| adhoc | morning all | 02:59 |
|---|---|---|
| golinux | Evening all. ;) | 03:02 |
| * adhoc sips the first coffee of the morning. | 03:04 | |
| adhoc | What is news in terms of coming releases/changes? | 03:04 |
| fsmithred | 423 packages can be upgraded. That's since 27July | 03:18 |
| adhoc | I better go patch things, BBL | 03:40 |
| onefang | I need to reboot, also BBL. | 03:40 |
| * adhoc needs to find a better way to WoL thirty ish machines, patch them and power most of them back down again... | 03:46 | |
| jmjl | oh testing | 04:24 |
| jmjl | I should've switched to daedalus, I was on chimaera | 04:25 |
| gnarface | daedalus is now current stable | 04:27 |
| adhoc | what is the current life expectancy of chimeara? | 04:36 |
| bradd | Hi. is there a way to make a 'persistant' network interface name, rather than the ethXX names.. devuan seems to be re-ordering my eth2 and eth3 interfaces | 04:45 |
| gnarface | yea, there's a way to do that... are they hotplug or something? | 05:05 |
| gnarface | usually it's not a problem for pci devices | 05:05 |
| gnarface | usually what's supposed to happen is udev should cache them in /etc/udev/rules.d/70-persistent-net.rules, keyed by MAC address | 05:06 |
| gnarface | oh, hmm... if you were using MAC address randomization maybe that could throw it off... not sure | 05:07 |
| gnarface | anyway, nothing will stop you from making a udev rule keyed by whatever you want that names them whatever you want. not sure if there's an easier way (maybe a module option or something? maybe a network-manager setting if you're using that...) | 05:08 |
| Xenguy | adhoc, Chimaera is good for a few years now | 05:09 |
| gnarface | normally it shouldn't be an issue unless they're usb devices | 05:09 |
| gnarface | usb devices just come up in the order they're plugged in, and behave notoriously unpredictably if plugged in at the same time or already plugged in from a previous boot | 05:10 |
| rustyaxe | usb ethernet devices are teh creation of the devil.. Mostly because they almost inevitably are realshrek chipsets inside :( | 05:17 |
| bradd | ok, no I've added a new nic to the system.. so maybe now (if I dont change anything), I can just use the ethXX device without fear of it changing on me? | 05:24 |
| amarsh04 | On the /usrmerge front I tried to install base-files:amd64 13.3devuan1 -> 13.4devuan1 but it failed due to /bin -> /usr/bin and /sbin -> /usr/sbin symbolic links still present | 05:24 |
| amarsh04 | removing those symbolic links caused booting to fail as the checking and mounting of root had the old paths hard-coded | 05:26 |
| bradd | gnarface, I'm getting in dmesg -> "Error changing net interface name eth2 to eth1: File exists" (although everything seems to be working) | 05:31 |
| adhoc | Xenguy: thanks for the update. | 05:45 |
| gnarface | bradd: find this file and just edit by hand then: ls -l /etc/udev/rules.d/*persistent-net* | 06:44 |
| gnarface | or, delete it and presumably it'll regenerate it correctly (probably a good idea to make a backup just in case) | 06:45 |
| bradd | ok, I'll give it a try | 06:48 |
| bradd | gnarface: well it got rid of the warning, but it didnt regenerate that persistant file | 07:11 |
| amarsh04 | updated the forum with my /usrmerge situation: https://dev1galaxy.org/viewtopic.php?pid=51466#p51466 | 07:20 |
| gnarface | bradd: that's weird... you sure you have udev or eudev actually installed? which release is this? | 07:35 |
| gnarface | if you've just recently upgraded it could have been removed accidentally | 07:38 |
| bradd | how do I tell? (uname -a gives debian 6.1.69-1 | 07:44 |
| bradd | (sources.list says its using daedalus) | 07:48 |
| CueXXIII | amarsh04: how did it fail exactly? those symlinks should be there after usrmerge | 07:51 |
| CueXXIII | amarsh04: or do you mean manually created symlinks from /bin/somefile -> /usr/bin/somefile? not sure if usrmerge can handle that situation | 07:55 |
| gnarface | bradd: try "dpkg -l |grep udev" | 08:12 |
| bradd | ii eudev 3.2.12-4+deb12u1 amd64 /dev/ and hotplug management daemon | 08:12 |
| bradd | ii libeudev1:amd64 3.2.12-4+deb12u1 amd64 libeudev shared library | 08:12 |
| gnarface | no no, don't paste | 08:12 |
| gnarface | yea, looks installed though | 08:12 |
| gnarface | hmm, weird that it's not working | 08:12 |
| gnarface | how about eth0? did it show up correctly at least? | 08:13 |
| gnarface | maybe that file doesn't actually get written right away, i'm not sure | 08:13 |
| bradd | oh. well the networking has been working now, so maybe at some point I'll just generate the udev file manually | 08:14 |
| bradd | i.e. with mac address mappings | 08:14 |
| gnarface | on mine it says it's generated by the /lib/udev/write_net_rules binary | 08:15 |
| gnarface | which presumably is referenced by /lib/udev/rules.d/75-persistent-net-generator.rules (not sure the index numbers are static on those) | 08:15 |
| gnarface | i've had problems before with those not getting updated right but not recently | 08:16 |
| gnarface | (the files in /lib/udev/ that is) | 08:16 |
| bradd | odd, but after removing that perisitant file, there is nothing in that directory | 08:16 |
| bradd | maybe I should try a apt upgrade ? | 08:17 |
| gnarface | to be clear, do not remove the /lib/udev/rules.d/75-persistent-net-generator.rules file, just the /etc/udev/rules.d/70-persistent-net.rules file | 08:17 |
| gnarface | i think you have the most recent udev, but if you're current and your /etc/apt/sources.list is correct, this shouldn't hurt anything: apt-get update && apt-get --no-install-recommends dist-upgrade | 08:18 |
| bradd | there was only the 70-* file.. no other rule files in that dir | 08:18 |
| bradd | ok, I'll give it a shot | 08:18 |
| gnarface | uh, well the only other auto-generated file that goes in there that i know of would be 70-persistent-cd.rules, which would only appear if you have an optical drive | 08:18 |
| gnarface | basically, the system's udev rules go in /lib/udev/rules.d, and /etc/udev/rules.d is for your own custom ones or auto-generated ones | 08:19 |
| bradd | oh, i see | 08:19 |
| gnarface | i don't recall any issues with it not getting created before... anything's possible though | 08:21 |
| bradd | ok. did the dist-upgrade. nothing in /etc/udev/rules.d | 08:23 |
| gnarface | make sure udev is running: ps aux |grep udev | 08:23 |
| bradd | yep. /sbin/udevd --daemon | 08:23 |
| gnarface | then maybe uh... "/etc/init.d/eudev reload" | 08:24 |
| gnarface | though i can't imagine why that would work if it didn't work on startup... maybe if some other relevant log error turns up | 08:24 |
| gnarface | there is a newer kernel than that in the repo, but that one doesn't look very much older so i doubt it's the issue | 08:25 |
| gnarface | if some other error happened to be in the logs that might help though, at this point i'm out of ideas | 08:25 |
| gnarface | just by way of sanity check, what's the md5sum of your /lib/udev/rules.d/75-persistent-net-generator.rules file? i've got 6578220eeaa8e957de1464d46a34c3f7 here | 08:26 |
| bradd | ok. I'll run it like this for now.. maybe tomorrow I'll make the rules.d file | 08:26 |
| gnarface | this is a pure daedalus system? no other stuff mixed in? the only other thing i could really do is sanity check your /etc/apt/sources.list | 08:27 |
| bradd | same md5 | 08:27 |
| gnarface | if you paste it at paste.debian.net i'll look | 08:27 |
| gnarface | oh, hmm... just thought of one more thing but it's a long shot: df -h | 08:28 |
| gnarface | make sure the partition /etc is on isn't full | 08:28 |
| gnarface | though i'd expect udev to complain about that i don't actually know... | 08:28 |
| bradd | /dev/md0 (/) is at 80% used | 08:29 |
| gnarface | hmm, that's probably not the issue then | 08:29 |
| gnarface | sorry, out of ideas. let me know if you figure it out, and stick around in case someone else has an idea. | 08:30 |
| bradd | ok, thats for helping | 08:30 |
| gnarface | there must be something obvious that i'm forgetting, i'd think... usually the problem isn't with that file not generating, usually the complaint is about stopping it from changing | 08:31 |
| AlexLikeRock | adios gnarface o/ | 08:40 |
| amarsh04 | CUEXXIII "The base-files package cannot be installed because /bin is a symbolic link and not pointing at usr/bin exactly. | 10:54 |
| CueXXIII | amarsh04: ls -ld /bin | 11:14 |
| CueXXIII | it should point to "usr/bin", not "/usr/bin" nor "../usr/bin" | 11:14 |
| amarsh04 | ah, thanks | 11:17 |
| AlexLikeRock | o/ | 23:30 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!