January 2022 - Arch-general - lists.archlinux.org

[arch-general] Linux server crash causing router switch to stop working
by David Rosenstrauch 16 Feb '22

16 Feb '22

Been experiencing a weird issue several times recently that's got me stumped. A couple of weeks ago, my entire home network went dead right in the middle of a zoom call. Same problem happened again late last night. The problem is intermittent/occasional: everything runs without issue for several days, then suddenly crashes. I figured the issue was a problem with my wifi router, but after much debugging I don't think that's the case. I have a new (Arch) linux server I built a few weeks ago. What seems to be happening is that my server crashes for some reason - and then looks like it's causing the ethernet LAN switch inside my router to stop working. (I verified that when I unplug the server's network cable the router starts working again, and when I plug it back in the router stops.) My best guess as to what's happening is that when the server crashes it's somehow either flooding the network, or sending out invalid network packets, and so making the router inoperable. I have no idea what could cause this, though, or how to fix it. (Or even if this is definitely the issue.) I also have no way to know if this is an issue with my server hardware vs. say an issue in a recent kernel upgrade, since the server is so new. Server is running Arch, and is kept very up to date. It's using the on-board LAN chip on the mobo - Realtek RTL8125 2.5GbE (network module r8169, RTL8125B chip/firmware). Router is a Netgear R8000, running dd-wrt (r47097). Any suggestions welcome! Thanks, DR

7 21

[arch-general] Bazel, reproducible builds and the out-of-date packages fcitx5-mozc & fcitx-mozc
by Alexander Michalopoulos 31 Jan '22

31 Jan '22

Hello, This is a request concerning two out-of-date packages in [community] called fcitx5-mozc and fcitx-mozc. I’m not sure if this is the proper channel to address such things, if it isn’t then I apologize. Fcitx/Fcitx5 is an input engine like IBus, and Mozc is a Japanese IME. There used to be more Mozc packages in [community] (e.g. ibus-mozc) but they were demoted to the AUR since a good while back, the reason being that upstream Mozc has opted to adopt Bazel as its build system, which among other things is a pain to make compatible with Arch’s reproducibility goals. It’s the same thing that has also happened with other packages like e.g. the one for Anki, which was also recently demoted to the AUR for the exact same reason. The result of all this though is that, while these other Mozc packages have in the meantime been adopted and kept up-to-date in the AUR by several people, the fcitx5-mozc and fcitx-mozc packages have been kept outdated in [community] since May 2021, in their last known working GYP version (GYP was the build system used before Bazel but was deprecated in favor of the latter). I understand that Bazel is… not a great piece of software, to put it mildly, and that upstream’s choice to use it is less than optimal; but keeping a package in an almost year-old outdated version is even less optimal in my opinion. Not to mention that even though for most people this kind of package (an input engine add-on for a non-Latin writing system) is obscure at best, for some of us it’s part of our everyday life and may even be crucial (for work or education purposes, etc) and thus keeping it up-to-date with upstream is a very desired goal. My request is, since the maintainer of fcitx5-mozc and fcitx-mozc is probably not interested in updating these two packages to their current upstream versions, that they be demoted to AUR packages so they can be adopted and updated by other interested parties (that would be me). Or, if for whatever reason this isn’t an option, that either an effort is made to update them in situ to their current upstream versions (if such an effort is not already underway) or at least that they are renamed to something different (e.g. fcitx5-mozc-legacy and fcitx-mozc-legacy) so their current names can be adopted and used by new and up-to-date packages in the AUR. Thanks for reading.

1 0

[arch-general] headless box audio with pipewire
by Javier 31 Jan '22

31 Jan '22

Hello ! I have a box, where I can login through SSH, start ncmpcpp to play something with mpd, and then logout, leaving the box playing music, but with no user logged in. At times, I also want to start Xorg (eventually Wayland), and use pipewire for sound and even using bluetooth audio devices (easy with pipewire/pulse). While logged in, I can use pipewire even in console. But once logged out, sound is gone. So pipewire doesn't really work headless by default, is there a way to make it work headless (no one logged in, and still having audio enabled and playing music for example), how so? With alsa, no pipewire, no pulse, all this is pretty simple, since it executes as root, headless audio (no one logged in), audio on console (no Xorg, no Wayland), audio on graphical environmet (Xorg or Wayland). Complex thing with alsa, like using bluetooth devices, or selecting devices is complex. An alternative would be having an easy way to use bluetooth devices with plain alsa (no need for command line for example, since not all users will be able to do that). I think bluez-alsa-git on AUR can help using bluetooth audio devices with plain alsa, but it's somehow complex for non CLI users. Perhaps it's easier to get a box, which when headless, somehow allows alsa to work, and when logged in, pipewire/pulse takes over. But things don't work that way when pipewire/pulse is present, audio is gone when not logged in. Thanks ! -- Javier

3 4

[arch-general] AppArmor
by Frank Zimmermann 28 Jan '22

28 Jan '22

Oh der I made a mess. Wanted to setup Apparmor and Firefox and Evolution aren't working anymore. The programs are starting but without an UI. Even when I undo things this behaviour retains. I don't see anything in the journal and now don't know where to look for what's going wrong. Frank

5 4

[arch-general] VLAN config works only partially
by Rainer Dorsch 22 Jan '22

22 Jan '22

Hi, I configured my network with multiple VLANs. I started with the description at https://wiki.archlinux.org/title/VLAN#systemd-networkd My current configuration: [n3150 ~]# cat /etc/systemd/network/enp2s0.network [Match] Name=enp2s0 [Link] ActivationPolicy=Manual [Network] Description=The unconfigured physical ethernet device # In case of 'tagged only' setups, you probably don't need any IP # configuration on the link without VLAN (or: default VLAN). # For that just omit an [Address] section and disable all the # autoconfiguration magic like this: LinkLocalAddressing=no LLDP=no EmitLLDP=no IPv6AcceptRA=no IPv6SendRA=no ;these are arbitrary names, but must match the *.netdev and *.network files VLAN=enp2s0.1 VLAN=enp2s0.2 VLAN=enp2s0.7 [DHCPv4] UseDNS=false [n3150 ~]# The VLAN config is simple, e.g. n3150% cat /etc/systemd/network/enp2s0.2.netdev [NetDev] Name=enp2s0.2 Kind=vlan [VLAN] Id=2% n3150% cat /etc/systemd/network/enp2s0.2.network [Match] Name=enp2s0.2 [Network] DHCP=yes Hostname=n3150-wlan n3150% The resulting routes: n3150% ip route default via 192.168.0.1 dev enp2s0.1 proto dhcp src 192.168.0.206 metric 1024 default via 192.168.2.1 dev enp2s0.2 proto dhcp src 192.168.2.206 metric 1024 default via 192.168.7.1 dev enp2s0.7 proto dhcp src 192.168.7.206 metric 1024 default via 192.168.0.1 dev enp2s0 proto dhcp metric 20100 192.168.0.0/24 dev enp2s0 proto kernel scope link src 192.168.0.206 metric 100 192.168.0.0/24 dev enp2s0.1 proto kernel scope link src 192.168.0.206 metric 1024 192.168.0.1 dev enp2s0.1 proto dhcp scope link src 192.168.0.206 metric 1024 192.168.2.0/24 dev enp2s0.2 proto kernel scope link src 192.168.2.206 metric 1024 192.168.2.1 dev enp2s0.2 proto dhcp scope link src 192.168.2.206 metric 1024 192.168.7.0/24 dev enp2s0.7 proto kernel scope link src 192.168.7.206 metric 1024 192.168.7.1 dev enp2s0.7 proto dhcp scope link src 192.168.7.206 metric 1024 n3150% Unfortunately, the 192.168.0.* network is not reachable without explicitly deleting the routes for enp2s0: [n3150 ~]# ping 192.168.0.30 PING 192.168.0.30 (192.168.0.30) 56(84) Bytes an Daten. ^C --- 192.168.0.30 ping-Statistik --- 2 Pakete übertragen, 0 empfangen, 100% packet loss, time 1016ms [n3150 ~]# ip route del default via 192.168.0.1 dev enp2s0 proto dhcp metric 20100 [n3150 ~]# ip route del 192.168.0.0/24 dev enp2s0 proto kernel scope link src 192.168.0.206 metric 100 [n3150 ~]# ping 192.168.0.30 PING 192.168.0.30 (192.168.0.30) 56(84) Bytes an Daten. 64 Bytes von 192.168.0.30: icmp_seq=1 ttl=64 Zeit=0.699 ms 64 Bytes von 192.168.0.30: icmp_seq=2 ttl=64 Zeit=0.287 ms ^C --- 192.168.0.30 ping-Statistik --- 2 Pakete übertragen, 2 empfangen, 0% packet loss, time 1010ms rtt min/avg/max/mdev = 0.287/0.493/0.699/0.206 ms [n3150 ~]# Can anybody tell how I can avoid the routes for enp2s0 get configured? Thanks Rainer

2 2

[arch-general] Env variable of user unit of ssh-agent not set
by Rainer Dorsch 22 Jan '22

22 Jan '22

Hi, I followed https://wiki.archlinux.org/title/SSH_keys#Start_ssh-agent_with_systemd_user It seems that SSH_AUTH_SOCK does not get set although it is in the user unit file. $ systemctl --user status ssh-agent.service ● ssh-agent.service - SSH key agent Loaded: loaded (/home/rd/.config/systemd/user/ssh-agent.service; enabled; vendor preset: enabled) Active: active (running) since Fri 2022-01-21 21:14:16 CET; 7min ago Main PID: 574 (ssh-agent) Tasks: 1 (limit: 9331) Memory: 920.0K CPU: 9ms CGroup: /user.slice/user-2809.slice/user(a)2809.service/app.slice/ssh- agent.service └─574 /usr/bin/ssh-agent -D -a /run/user/2809/ssh-agent.socket Jan 21 21:14:16 n3150 ssh-agent[574]: SSH_AUTH_SOCK=/run/user/2809/ssh- agent.socket; export SSH_AUTH_SOCK; Jan 21 21:14:16 n3150 ssh-agent[574]: echo Agent pid 574; Jan 21 21:14:16 n3150 systemd[564]: Started SSH key agent. $ cat .config/systemd/user/ssh-agent.service [Unit] Description=SSH key agent [Service] Type=simple Environment=SSH_AUTH_SOCK=%t/ssh-agent.socket # DISPLAY required for ssh-askpass to work Environment=DISPLAY=:0 ExecStart=/usr/bin/ssh-agent -D -a $SSH_AUTH_SOCK [Install] WantedBy=default.target $ ssh-add Could not open a connection to your authentication agent. $ echo $SSH_AUTH_SOCK $ Can anybody tell what I am missing? Any hint or advice is welcome. Thanks Rainer

3 3

[arch-general] Can't mount vfat filesystem
by Jeanette C. 22 Jan '22

22 Jan '22

Hey hey, I just updated my system yesterday and now I can't mount a VFAT filesystem. Installed packages: linux 5.16.2.arch1-1 util-linux 2.37.2-5 There is no mount.vfat or mount.fat program/script and apparently no kernel module installed that contains "fat" in its name. Is there a change in the way to mount such filesystems? Best wishes and thanks for any hints, Jeanette -- * Website: http://juliencoder.de - for summer is a state of sound * Youtube: https://www.youtube.com/channel/UCMS4rfGrTwz8W7jhC1Jnv7g * Audiobombs: https://www.audiobombs.com/users/jeanette_c * GitHub: https://github.com/jeanette-c Baby, there's something about you Show me, that nothing can hold you down <3 (Britney Spears)

3 4

[arch-general] Grub installiert issue
by Frank Zimmermann 19 Jan '22

19 Jan '22

I'm doing my first Arch install on an UEFI system and have troubles with GRUB. Following the instruction on the Wiki grub-mkconfig complained about a missing /boot/grub/grub.cfg.new. So I manually created a grub.cfg file but when rebooting Grub says error: unknown file system. In rescue mode I type set root=(hd0,gpt1) set prefix=(0,1)/boot/grub insmod normal error: disk '0,1' not found set prefix=(hd0,gpt1)/boot/grub insmod normal error: unknown filesystem set prefix=(0,gpt1)/boot/grub insmod normal error: disk 0,gpt1 not found My efi partition is the first partition on the first ssd. I think I did all I found on the Wiki and search on the web but am lost now with generating the grub.cfg Any hints please? Frank

5 6

[arch-general] pacman caches packages with u=rw,g=,o= permissions
by u34＠net9.ga 19 Jan '22

19 Jan '22

Lately, possibly starting at mid December, files in /var/cache/pacman/pkg/ are kept with mode u=rw,g=,o= . Am I the only one having it? -- jadon

4 8

[arch-general] A correction for my response about permissions of pacman packages cache
by u34＠net9.ga 19 Jan '22

19 Jan '22

At Tue Jan 18 23:44:56 UTC 2022, at the bottom of https://lists.archlinux.org/pipermail/arch-general/2022-January/049521.html , I wrote that my understanding is that checkupdates -d is equivalent to pacman -Sywu && pacman -Su . I haven't read checkupdates manual page. It could be that other then a safe upgrade it does additional work. I still haven't read checkupdates manual page. Rereading mpan, who is the author of the message I responded to, message, I should have wrote that my understanding is that checkupdates -d is equivalent to pacman -Sywu . Not to pacman -Sywu && pacman -Su . -- jadon

1 0