| eyalroz | Has anyone noticed crashes of transmission-gtk on Excalibur recently? | 00:24 |
|---|---|---|
| plasma41 | eyalroz: Does it give any error message when it crashes? | 00:27 |
| eyalroz | plasma41: GUI-wise, no, it crashes silently. I need to start it from a terminal and see. | 00:27 |
| eyalroz | I don't suppose it logs failures anywhere? | 00:27 |
| eyalroz | Now, it doesn't crash immediately, it's just that whenever I look for it after I haven't touched it for a few hours, it's gone... | 00:28 |
| plasma41 | I don't know. Whatever the manpage says | 00:28 |
| Xenguy | eyalroz, qbittorrent may be worth a look, if you need an alternative | 00:29 |
| eyalroz | Xenguy: I was actually quite ok with transmission-gtk, though if I can't get it to stay up, I suppose I'll try qbt. | 00:29 |
| eyalroz | but - this is a distro channel, the answer should be "hmm, let us try and fix that" :-) | 00:30 |
| Xenguy | Who's to say the problem is not upstream? Hard to know, unless you gather more info | 00:31 |
| plasma41 | eyalroz: Were you able to get an error message upon a crash? | 01:11 |
| soolapimo | hello. I'm running Ceres and on my last update, I noticed that some things started to fail: xorg doesn't read my inputs, no sound card detected, etc. I've also noticed that kernel 6.6.9 fails to install. What could I do to get around this? | 11:43 |
| soolapimo | Error log: https://termbin.com/awnc | 11:45 |
| soolapimo | I think the two issues (kernel not being installed and devices not being detected) are related, but I wonder what could be the fix for this; wait for the headers package? | 11:46 |
| Joril | soolapimo: Are you running a usrmerged system? | 11:48 |
| soolapimo | Joril: I'm not sure, actually. I suspect that's part of the issue, since before these logs, a problem existed that things like /sbin/modprobe weren't found | 11:53 |
| soolapimo | how can I check? | 11:53 |
| Joril | check if /bin is a symlink to /usr/bin | 11:54 |
| soolapimo | I ran `file /bin` and it says "directory" so I guess not? | 11:59 |
| fsmithred | I ran into this a couple days ago. I upgraded and got 6.6.9 kernel and also added usrmerge. Files were missing and I had to reinstall lvm2 and dmsetup. That got rid of two errors. Also reinstalled xorg xinit xauth and every xserver-xorg-* that was installed. Then reinstalled elogind libpam-elogind and policykit-1-gnome. I can now boot the 6.6.9 kernel and use startx, but lightdm is caught in an endless login loop. | 11:59 |
| fsmithred | ls -l / | 11:59 |
| fsmithred | will show you if /bin /sbin and /lib* are symlinks | 11:59 |
| avir327 | Hallo Devuaneers! After the latest Excalibur update, I get boot errors 'sed: not found' and (probably in consequence), the NICs are not brought up. After reverting to the "base installation" snapshot from ~one month ago, everything is fine again. | 12:00 |
| avir327 | This is reproducable. | 12:00 |
| fsmithred | sed and other executables are now under /usr | 12:00 |
| fsmithred | You should be able to add the usrmerge package to set all the links, but it seems to be broken. | 12:02 |
| fsmithred | Several people have had similar problems in the past week. | 12:02 |
| avir327 | Did I miss any changelog? I think no... | 12:02 |
| avir327 | Is there any documentation? | 12:03 |
| fsmithred | check with debian | 12:03 |
| avir327 | Is this an upstream issue? | 12:03 |
| fsmithred | yeah, I think they want to prove that they can force their stupid decisions onto us, but telling them that won't make it easier to get help. | 12:04 |
| fsmithred | we're working on solving it | 12:04 |
| avir327 | OK thank you. | 12:06 |
| avir327 | I mean: Nice. Thank you! | 12:06 |
| soolapimo | fsmithred: thanks for the information. I am not yet on that level of debugging yet, but your xorg problems seem similar. ls -l tells me that both /bin and /sbin are indeed directories as well. I have not heard about usrmerge before. Should I install that package? | 12:11 |
| soolapimo | (and so are /lib*) | 12:11 |
| avir327 | soolapimo: installing 'usrmerge' seems to work fine. Symlinks are created, right now I am running the update again. | 12:14 |
| avir327 | up and running, 'usrmerge' can then be purged. | 12:19 |
| avir327 | Damn arrogance of power. I am curious for the Devuan fix. | 12:22 |
| fsmithred | there are some recent discussions about usrmerge on the forum - dev1galaxy.org | 12:24 |
| fsmithred | I have to go. See you all later. | 12:24 |
| soolapimo | avir327: thanks. I'll give it a shot now | 12:28 |
| soolapimo | and thanks fsmithred for the background information! | 12:28 |
| soolapimo | avir327: did you just install the package and that was it? No need to run anything? I'm still getting the same errors from dpkg wrt installing 6.6.9 | 12:35 |
| eyalroz | plasma41: So, my transmission-gtk crashes are apparently due to a segmentation fault (!) | 12:37 |
| eyalroz | Xenguy: ping in case this tells you something | 12:37 |
| avir327 | soolapimo: Yes, I just installed (and then purged) it. | 12:39 |
| avir327 | soolapimo: I missed your errors, joined this channel later. | 12:40 |
| soolapimo | I see, I'm out of things to try then, I guess. This is the error message (which still looks the same after usrmerge) https://termbin.com/awnc | 12:48 |
| avir327 | soolapimo: did you try with 'apt --fix-broken install' and 'dpkg --configure --pending'? | 13:15 |
| soolapimo | avir327: just did, but the error messages remain... 'guess I should try a reboot eh | 13:25 |
| avir327 | do you have a working kernel? | 13:29 |
| soolapimo | yes, 6.6.8 still loaded and working here. see you in a bit | 13:36 |
| al1r4d | hey, what is purpose of #devuan-infra :/ | 15:38 |
| onefang | That's for discussing the Devuan infrastructure. | 15:50 |
| soolapimo | hello, earlier today I reported a problem here with upgrading to kernel 6.6.9 that broke my device drivers. This had to do with the usrmerge thing that I hadn't done prior, but even after installing the package it wouldn't be resolved. I'm now happy to report that I've solved the issue myself, and would like to share the procedure I took to do so: https://termbin.com/lpug | 22:21 |
| soolapimo | thank you all who cued me on this issue. I might do a longer writeup later, but this is the sparknotes version of it | 22:22 |
| Nietz | we are grateful for your contribution! thank you for being a part of our community | 22:24 |
| gnu_srs1 | soolapimo: Couldn't you avoid usrmerge by symlinking? | 22:26 |
| soolapimo | thanks Nietz! Frankly, I don't hang out here very often, though. I should try to give back a little more, or in the forums. I'm glad to help, anyway :) | 22:32 |
| soolapimo | gnu_srs1: I tried it for a few files (modprobe, modinfo, etc) but eventually realized there were too many of them. It would become unmanageable. Also, a lot of the files that were duplicated belonged to the most random of packages, like sane. I could not possibly guess that many files without that script. | 22:34 |
| onefang | gnu_srs1: or just go with the symlinking that usrmerge is currently doing. Simples! BTW that symlinking is just a handful of directories, not hundreds of files. | 23:21 |
| gnu_srs1 | I meant doing it reverse: ln -s /usr/sbin/modprobe /sbin etc | 23:39 |
| rwp | I might suggest using relative links in these cases. Because then if the storage is mounted on a rescue media or something at say /mnt then the relative links still point to the correct place. Just makes things work easier then. | 23:48 |
| gnu_srs1 | rwp: Is (cd /usr/sbin; ln -s ./modprobe ../sbin) etc OK? | 23:54 |
| rrq | it's the link that should be relative, not the command that sets it | 23:59 |
| rwp | I suggest the easiest way to mentally get things correct is to cd to the destination directory and then "ln -s ../from/the/source .". | 23:59 |
| rrq | you would be thinking: cd /sbin ; ln -s ../usr/sbin/modprobe . | 23:59 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!