I have ascii-backports in my sources.list and it does not want to go higher than it is now | 19:35 |
_abc_ | irssi 1.0.7-1~deb9u1 (20180213 0139) | 19:36 |
Jjp137 | ascii-backports has 1.2.0-2~bpo9+1 | 19:36 |
_abc_ | Strange. Let me try to pull it in. Brb | 19:36 |
Jjp137 | ascii-backports never has priority over ascii unless the current installed version is from ascii-backports itself (or something else broke) | 19:37 |
_abc_ | It does not even show it. Just a sec. | 19:37 |
yeti | apt-cache policy irssi | 19:37 |
Jjp137 | (or more technically, it has a pin priority of 100) | 19:37 |
_abc_ | aptitude shows me only 1.0.7-1~ as above | 19:38 |
_abc_ | So what is the way to make it upgrade to irssi-1.2.2 ? | 19:39 |
_abc_ | I updated the packages list and backports was in the list of updated items. | 19:39 |
yeti | aptitude should list the different versions | 19:39 |
yeti | loot at the details if irssi | 19:39 |
_abc_ | It does not. Let's see if aptitude is at the highest version 1st | 19:40 |
Jjp137 | yea go to the bottom and it should have the list of available versions there | 19:40 |
golinux | _abc_: pkginfo.devuan.org | 19:40 |
_abc_ | golinux: ? | 19:40 |
golinux | To find if we have a package | 19:41 |
_abc_ | Ah now it shows it as 'p" | 19:41 |
_abc_ | Let's see if I can upgrade it | 19:41 |
_abc_ | yay yes. Brb | 19:41 |
_abc_ | Re. Ok I'm on 1.2.2 | 19:44 |
_abc_ | Why did it work when I selected upgrade in the versions in the package info and not when I tried + on the package name in the main list? | 19:44 |
_abc_ | Will it not upgrade over major (?) version numbers in such cases? | 19:44 |
Jjp137 | b/c the ascii version had higher priority than ascii-backports | 19:45 |
_abc_ | I did not know this. I need to check a number of packages I tried to upgrade with '+' in the main list. | 19:45 |
_abc_ | Jjp137: is this not a little counter productive? Also, I hope security has higher priority than the base distribution? | 19:45 |
Jjp137 | well you're meant to pick and choose from backports and not install everything from it | 19:46 |
Jjp137 | oh and apt-cache policy by itself shows that, at least for me, ascii-security and ascii both have a priority of 500, but in that case then the most recent version applies | 19:50 |
_abc_ | Ok, your advice helped me upgrade a bunch of things from backports. | 20:30 |
_abc_ | Thanks. | 20:30 |
_abc_ | Had to do it manually one by one but I think I covered the important things. | 20:30 |
_abc_ | For now. | 20:33 |
buZz | _abc_: or just distupgrade to beowulf? :D that doesnt have backports yet | 20:36 |
buZz | i think .. | 20:36 |
_abc_ | Some day | 20:39 |
buZz | i upgraded my main workstation some while ago, its nice to suddenly have a whole pile of up2date(r) software | 20:44 |
buZz | it had ~4000 packages installed, so it was quite an upgrade | 20:45 |
adlpaf | #django | 20:47 |
gnarface | nemo: yea there's some known issue and you gotta use the unsigned one for now. it's working for you though, right? | 20:50 |
gnarface | james1138: if it is different from debian's options, it probably had something to do with availability of non-systemd patches and volunteer help leading up to launch | 20:50 |
gnarface | FatPhil: uh, the rpi needs some patched ntpd that allows it to do that immediate jump at boot time. not sure how you don't have it. | 20:51 |
gnarface | FatPhil: (but you could use ntpdate to set it once yourself then start ntpd and watch it to see if it behaves right. maybe you had just lost the wifi connection to the ntp server or something like that? | 20:52 |
buZz | gnarface handling backlog like a pro | 20:52 |
gnarface | ) | 20:52 |
buZz | :) | 20:52 |
yeti | and the forwardlogs? | 20:52 |
yeti | aren't logs always "back"? (unless your ntp goes postal...) | 20:53 |
buZz | :) | 20:53 |
gnarface | unless your irc client can see the future... | 20:53 |
buZz | backlog is a term describing open questions from previous | 20:54 |
gnarface | FatPhil: -g Normally, ntpd exits with a message to the system log if the offset exceeds the panic threshold, which is 1000 s by default. This option allows the time to be set to any value without restriction; however, this can happen | 20:54 |
gnarface | only once. If the threshold is exceeded after that, ntpd will exit with a message to the system log. This option can be used with the -q and -x options. | 20:54 |
gnarface | FatPhil: (setting that is supposed to do it) | 20:54 |
gnarface | FatPhil: (i don't know if that patch ever made it from raspbian to upstream ntpd but it did make it into the devuan ascii image for rpi1) | 20:55 |
gnarface | FatPhil: i have never had a problem with it but i am also not using wireless | 20:56 |
gnarface | yeti: when i try to handle the forwardlogs people get all gripped out and start yelling weird stuff at me like "get out of my brain! shut up! you're not God!" | 20:58 |
yeti | wouldn't openntpd's -s be enough to fix it? | 20:58 |
yeti | set time from ntp at boot | 20:59 |
gnarface | i don't even know if that is openntpd or the bsd one... | 20:59 |
gnarface | on the rpi image | 20:59 |
gnarface | i just remember raspbian had to do something weird to it to address this | 20:59 |
yeti | "the foundaion" sometimes does weitd things | 20:59 |
gnarface | FatPhil: uh, yea so that's a breakthrough. openntpd *does* show up in the repos but that is *not* the one i'm using | 21:00 |
yeti | a while ago they disabled fsck because it would confuse the user if bootimg sometimes would take longer | 21:00 |
gnarface | FatPhil: i'm just using the one called plain old "ntp" | 21:00 |
gnarface | and i'm using a local time server elsewhere on my LAN to reduce outside influence on the drift and jitter rates | 21:01 |
gnarface | yeti: ugh. | 21:01 |
buZz | yeti: lol wtf | 21:04 |
buZz | keep in mind, raspbian is -not- a raspberry foundation creation ;) | 21:04 |
buZz | they only appropriated it to themselves later :P | 21:04 |
yeti | "a while" was belittled... some years ago would be better wording | 21:04 |
buZz | you can run plain debian or devuan on a pi -easily- | 21:04 |
buZz | and always could have, almost | 21:05 |
buZz | -only- the pi1 had such ancient cpu that it wasnt supported | 21:05 |
yeti | I had 2 PIs committing suicide under debian few weeks ago | 21:05 |
yeti | unbooatble after update | 21:05 |
buZz | by having /var on SD? :P | 21:05 |
yeti | one pi1 and one oi3 | 21:05 |
buZz | ah, cute | 21:05 |
yeti | I'm waiting for newer deboian images before I retry | 21:06 |
yeti | I wantred to understand debian's kenrel updates to patch that into my devuan pis | 21:06 |
yeti | but lost drive when they fell apart on the 1st update | 21:07 |
yeti | my cubies run fine... started as debian and then transgraded to devuan | 21:07 |
yeti | ok... time for some kilojoules... | 21:08 |
yeti | calories? | 21:08 |
_abc_ | yeti: what foundation disabled fsck?! | 21:33 |
phillipsjk | _abc_, I have had systemd time-out on a fsck (with debian): they probably assumed everbody is using journalling fielsystems now. | 22:01 |
_abc_ | I understand a journal replay on an unclean umount on a 1-2TB disk is quite long too. | 22:19 |
gnarface | i hear that meme alot, like complaints that it takes hours but i haven't managed to get anyone to prove to me that happens when you're not using a shitty raid array that boots in PIO mode | 22:24 |
gnarface | but i have been able to prove there are some conspicuous misbehaviors in the e2fsprogs tools in certain versions | 22:26 |
gnarface | and the only clear motive for sabotaging everyone's boot up process seems to be to frame that mistake for the cause of a filesystem corruption glitch that would have happened anyway but someone is trying to astroturf over | 22:27 |
gnarface | ext4 was just not ready for prime time but raspbian project leads seemed to respond as though they had been paid to hide the evidence from noob users | 22:28 |
buZz | hehe PIO mode :D | 22:29 |
gnarface | at the time i just assumed arrogance and incompetence and a pathological inability to own up to even trivial errors in their own judgement, but now i frankly wouldn't be surprised if they were actually paid | 22:29 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!