Bugzilla – Bug 1055834
[FAILED] Failed to start Setup Virtual Console.
Last modified: 2021-03-02 16:28:58 UTC
My first bug report. Please excuse irregularities. After a recent Tumbleweed update (20170822?) I began to receive this error on boot: [FAILED] Failed to start Setup Virtual Console. The details as reported: eric@linux-epth:~$ sudo systemctl status systemd-vconsole-setup.service ● systemd-vconsole-setup.service - Setup Virtual Console Loaded: loaded (/usr/lib/systemd/system/systemd-vconsole-setup.service; static; vendor preset: disabled) Active: inactive (dead) since Thu 2017-08-24 20:04:51 CDT; 26min ago Docs: man:systemd-vconsole-setup.service(8) man:vconsole.conf(5) Process: 467 ExecStart=/usr/lib/systemd/systemd-vconsole-setup (code=exited, status=0/SUCCESS) Main PID: 467 (code=exited, status=0/SUCCESS) Aug 24 20:04:51 linux-epth systemd-vconsole-setup[467]: KD_FONT_OP_SET failed, fonts will not be copied to tty7: Invalid argument Aug 24 20:04:51 linux-epth systemd[1]: Started Setup Virtual Console. eric@linux-epth:~$ More research and think this is tied to a boot problem as systemctl work manually: eric@linux-epth:~$ sudo systemctl status systemd-vconsole-setup.service [sudo] password for root: ● systemd-vconsole-setup.service - Setup Virtual Console Loaded: loaded (/usr/lib/systemd/system/systemd-vconsole-setup.service; static; vendor preset: disabled) Active: inactive (dead) since Sat 2017-08-26 08:54:42 CDT; 1min 51s ago Docs: man:systemd-vconsole-setup.service(8) man:vconsole.conf(5) Process: 441 ExecStart=/usr/lib/systemd/systemd-vconsole-setup (code=exited, status=0/SUCCESS) Main PID: 441 (code=exited, status=0/SUCCESS) Aug 26 08:54:42 linux-epth systemd-vconsole-setup[441]: KD_FONT_OP_SET failed, fonts will not be copied to tty7: Invalid argument Aug 26 08:54:42 linux-epth systemd[1]: Started Setup Virtual Console. eric@linux-epth:~$ sudo systemctl start systemd-vconsole-setup.service eric@linux-epth:~$ sudo systemctl status systemd-vconsole-setup.service ● systemd-vconsole-setup.service - Setup Virtual Console Loaded: loaded (/usr/lib/systemd/system/systemd-vconsole-setup.service; static; vendor preset: disabled) Active: inactive (dead) since Sat 2017-08-26 08:57:03 CDT; 4s ago Docs: man:systemd-vconsole-setup.service(8) man:vconsole.conf(5) Process: 2637 ExecStart=/usr/lib/systemd/systemd-vconsole-setup (code=exited, status=0/SUCCESS) Main PID: 2637 (code=exited, status=0/SUCCESS) Aug 26 08:57:02 linux-epth systemd[1]: Starting Setup Virtual Console... Aug 26 08:57:03 linux-epth systemd[1]: Started Setup Virtual Console. eric@linux-epth:~$ My thanks to tsu2 on the user forum (Re: New Boot Error -"[FAILED] Failed to start Setup Virtual Console."): "Looks like this is likely a systemd upstream error. I found where your error is generated, and I doubt anyone but upstream can do anything about it https://github.com/systemd/systemd/blob/master/src/vconsole/vconsole-setup.c Best to just submit a bug to https://bugzilla.opensuse.org TSU" Please note line 306 of referenced code as it pertains to warning I received: log_warning_errno(errno, "KD_FONT_OP_SET failed, fonts will not be copied to tty%u: %m", i); KD_FONT_OP_SET failed, fonts will not be copied to tty7: Invalid argument Thank you for your help with this.
.../usr/bin/loadkeys tool fails... "poettering commented 30 minutes ago Ah, this reveals it. It's not actually vconsole at fault here, but your /usr/bin/loadkeys tool fails with a failure exit code (and we invoke that, and propagate the exit code). The precise command we invoke is this: /usr/bin/loadkeys -q -C /dev/tty1 -u us Which results in a tonload "Invalid argument" messages from loadkeys. Does it work if you invoke it locally from a shell? It appears to me you should contact your distro for help around this, as it's the kbd package at fault here, not systemd/vconsole" Please see this thread for the error as reported by systemd (above): https://github.com/systemd/systemd/issues/6689
Adding kbd maintainers to cc, should possibly reassign bug away from systemdmaintainers too, but leaving that in place for now.
Hi Eric, for debugging this. Could you try to break in the initramfs and try to execute the following command: $ loadkeys -C /dev/tty1 -u us In order to break in the initramfs you can add to the kernel command line "rd.break=pre-mount" option. Thanks.
I have read the whole upstream bug. The error "KD_FONT_OP_SET failed,..." comes from systemd-vconsole-setup and it is displayed if ioctl(, KDFONTOP,) fails. It has nothing to do with kbd. Aug 30 11:57:35 linux-epth systemd-vconsole-setup[155]: Executing "/usr/bin/setfont -C /dev/tty1 -m trivial lat9w-16.psfu"...Aug 30 11:57:37 linux-epth systemd-vconsole-setup[155]: /usr/bin/loadkeys failed with error code 1. Really comes from kbd. Aug 30 11:57:37 linux-epth systemd-vconsole-setup[155]: KD_FONT_OP_GET failed while trying to get the font metadata: Input/output error Aug 30 11:57:* linux-epth systemd-vconsole-setup[155]: Input/output error (many times) Aug 30 11:57:37 linux-epth systemd-vconsole-setup[155]: KDSKBENT: Input/output error: could not deallocate keymap 128 But as both errors started to appear together, so I suspect that they are caused by changes in the graphic subsystem (we seen similar problems in past, after introduction of a new graphic splash). Let's see whether it works from the early command line.
I'm sorry but I do not have access to this computer (Tumbleweed) any longer and am afraid I can't help you with further debugging. Hope you fix it, Eric
Well in that case I'm afraid we won't be able to debug it further, therefore closing.
Hello, I have the same bug and I have always access to the tumbleweed PC, also, I have performed the steps you requested. After breaking in initramfs, the loadkeys -C /dev/tty1 -u us fails, but this command would work: loadkeys -C /dev/tty1 -u gb note that inside initramfs, the /etc/vconsole.conf file contains a single line: KEYMAP=gb I'm attaching a photo of the entire session.
Created attachment 740462 [details] loadkeys commands
Created attachment 740463 [details] etc vconsole conf
Comment 7: So manual setting of the keymap works, but call from systemd-vconsole-setup does not. It looks like a race condition: Keymap and font cannot be set after initializing of the graphic subsystem. (The us keymap is probably not present, which is irrelevant to the reported bug.)
Please let me know if you need any additional testing (I have just updated my email address that was obsolete).
(In reply to Lorenzo Paulatto from comment #11) > Please let me know if you need any additional testing (I have just updated > my email address that was obsolete). Could you try to disable plymouth (add "plymouth.enable=0" to the kernel command line) and see if the error fades away ?
Also can you please boot with the following options added to tje kernel command line "debug printk.devkmsg=on" and post the result of "journalctl -b" ? Thanks.
Created attachment 740771 [details] output of "journalctl -b " without changing any kernel parameter
Created attachment 740773 [details] output of "journalctl -b " without plymouth disabled Here is the log with all the options, i.e. "plymouth.enable=0" and "debug" and "printk.devkmsg=on". Adding just "plymouth.enable=0" the error "Failed to start Setup Virtual Console." I guess, also with devkmsg enabled, but it is difficult to see because of too much output. P.S. I won't be able to do more kernel testing until Monday, as the PC is at work. P.P.S I am using nVIdia drivers with dkms , which causes plymouth to not properly work anyway, but disabling it by hand I see the boot messages on screen, which I usually do not
Sorry for the garbled sentence: Adding just "plymouth.enable=0" the error "Failed to start Setup Virtual Console." I mean that the error occurs even with plymouth.enable=0. I also want to point out that I saw the error 100% of the times, over around 10 reboots. If it is a race condition, it is a rigged race.
Thank you Lorenzo for taking this bug over. For what it's worth, I had the error far more times than I did not so I also thought it was a rigged race! Also - the error began (I think) after a change to dracut - it is a relatively new problem. Last, I noticed this error on a pc which has a built-in graphics chip (i915 driver) and it displays the plymouth tumbleweed animation correctly most boots. I have another machine also running the identical TW setup but it has an Nvidia chip and runs with nouveau. The second machine rarely displays the correct plymouth animation - I get a black screen with three small rectangles with flashing "?" where the the flashing blue dots would be if plymouth were working. Hope this helps...
Oh yes... the second pc very, very rarely reports the bug on boot...
(In reply to Lorenzo Paulatto from comment #16) > Sorry for the garbled sentence: Adding just "plymouth.enable=0" the error > "Failed to start Setup Virtual Console." > > I mean that the error occurs even with plymouth.enable=0. > This is still not coherent... According to the logs you attached: - with "debug printk.devkmsg=on" -> the error DOES occur - with "plymouth.enable=0 debug printk.devkmsg=on" -> the error does NOT occur
Could you try the following with plymouth *enabled*: - modify /usr/lib/systemd/system/plymouth-start.service in order to have: Wants=systemd-ask-password-plymouth.path IOW remove systemd-vconsole-setup.service from Wants= - regenerate the initrd with "mkinitrd" - reboot and see if the error is gone (with plymouth enabled) Thanks.
This seems to have worked, after editing plymouth-start.service and regenerating the init image, I've rebooted a few times and never got the error message.
Thanks for the report. I've updated https://github.com/systemd/systemd/issues/6689 and waiting for some feedbacks before making conclusion.
Lorenzo, could you try to boot with "init=/bin/sh" added to the kernel command line and once the shell appears, try: loadkeys -q -C /dev/tty1 -u us Does it fail ?
Created attachment 741183 [details] loadkeys suceeds in minimal sh after editing vconsole.service This took a bit longer than expected, because my USB keyboard did not work in the minimal shell, I had to find one PS2, which turned out to be French. Anyway, I confirm that loadkeys succeeds now. I've attached a screen picture
I have the same issue: http://susepaste.org/60562051
I have the same problem after my today's upgrade.
@Vinicius, @Petr, could you report that back to the ticket opened at github: https://github.com/systemd/systemd/issues/6689 Could you see if the workaround posted in comment #20 works for you ? Thanks.
Same issue on an older PC of mine. Funny enough: > - modify /usr/lib/systemd/system/plymouth-start.service in order to have: > Wants=systemd-ask-password-plymouth.path > > - regenerate the initrd with "mkinitrd" didnt change anything.
Problem still present in Tumbleweed 20171125. Will try suggestion in comment #20.
I have the same issue. The solution from comment #20 doesn't help as /usr/lib/systemd/system/plymouth-start.service already have systemd-ask-password-plymouth.path in Wants list.
I found the following messages in system journal: Dec 01 20:29:40 vz-lapbook systemd[1]: systemd-vconsole-setup.service: Unit entered failed state. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: systemd-vconsole-setup.service: Failed with result 'exit-code'. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Starting Show Plymouth Boot Screen... Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Started Show Plymouth Boot Screen. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook kernel: usb 1-8: new high-speed USB device number 6 using xhci_hcd Dec 01 20:29:40 vz-lapbook systemd[1]: Reached target Paths. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Started Forward Password Requests to Plymouth Directory Watch. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Reached target Basic System. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Reached target Initrd Default Target. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Starting Cleaning Up and Shutting Down Daemons... Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Stopped target Timers. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Starting Setup Virtual Console... Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Stopped target Initrd Default Target. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Stopped target Remote File Systems. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Stopped target Remote File Systems (Pre). Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Stopped target Basic System. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Stopped target System Initialization. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Stopped udev Coldplug all Devices. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Stopped dracut pre-trigger hook. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Stopped Apply Kernel Variables. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error .... Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook kernel: systemd: 20 output lines suppressed due to ratelimiting Dec 01 20:29:40 vz-lapbook systemd[1]: Stopped dracut ask for additional cmdline parameters. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Started Cleaning Up and Shutting Down Daemons. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Started Plymouth switch root service. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Stopped udev Kernel Device Manager. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Stopped Create Static Device Nodes in /dev. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Stopped Create list of required static device nodes for the current kernel. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Closed udev Kernel Socket. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Closed udev Control Socket. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Starting Cleanup udevd DB... Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Started Cleanup udevd DB. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook kernel: mmc2: new ultra high speed SDR50 SDXC card at address aaaa Dec 01 20:29:40 vz-lapbook kernel: mmcblk2: mmc2:aaaa SL64G 59.5 GiB Dec 01 20:29:40 vz-lapbook systemd[1]: Started Setup Virtual Console. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook kernel: mmcblk2: p1 Dec 01 20:29:40 vz-lapbook kernel: sdhci-pci 0000:00:1b.0: SDHCI controller found [8086:5aca] (rev b) Dec 01 20:29:40 vz-lapbook systemd[1]: RTC configured in localtime, applying delta of 180 minutes to system time. Dec 01 20:29:40 vz-lapbook systemd[1]: Reached target Switch Root. Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd[1]: Starting Switch Root... Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: Input/output error Dec 01 20:29:40 vz-lapbook systemd-vconsole-setup[159]: KDSKBENT: Input/output error: could not deallocate keymap 128
*** Bug 1073066 has been marked as a duplicate of this bug. ***
I am experiencing the exact issue on Asus laptop with Nvidia Optimus/Intel combo graphics. It started after 01/10 update. I've tried everything mentioned here, but so far nothing fixes the issue. No longer getting the message after I modified systemd plymouth service. only seeing 3 dots but no loggin menu. Please advise if someone knows how to fix this issue. Thanks
I am experiencing the same bug since a Tumbleweed Update back in August 2017. It has persisted on my system (Tumbleweed, KDE / Plasma, German, no manual interventions) despite the huge number of system updates since. I should also like to ask for any advice towards a solution, please. Thanks. Please do also refer to the Forum thread https://forums.opensuse.org/showthread.php/526731-systemd-vconsole-setup-failed-(-Fonts-will-not-be-copied-to-remaining-consoles).
(In reply to Ludger Ricken from comment #34) > I am experiencing the same bug since a Tumbleweed Update back in August > 2017. It has persisted on my system (Tumbleweed, KDE / Plasma, German, no > manual interventions) despite the huge number of system updates since. I > should also like to ask for any advice towards a solution, please. > > Thanks. > > Please do also refer to the Forum thread > https://forums.opensuse.org/showthread.php/526731-systemd-vconsole-setup- > failed-(-Fonts-will-not-be-copied-to-remaining-consoles). Adding info: NVIDIA graphics card, nouveau driver.
I can also confirm this bug on a VirtualBox guest running on EFI emulation, with a fast SSD disk on the host. It started happening sometime in January/2018. The fix in #20 works.
Unfortunately, no progress on my system. In short, the fix from Comment #20 doesn't help. Now a longer write-up, hoping for some clarification. Please excuse any irregularities in this bug report, haven't gotten deeper into the attaching and quoting rules here. My Tumbleweed was set-up from scratch in July 2017. It worked fine (including all updates coming in) until late in August 2017, when the error message [FAILED] Failed to start Setup Virtual Console. See 'systemctl status systemd-vconsole-setup.service' for details. Starting Show Plymouth Boot Screen... suddenly appeared and never went away despite all updates since. ---------- This is my system: inxi -Fxzd System: Host: susytmblwdke.fritz.box Kernel: 4.14.11-1-default x86_64 bits: 64 gcc: 7.2.1 Desktop: KDE Plasma 5.11.5 (Qt 5.10.0) Distro: openSUSE Tumbleweed 20180106 Machine: Device: laptop System: Hewlett-Packard product: HP EliteBook 8560w v: A0001C02 serial: N/A Mobo: Hewlett-Packard model: 1631 v: KBC Version 01.3F serial: N/A UEFI: Hewlett-Packard v: 68SVD Ver. F.63 date: 10/27/2016 Battery BAT0: charge: 54.3 Wh 100.0% condition: 54.3/54.3 Wh (100%) model: Hewlett-Packard Primary status: Full hidpp__0: charge: 95% condition: NA/NA Wh model: Logitech M705 status: Discharging CPU: Dual core Intel Core i5-2540M (-HT-MCP-) arch: Sandy Bridge rev.7 cache: 3072 KB flags: (lm nx sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx) bmips: 10376 clock speeds: max: 3300 MHz 1: 2594 MHz 2: 2594 MHz 3: 2594 MHz 4: 2594 MHz Graphics: Card: NVIDIA GF108GLM [Quadro 1000M] bus-ID: 01:00.0 Display Server: x11 (X.Org 1.19.5 ) drivers: nouveau (unloaded: modesetting,fbdev,vesa) Resolution: 1600x900@59.98hz OpenGL: renderer: llvmpipe (LLVM 5.0, 256 bits) version: 3.3 Mesa 17.2.6 Direct Render: Yes Audio: Card-1 Intel 6 Series/C200 Series Family High Def. Audio Controller driver: snd_hda_intel bus-ID: 00:1b.0 Card-2 NVIDIA GF108 High Def. Audio Controller driver: snd_hda_intel bus-ID: 01:00.1 Sound: Advanced Linux Sound Architecture v: k4.14.11-1-default Network: Card-1: Intel 82579LM Gigabit Network Connection (Lewisville) driver: e1000e v: 3.2.6-k port: 5020 bus-ID: 00:19.0 IF: enp0s25 state: down mac: <filter> Card-2: Intel Centrino Ultimate-N 6300 driver: iwlwifi bus-ID: 25:00.0 IF: wlo1 state: up mac: <filter> Drives: HDD Total Size: 2048.4GB (47.1% used) ID-1: /dev/sda model: Samsung_SSD_850 size: 2048.4GB Optical-1: /dev/sr0 model: PIOR BD-RW BDR-TD05AS rev: 1.01 dev-links: cdrom,cdrw,dvd,dvdrw Features: speed: 62x multisession: yes audio: yes dvd: yes rw: cd-r,cd-rw,dvd-r,dvd-ram state: running Partition: ID-1: / size: 94G used: 8.6G (10%) fs: ext4 dev: /dev/sda4 ID-2: /home size: 47G used: 430M (1%) fs: ext4 dev: /dev/sda5 ID-3: swap-1 size: 20.15GB used: 0.00GB (0%) fs: swap dev: /dev/sda8 Sensors: System Temperatures: cpu: 48.0C mobo: 43.0C gpu: 42.0 Fan Speeds (in rpm): cpu: N/A Info: Processes: 187 Uptime: 0:52 Memory: 1349.6/15987.9MB Init: systemd runlevel: 5 Gcc sys: N/A Client: Shell (bash 4.4.121) inxi: 2.3.40 ---------- System's suggestion to look at when the boot process throws the FAILED message: sudo systemctl status systemd-vconsole-setup.service [sudo] Passwort für root: ● systemd-vconsole-setup.service - Setup Virtual Console Loaded: loaded (/usr/lib/systemd/system/systemd-vconsole-setup.service; static; vendor preset Active: inactive (dead) since Mon 2018-01-15 09:32:18 CET; 26min ago Docs: man:systemd-vconsole-setup.service(8) man:vconsole.conf(5) Main PID: 399 (code=exited, status=0/SUCCESS) Jan 15 09:32:18 susytmblwdke systemd[1]: Started Setup Virtual Console. Nothing else there. ---------- This is my /usr/lib/systemd/system/plymouth-start.service: [Unit] Description=Show Plymouth Boot Screen DefaultDependencies=no Wants=systemd-ask-password-plymouth.path systemd-vconsole-setup.service After=systemd-vconsole-setup.service systemd-udev-trigger.service systemd-udevd.service Before=systemd-ask-password-plymouth.service ConditionKernelCommandLine=!plymouth.enable=0 [Service] ExecStart=/usr/sbin/plymouthd --mode=boot --pid-file=/run/plymouth/pid --attach-to-session ExecStartPost=-/usr/bin/plymouth show-splash Type=forking KillMode=none SendSIGKILL=no Removing the "systemd-vconsole-setup.service" from the "Wants=" list as per Comment #20 and rebuilding the initial ram disk with "sudo mkinitrd" doesn't help. ---------- This is the terminal output from mkinitrd: sudo mkinitrd [sudo] Passwort für root: Creating initrd: /boot/initrd-4.14.11-1-default dracut: Executing: /usr/bin/dracut --logfile /var/log/YaST2/mkinitrd.log --force /boot/initrd-4.14.11-1-default 4.14.11-1-default dracut: *** Including module: bash *** dracut: *** Including module: systemd *** dracut: *** Including module: warpclock *** dracut: *** Including module: systemd-initrd *** dracut: *** Including module: i18n *** dracut: Could not find FONT_MAP none! dracut: *** Including module: drm *** dracut: *** Including module: plymouth *** dracut: *** Including module: kernel-modules *** dracut: *** Including module: resume *** dracut: *** Including module: rootfs-block *** dracut: *** Including module: suse-xfs *** dracut: *** Including module: terminfo *** dracut: *** Including module: udev-rules *** dracut: Skipping udev rule: 40-redhat.rules dracut: Skipping udev rule: 50-firmware.rules dracut: Skipping udev rule: 50-udev.rules dracut: Skipping udev rule: 91-permissions.rules dracut: Skipping udev rule: 80-drivers-modprobe.rules dracut: *** Including module: dracut-systemd *** dracut: *** Including module: haveged *** dracut: *** Including module: usrmount *** dracut: *** Including module: base *** dracut: *** Including module: fs-lib *** dracut: *** Including module: shutdown *** dracut: *** Including module: suse *** dracut: *** Including modules done *** dracut: *** Installing kernel module dependencies and firmware *** dracut: *** Installing kernel module dependencies and firmware done *** dracut: *** Resolving executable dependencies *** dracut: *** Resolving executable dependencies done*** dracut: *** Hardlinking files *** dracut: *** Hardlinking files done *** dracut: *** Stripping files *** dracut: *** Stripping files done *** dracut: *** Generating early-microcode cpio image *** dracut: *** Constructing GenuineIntel.bin **** dracut: *** Store current command line parameters *** dracut: Stored kernel commandline: dracut: resume=UUID=d0e52c5d-112b-47dc-aa2c-7e1e31e38323 dracut: root=UUID=9bc866d3-2c3e-4934-8c8f-4e9adf94605b rootfstype=ext4 rootflags=rw,noatime,data=ordered dracut: *** Creating image file '/boot/initrd-4.14.11-1-default' *** dracut: *** Creating initramfs image file '/boot/initrd-4.14.11-1-default' done *** It throws the error line dracut: Could not find FONT_MAP none! ---------- In search for the reason of the failure during boot, I now looked at /etc/vconsole.conf: KEYMAP=de-nodeadkeys This is what I had chosen back in July at installation. Maybe there would be a better choice for a German system, please advise if so. Changing this to KEYMAP=de doesn't help. "localectl list-keymaps" shows that both de and de-nodeadkeys do exist. In the openSUSE forum thread https://forums.opensuse.org/showthread.php/526731-systemd-vconsole-setup-failed-(-Fonts-will-not-be-copied-to-remaining-consoles), someone mentioned having also FONT=latarcyrheb-sun16 in /etc/vconsole.conf. I tried this, but it didn't help, neither for the "dracut Could not find FONT_MAP none!" nor for the "Failed to start Setup Virtual Console" issue. ---------- Thus, I am currently down to the two questions: 1. What's the reason for "dracut: Could not find FONT_MAP none!" ? Is it related to the infamous original failure message? Does it matter? 2. What else could I try to get rid of the "Failed to start Setup Virtual Console" failure message during boot, when the fix in Comment #20 doesn't help? Any further help is gratefully appreciated.
I'm suspecting to have the same error situation at every boot since some months. (tumbleweed, kernel 4.14.13-3.gb5e51f3-default, Ryzen 7 1700X, Radeon 460). Please contact me, if it's another case or interesting otherwise. Excerpt from journalctl -b: Jan 15 14:37:56 linux-u4te systemd-vconsole-setup[270]: KD_FONT_OP_GET failed while trying to get the font metadata: Input/output error Jan 15 14:37:56 linux-u4te systemd-vconsole-setup[270]: Input/output error Jan 15 14:37:56 linux-u4te systemd-vconsole-setup[270]: Fonts will not be copied to remaining consoles Jan 15 14:37:56 linux-u4te systemd-vconsole-setup[270]: Input/output error Jan 15 14:37:56 linux-u4te systemd[1]: systemd-vconsole-setup.service: Main process exited, code=exited, status=1/FAILURE Jan 15 14:37:56 linux-u4te systemd-vconsole-setup[270]: Input/output error Jan 15 14:37:56 linux-u4te systemd[1]: Failed to start Setup Virtual Console. (As a minor problem with this hardware I haven't addressed it before).
In continuation of Comment #37 and answering to Comment #38 (Heinz Kohl): Yes, I figure it's the same issue here as in your case. This is a very short excerpt from sudo journalctl -b: Jan 16 11:37:45 susytmblwdke systemd-vconsole-setup[169]: /usr/bin/loadkeys failed with error code 1. Jan 16 11:37:45 susytmblwdke systemd-vconsole-setup[169]: KD_FONT_OP_GET failed while trying to get the font metadata: Input/output error Jan 16 11:37:45 susytmblwdke systemd-vconsole-setup[169]: Fonts will not be copied to remaining consoles Jan 16 11:37:45 susytmblwdke systemd[1]: systemd-vconsole-setup.service: Main process exited, code=exited, status=1/FAILURE Jan 16 11:37:45 susytmblwdke systemd[1]: Failed to start Setup Virtual Console. Jan 16 11:37:46 susytmblwdke systemd-vconsole-setup[169]: KDSKBENT: Input/output error: could not deallocate keymap 128 Jan 16 11:37:47 susytmblwdke systemd[1]: systemd-vconsole-setup.service: Unit entered failed state. Jan 16 11:37:47 susytmblwdke systemd[1]: systemd-vconsole-setup.service: Failed with result 'exit-code'. Jan 16 11:37:49 susytmblwdke.fritz.box display-manager[1392]: /etc/vconsole.conf available Jan 16 11:37:49 susytmblwdke.fritz.box display-manager[1392]: KEYMAP: de-nodeadkeys Jan 16 11:37:49 susytmblwdke.fritz.box display-manager[1392]: Command: localectl set-keymap de-nodeadkeys ---------- Here's another abridged output from sudo journalctl -b, a bit more than the previous one: -- Logs begin at Tue 2017-07-18 10:39:08 CEST, end at Tue 2018-01-16 11:43:19 CET. -- Jan 16 11:37:44 susytmblwdke kernel: microcode: microcode updated early to revision 0x29, date = 2013-06-12 Jan 16 11:37:44 susytmblwdke kernel: Linux version 4.14.12-1-default (geeko@buildhost) (gcc version 7.2.1 20171020 [gcc-7-branch revision 253932] (SUSE Linux)) #1 SMP PREEMPT Fri Jan 5 18:15:55 UTC 2018 (3cf399e) Jan 16 11:37:44 susytmblwdke kernel: Command line: BOOT_IMAGE=/boot/vmlinuz-4.14.12-1-default root=UUID=9bc866d3-2c3e-4934-8c8f-4e9adf94605b resume=/dev/disk/by-uuid/d0e52c5d-112b-47dc-aa2c-7e1e31e38323 splash=silent quiet showopts ... Jan 16 11:37:44 susytmblwdke dracut-cmdline[186]: dracut- dracut-044-10.2 dracut-044-10.2 Jan 16 11:37:44 susytmblwdke dracut-cmdline[186]: Using kernel command line parameters: resume=UUID=d0e52c5d-112b-47dc-aa2c-7e1e31e38323 root=UUID=9bc866d3-2c3e-4934-8c8f-4e9adf94605b rootfstype=ext4 rootflags=rw,noatime,data=ordered BOOT_IMAGE=/boot/vmlinuz-4.14.12-1-default root=UUID=9bc866d3-2c3e-4934-8c8f-4e9adf94605b resume=/dev/disk/by-uuid/d0e52c5d-112b-47dc-aa2c-7e1e31e38323 splash=silent quiet showopts Jan 16 11:37:44 susytmblwdke systemd-vconsole-setup[169]: Input/output error Jan 16 11:37:44 susytmblwdke systemd[1]: Started dracut cmdline hook. Jan 16 11:37:44 susytmblwdke systemd-vconsole-setup[169]: Input/output error Jan 16 11:37:44 susytmblwdke systemd[1]: Starting udev Kernel Device Manager... Jan 16 11:37:44 susytmblwdke systemd-vconsole-setup[169]: Input/output error Jan 16 11:37:44 susytmblwdke systemd[1]: Started udev Kernel Device Manager. Jan 16 11:37:44 susytmblwdke systemd-vconsole-setup[169]: Input/output error Jan 16 11:37:44 susytmblwdke systemd[1]: Starting dracut pre-trigger hook... Jan 16 11:37:45 susytmblwdke systemd-vconsole-setup[169]: Input/output error Jan 16 11:37:44 susytmblwdke systemd[1]: Started dracut pre-trigger hook. Jan 16 11:37:45 susytmblwdke systemd-vconsole-setup[169]: Input/output error Jan 16 11:37:44 susytmblwdke systemd[1]: Starting udev Coldplug all Devices... Jan 16 11:37:45 susytmblwdke systemd-vconsole-setup[169]: Input/output error Jan 16 11:37:45 susytmblwdke systemd[1]: Reached target Initrd File Systems. Jan 16 11:37:45 susytmblwdke systemd-vconsole-setup[169]: Input/output error ... Jan 16 11:37:45 susytmblwdke systemd-vconsole-setup[169]: Input/output error Jan 16 11:37:45 susytmblwdke systemd-vconsole-setup[169]: /usr/bin/loadkeys failed with error code 1. Jan 16 11:37:45 susytmblwdke systemd-vconsole-setup[169]: Input/output error Jan 16 11:37:45 susytmblwdke systemd-vconsole-setup[169]: KD_FONT_OP_GET failed while trying to get the font metadata: Input/output error Jan 16 11:37:45 susytmblwdke systemd-vconsole-setup[169]: Input/output error Jan 16 11:37:45 susytmblwdke systemd-vconsole-setup[169]: Fonts will not be copied to remaining consoles Jan 16 11:37:45 susytmblwdke systemd-vconsole-setup[169]: Input/output error Jan 16 11:37:45 susytmblwdke systemd[1]: systemd-vconsole-setup.service: Main process exited, code=exited, status=1/FAILURE Jan 16 11:37:45 susytmblwdke systemd-vconsole-setup[169]: Input/output error Jan 16 11:37:45 susytmblwdke systemd[1]: Failed to start Setup Virtual Console. ... Jan 16 11:37:46 susytmblwdke systemd-vconsole-setup[169]: Input/output error Jan 16 11:37:46 susytmblwdke systemd-vconsole-setup[169]: KDSKBENT: Input/output error: could not deallocate keymap 128 Jan 16 11:37:47 susytmblwdke systemd[1]: systemd-vconsole-setup.service: Unit entered failed state. Jan 16 11:37:47 susytmblwdke systemd[1]: systemd-vconsole-setup.service: Failed with result 'exit-code'. Jan 16 11:37:47 susytmblwdke systemd[1]: Starting Show Plymouth Boot Screen... ... Jan 16 11:37:47 susytmblwdke kernel: Console: switching to colour frame buffer device 200x56 Jan 16 11:37:47 susytmblwdke kernel: nouveau 0000:01:00.0: fb0: nouveaufb frame buffer device Jan 16 11:37:47 susytmblwdke kernel: [drm] Initialized nouveau 1.3.1 20120801 for 0000:01:00.0 on minor 0 ... Jan 16 11:37:47 susytmblwdke systemd[1]: Started Show Plymouth Boot Screen. Jan 16 11:37:47 susytmblwdke systemd[1]: Reached target Paths. Jan 16 11:37:47 susytmblwdke systemd[1]: Started Forward Password Requests to Plymouth Directory Watch. Jan 16 11:37:47 susytmblwdke systemd[1]: Reached target Basic System. Jan 16 11:37:47 susytmblwdke systemd[1]: Reached target Initrd Default Target. Jan 16 11:37:47 susytmblwdke systemd[1]: Starting Cleaning Up and Shutting Down Daemons... Jan 16 11:37:47 susytmblwdke systemd[1]: Stopped target Initrd Default Target. Jan 16 11:37:47 susytmblwdke systemd[1]: Starting Setup Virtual Console... ... Jan 16 11:37:47 susytmblwdke systemd[1]: Started Setup Virtual Console. ... Jan 16 11:37:47 susytmblwdke systemd[1]: Started Setup Virtual Console. ... Jan 16 11:37:49 susytmblwdke.fritz.box display-manager[1392]: /etc/vconsole.conf available Jan 16 11:37:49 susytmblwdke.fritz.box display-manager[1392]: KEYMAP: de-nodeadkeys Jan 16 11:37:49 susytmblwdke.fritz.box display-manager[1392]: Command: localectl set-keymap de-nodeadkeys ... ---------- No remedy around yet, it seems. Yes, I agree, it does not seem to harm the Tumbleweed system, but it's annoying nevertheless. So I/we all would like to get this issue solved. Please ... Please let me remind you of the dracut error "dracut: Could not find FONT_MAP none!" during mkinitrd. Any sense to be made of this?
Made the change recommended in #20 on Tumbleweed 20180110. It made the message disappear.
Finally (sound of a sigh)... Again made the change as per #20, today on Tumbleweed 20180110. It makes the „Failure“ message go away. I still had the previous Tmblwd version yesterday, but that fix didn‘t work then. I keep my fingers crossed - hope the next versions of Tmblwd won‘t break things again. Thanks, Freek de Kruijf, for making me try again today. And thanks to Frank Bui for #20!
Ludger- Are you saying that the steps in #20 completely resolves the issues or just the error message? Please confirm? Thanks
(In reply to Indy chery from comment #42) > Ludger- Are you saying that the steps in #20 completely resolves the issues > or just the error message? Please confirm? Thanks I'm for sure not knowledgeable enough to answer this question. What I can say is this: Before putting the fix in #20 into place, (A) quite some folks got the obnoxious message "Failed to start Setup Virtual Console" on the screen during boot, and (B) my "sudo journalctl -b" log was literally flooded with error messages "systemd-vconsole-setup[nnn]: Input/output error" as appended to my comment #39. With the #20 fix in place now, I don't get the a.m. messages (A) and (B) any longer. In particular, my journalctl now has none of those vconsole-setup errors anymore. I would like to leave an interpretation of these observations to the more knowledgeable folks.
After recent upgrades the error message reappeared, at least now in 20180127. It definitely has something to do with being run too early in the boot process. After finishing the boot process, the command to start this service completes successfully. Applied the suggestion in comment #20 again. If there is no further comment, applying that suggestion fixed the problem.
Same here after Update to 20180206 (or 07? Not at my PC now), having skipped a round or two. Applying Comment #20 fixed it again. Interesting quirk: Using YaST -> Bootloader yielded a new initrd that still had the bug. Only sudo mkinitrd worked. I didn‘t dig deeper. Hope someone will fix the bug discussed in this thread for good someday.
Also had to follow comment #20 with 20180222.
Same failure on Leap 15.0 beta.
(In reply to Freek de Kruijf from comment #47) > Same failure on Leap 15.0 beta. Applying suggestion in Comment 20 makes the error message to disappear.
Just wanted to confirm: I removed plymouth* and installed the latest snapshot (so getting new kernel/initrd). Problem is gone. So I guess it has something to do with plymouth. Anyways this is a more long term fix for people who don't care about the splashscreen.
Is there anybody that can reproduce this reliably and have some time to debug this ? I can't reproduce here so I need some help. Thanks.
(In reply to Franck Bui from comment #50) > Is there anybody that can reproduce this reliably and have some time to > debug this ? > > I can't reproduce here so I need some help. > > Thanks. In my view this problem occurs because it is executed too early in the boot process. Probably a file is not yet available at that time.
(In reply to Franck Bui from comment #50) > Is there anybody that can reproduce this reliably and have some time to > debug this ? > > I can't reproduce here so I need some help. > > Thanks. Maybe I can help. I have dual boot with working and non-working release.
(In reply to Seppo Lehtikangas from comment #52) > (In reply to Franck Bui from comment #50) > > Is there anybody that can reproduce this reliably and have some time to > > debug this ? > > > > I can't reproduce here so I need some help. > > > > Thanks. > > Maybe I can help. I have dual boot with working and non-working release. Maybe I can help too. I have two Leap 15.0 Beta systems. On one the patch of Comment #20 has been applied, which does nor show the problem. On the other one the patch has not been applied and shows the error message. Performing "journalctl -b -u systemd-vconsole-setup.service" shows for about 3 seconds a number of error messages like: Mar 28 22:00:13 eik150t systemd-vconsole-setup[148]: Input/output error However at the end I have: Mar 28 22:00:14 eik150t systemd-vconsole-setup[148]: KDSKBENT: Input/output error: could not deallocate keymap 128 Mar 28 22:00:16 eik150t systemd[1]: Starting Setup Virtual Console... Mar 28 22:00:16 eik150t systemd[1]: Started Setup Virtual Console. Mar 28 22:00:21 eik150t systemd[1]: Started Setup Virtual Console. Mar 28 22:00:26 eik150t systemd[1]: Starting Setup Virtual Console... Mar 28 22:00:26 eik150t systemd-vconsole-setup[880]: KD_FONT_OP_SET failed, fonts will not be copied to tty7: Illegal argument Mar 28 22:00:26 eik150t systemd[1]: Started Setup Virtual Console.
The same here in Leap 15 Beta. The suggestion in #20 didn't work. systemctl status systemd-vconsole-setup.service ● systemd-vconsole-setup.service - Setup Virtual Console Loaded: loaded (/usr/lib/systemd/system/systemd-vconsole-setup.service; static; vendor preset: disabled) Active: inactive (dead) since Mon 2018-04-09 17:24:19 -03; 5min ago Docs: man:systemd-vconsole-setup.service(8) man:vconsole.conf(5) Main PID: 453 (code=exited, status=0/SUCCESS) Warning: Journal has been rotated since unit was started. Log output is incomplete or unavailable.
Jyi, from the tumbleweed update providing kernel 4.16.0 I don't see this issue any more. But It may also related to a problem I tried to solve within the vconsole settings were I switch my keyboard to German deadkeys and used localectl for doing so. May be this has been correct something else in the vconsole configuration with leads to a proper vconsole service startup. If someone of the commentators who has still the issue can create a snapshot and try to change the keyboard using localectl may give an idea if this has any impact. Regards Alex
*** Bug 1089515 has been marked as a duplicate of this bug. ***
I have 2 systems with the packages almost the same versions, but one of the system has such issue, the other one doesn't There are 2 main difference between systems: 1. Systems having the issue is installed on HDD, while the other on SSD and seems loading twice longer. 2. System having the issue was installed day ago, but all repos are the same. The system have the issue from the first start. I've attached 2 SVGs from systemd-analize, probably the order of loading the service make sense.
Created attachment 768237 [details] systemd-analize plot output from two systems
Thank you Franck Bui via Comment #20 - your fix worked perfectly; and all journal fail output is moreover gone now! Your fix also works on Beta Leap 15.0.
Also found in openSUSE Leap 15.0 I tried the recommendation of Comment #20 and didn't worked.
Likewise, after the 42.3->15.0 upgrade. Not there on a fresh install though.
A user reported that it happened to him on a fresh install of leap. solution proposed in #20 worked aswell.
(In reply to Santiago Zarate from comment #62) > A user reported that it happened to him on a fresh install of leap. solution > proposed in #20 worked aswell. I never saw this bug solved in Tumbleweed and Leap 15.0. Although there are reports that the workaround in #20 did not work, I still think that workaround should be implemented in a bugfix.
The error messages for me start with about 30 lines saying: systemd-vconsole-setup[175]: Input/output error But it is not clear to me what vconsole is trying to access, with that information debugging would be easier. Any idea on how to get more information out of this?
Created attachment 776834 [details] Boot log with systemd.log_level=debug
Did today boot with "systemd.log_level=debug" added to the boot line and see the first systemd-vconsole-setup[180] is running fine on /dev/tty1, then it runs into errors and after that other systemd-vconsole-setup instances are started (414/419/499/537/652/703) and these are all fine. See the "Boot log with systemd.log_level=debug" above. Still, we debug enabled no clue on what is triggering the Input/output error, could also not find it in https://github.com/systemd/systemd/blob/master/src/vconsole/vconsole-setup.c
*** Bug 1101245 has been marked as a duplicate of this bug. ***
Since I could reproduce this easily, I tried debugging a bit. Luckily, I could spot out the error at initrd now, while tracking another bug in bsc#1101591. The problem seems coming from dracut-cmdline-ask.service in /usr/lib/dracut/modules.d/98dracut-systemd, which invokes dracut-cmdline-ask.sh. This script and systemd-vconsole-setup are started in parallel for the same console (tty1). Then dracut-cmdline-ask quits immediately without doing anything (unless rd.cmdline=ask is given). As this is a bash script and it gets tty as stdin as specified in its *.service, this triggers the hangup of tty1 at its exit, as it seems. Meanwhile systemd-vconsole-setup continues and tries some ioctls after that, but they fail because of the hung up tty1. As a workaround, edit /usr/lib/dracut/modules.d/98dracut-systemd/module-setup.sh, and remove two lines containing "dracut-cmdline-ask" (one for script and one for service), then rebuild initrd via "dracut --force --local". Maybe it's not a fault of dracut itself, but the tty session has to be handled more carefully.
(In reply to Takashi Iwai from comment #68) > The problem seems coming from dracut-cmdline-ask.service in > /usr/lib/dracut/modules.d/98dracut-systemd, which invokes > dracut-cmdline-ask.sh. > This script and systemd-vconsole-setup are started in parallel for the same > console (tty1). > > Then dracut-cmdline-ask quits immediately without doing anything (unless > rd.cmdline=ask is given). As this is a bash script and it gets tty as stdin > as specified in its *.service, this triggers the hangup of tty1 at its exit, > as it seems. > > Meanwhile systemd-vconsole-setup continues and tries some ioctls after that, > but they fail because of the hung up tty1. Thanks Takashi for bringing more information on this issue. > > As a workaround, edit > /usr/lib/dracut/modules.d/98dracut-systemd/module-setup.sh, and remove two > lines containing "dracut-cmdline-ask" (one for script and one for service), > then rebuild initrd via "dracut --force --local". > Or make dracut-cmdline-ask.service run after systemd-vconsole-setup.service, this should do the trick too. > Maybe it's not a fault of dracut itself, but the tty session has to be > handled more carefully. Maybe systemd-vconsole-setup should be more careful and try to operate on a tty which is not already acquired by another process ? I'll investigate. Thanks !
After update on opensuse leap 15.0 I got same problem ... I made the changes that Franck Bui told us in recently comment but the same error show in bootup
*** Bug 1106331 has been marked as a duplicate of this bug. ***
I just had this exact same problem when installing the current Tumbleweed Snapshot-20180828 and happened to stumble on a solution by sheer coincidence. The issue has something to do with English language system keyboard layouts. The default keyboard layout for an installation happens to be English (US) which I had forgotten to change in the first screen where the user is prompted to accept the licensing agreement. Upon first booting I was immediately greeted by the the "Failed to start Setup Virtual Console" error message discussed in this tread. The good news is that I was able to hit Ctrl-D to bypass the error. In Yast I then changed the system keyboard configuration to German to match my keyboard. The error no longer appeared after rebooting. The bad news is that this issue seems to affect all English language keyboard layouts: UK/UK and US International. I hope this helps.
(In reply to John Mullen from comment #72) > I just had this exact same problem when installing the current Tumbleweed > Snapshot-20180828 and happened to stumble on a solution by sheer coincidence. > > The issue has something to do with English language system keyboard layouts. > > The default keyboard layout for an installation happens to be English (US) > which I had forgotten to change in the first screen where the user is > prompted to accept the licensing agreement. Upon first booting I was > immediately greeted by the the "Failed to start Setup Virtual Console" error > message discussed in this tread. The good news is that I was able to hit > Ctrl-D to bypass the error. In Yast I then changed the system keyboard > configuration to German to match my keyboard. The error no longer appeared > after rebooting. The bad news is that this issue seems to affect all English > language keyboard layouts: UK/UK and US International. > > I hope this helps. Hi John, yes it certainly does help! I switched in YAST from UK to French keyboard layout and back to UK English and the error no longer appears. Thanks! (This was in LEAP 15, and for what it's worth the boot was from an SSD. Maybe a timing clash?
I was using the Italian layout, switched to US International and the problem disappeared. Could the rule be: "switch from the layout you used during the installation at least once"?
I am on Tubleweed and switched from English (US) to Dutch and back again. Rebooted but still the same problem. systemd-vconsole-setup[179]: /usr/bin/loadkeys failed with exit status 1 What is bad in my opinion is that the system says "Input/output error" while it should have indicated which file it wants to access, that way finding what is wrong would be faster/better doable.
Right. The problem is back for me as well, occasionally.
(In reply to Franck Bui from comment #69) > > Or make dracut-cmdline-ask.service run after systemd-vconsole-setup.service, > this should do the trick too. I confirm that this fixed it for me on new clean Leap 15 install in QEMU KVM VM (XFCE pattern).
@Andrei Borzenkov shouldn't that be fixed with a maintenance request then? Is what causes the problem clear at this point? Or is it the solution that is unclear?
(In reply to Franck Bui from comment #69) > > Or make dracut-cmdline-ask.service run after systemd-vconsole-setup.service Note that 90-vconsole.rules runs systemd-vconsole-setup directly. I am not sure why it does not call systemd-vconsole-setup.service via SYSTEMD_WANTS instead. It probably should, otherwise there is no way to order this call against anything else. Moreover, given that systemd-vconsole-setup is actually prerequisite for any interactive service (shells, anything that queries passwords etc) the obvious question is - how are we supposed to order these components against systemd-vconsole-setup in the first place? It *must* complete before anything else, but it is started asynchronously from udev rule leaving no way to directly add dependencies on it.
I really wonder why in /usr/lib/udev/rules.d/90-vconsole.rules does not use RUN+="/usr/bin/systemctl restart systemd-vconsole-setup.service" instead the systemd-vconsole-setup binary its self. (IMHO) The systemctl approach would at least fulfill the dependencies like those given in kbdsettings.service
(In reply to Andrei Borzenkov from comment #79) > I am not sure why it does not call systemd-vconsole-setup.service via > SYSTEMD_WANTS instead. Perhaps it's been assumed that all services needing console accesses would run "long enough" after udev has finished to init the virtual consoles ? I can see only one (minor) advantage to do that: it removes the need for services that requires console accesses to explicitly add dependencies on systemd-virtual-console.service. But again this only works for services that run "long enough" after udev, so it's pretty fragile. To remove the need to explicitly track dependencies on systemd-vconsole-setup.service we could have added implicit default dependencies to units that uses Standard*=tty* but that wouldn't have been very helpful I guess as most units that define Standard*=tty* also set DefautDependencies=no.
But that said it seems to me that dracut-cmdline-ask.service is a *special* case as it seems that this service wants to be run even before udev is. I don't know if it really needs that (actully I'm not sure to see the service's use-fullness) but if so it seems that it should need to be more careful with its dependencies and explicitly use "{Requires,After}=systemd-vconsole-setup.service". Daniel, any idea ?
(In reply to Franck Bui from comment #81) > To remove the need to explicitly track dependencies on > systemd-vconsole-setup.service we could have added implicit default > dependencies to units that uses Standard*=tty* but that wouldn't have > been very helpful I guess as most units that define Standard*=tty* > also set DefautDependencies=no. But for services that doesn't set DefautDependencies=no, SYSTEMD_WANTS would be useful as Werner pointed out. I'm wondering if we couldn't even make sysinit.target scheduled after systemd-vconsole-setup.service so there won't be any need for those services to add any explicit constraints...
(In reply to Franck Bui from comment #82) > But that said it seems to me that dracut-cmdline-ask.service is a > *special* case as it seems that this service wants to be run even before > udev is. Yes, the idea is arguments can be passed even for the udev cmdline. Hence this has to be run really early on. This seems to be for really quirky cases where you want to pass arguments to dracut that should not show up on the kernel commandline. I'd need to ask Harald about the motivation. > > I don't know if it really needs that (actully I'm not sure to see the > service's use-fullness) but if so it seems that it should need to be > more careful with its dependencies and explicitly use > "{Requires,After}=systemd-vconsole-setup.service". As systemd-vconsole-setup.service has no dependencies on its own, I would agree.
(In reply to Daniel Molkentin from comment #84) > (In reply to Franck Bui from comment #82) > > I don't know if it really needs that (actully I'm not sure to see the > > service's use-fullness) but if so it seems that it should need to be > > more careful with its dependencies and explicitly use > > "{Requires,After}=systemd-vconsole-setup.service". > > As systemd-vconsole-setup.service has no dependencies on its own, I would > agree. Can you please take care of this ?
Regarding the issue about systemd-vconsole-setup binary which is spawned via an udev rule making the ordering against the relevant service difficult, I'll create a new upstream bug as it should be treated independently I think.
(In reply to Daniel Molkentin from comment #84) > As systemd-vconsole-setup.service has no dependencies on its own, I would > agree. Question: doesn't this service needs a file, so it depends on the availability of a file system.
(In reply to Freek de Kruijf from comment #87) > (In reply to Daniel Molkentin from comment #84) > > As systemd-vconsole-setup.service has no dependencies on its own, I would > > agree. > > Question: doesn't this service needs a file, so it depends on the > availability of a file system. Only initramfs, which should be present at this point.
Created attachment 781830 [details] proposed patch Please give this patch a try (in /usr/lib/dracut): patch -p1 < systemd-vconsole-setup-race-fix.patch Luigi, can you give it a test?
I added #781830 patch to the dracut package and booted/rebooted 10 times. No sign of the dreaded vconsole alert, but on two occasions instead of SDDM I saw only a pointer on a black screen. Presumably an unrelated glitch.
(In reply to Daniel Molkentin from comment #84) > (In reply to Franck Bui from comment #82) > > But that said it seems to me that dracut-cmdline-ask.service is a > > *special* case as it seems that this service wants to be run even before > > udev is. > > Yes, the idea is arguments can be passed even for the udev cmdline. BTW currently there's nothing that would prevent udev from starting before dracut-cmdline-ask.service and that leads to this issue: basically udev and dracut-cmdline-ask.service are started in //, which results in both systemd-vconsole-setup (binary) and dracut-cmdline-ask.service accessing to the console.
Is it safe to add Before=systemd-udevd.service?
(In reply to Daniel Molkentin from comment #92) > Is it safe to add Before=systemd-udevd.service? Since sysinit.target depends on systemd-udevd.service, I would say so.
(In reply to Franck Bui from comment #91) > (In reply to Daniel Molkentin from comment #84) > > (In reply to Franck Bui from comment #82) > > > But that said it seems to me that dracut-cmdline-ask.service is a > > > *special* case as it seems that this service wants to be run even before > > > udev is. > > > > Yes, the idea is arguments can be passed even for the udev cmdline. > > BTW currently there's nothing that would prevent udev from starting before > dracut-cmdline-ask.service and that leads to this issue: basically udev and > dracut-cmdline-ask.service are started in //, which results in both > systemd-vconsole-setup (binary) and dracut-cmdline-ask.service accessing to > the console. Not quiet true it seems dracut-cmdline-ask.service has Before=dracut-cmdline.service which has Before=dracut-pre-udev.service which has Before=systemd-udevd.service How does systemd-vconsole-setup.service ensure it's started after udev? Before=initrd-switch-root.target seems too relaxed.
(In reply to Daniel Molkentin from comment #94) > Not quiet true it seems Indeed that's not needed in your case. I was wrong, this issue is not due to dracut-cmdline-ask.service racing with udev (and indirectly with systemd-vconsole-setup) but is due to plymouth-start.service pulling systemd-vconsole-setup.service in (even if plymouth.enable=0 BTW). Since systemd-vconsole-setup.service has basically no ordering constraints, it can be run at mostly any time and in this case it's run very early, racing with dracut-cmdline-ask.service. > How does systemd-vconsole-setup.service ensure it's started after udev? There's no such enforcement as the console can be setup and used even before udev is run, I guess. At least it allows services, which needs to be run very early such as dracut-cmdline-ask.service to request the initialization of the console (if that's possible).
I've submitted the patch upstream. Once accepted, I'll backport.
(In reply to Franck Bui from comment #86) > I'll create a new upstream bug as it should be treated independently I think. FTR it's there: https://github.com/systemd/systemd/issues/10019
> Please give this patch a try (in /usr/lib/dracut): > > patch -p1 < systemd-vconsole-setup-race-fix.patch I tried the patch on my Tumbleweed system but no difference, still many 'systemd-vconsole-setup[178]: Input/output error'-s and 'systemd-vconsole-setup[178]: /usr/bin/loadkeys failed with exit status 1'
(In reply to M Fredericks from comment #98) > > Please give this patch a try (in /usr/lib/dracut): > > > > patch -p1 < systemd-vconsole-setup-race-fix.patch > > I tried the patch on my Tumbleweed system but no difference, still many > 'systemd-vconsole-setup[178]: Input/output error'-s and > 'systemd-vconsole-setup[178]: /usr/bin/loadkeys failed with exit status 1' Did you recreate your initramfs (by running "mkinitrd") ?
(In reply to Franck Bui from comment #99) > Did you recreate your initramfs (by running "mkinitrd") ? No I did not, thanks for the feedback. After running mkinitrd, the problem is gone, so the patch is working, at least for me, thanks! After the patch and mkinitrd the bootlog only messages on vconsole are: Sep 06 22:21:00 systemd-vconsole-setup[313]: KD_FONT_OP_GET failed while trying to get the font metadata: Function not implemented Sep 06 22:21:00 systemd-vconsole-setup[313]: Fonts will not be copied to remaining consoles Running mkinitrd I noticed a message that might be related to the problem (Could not find FONT_MAP none!): Creating initrd: /boot/initrd-4.18.0-1-default dracut: Executing: /usr/bin/dracut --logfile /var/log/YaST2/mkinitrd.log --force /boot/initrd-4.18.0-1-default 4.18.0-1-default dracut: *** Including module: bash *** dracut: *** Including module: systemd *** dracut: *** Including module: warpclock *** dracut: *** Including module: systemd-initrd *** dracut: *** Including module: i18n *** dracut: Could not find FONT_MAP none! dracut: *** Including module: drm *** dracut: *** Including module: plymouth *** dracut: *** Including module: btrfs *** dracut: *** Including module: kernel-modules *** dracut: *** Including module: resume *** dracut: *** Including module: rootfs-block *** dracut: *** Including module: suse-xfs *** dracut: *** Including module: terminfo *** dracut: *** Including module: udev-rules *** ...
(In reply to M Fredericks from comment #100) > Running mkinitrd I noticed a message that might be related to the problem > (Could not find FONT_MAP none!): https://bugzilla.opensuse.org/show_bug.cgi?id=1013573
With the patched /usr/lib/dracut/modules.d/98dracut-systemd/dracut-cmdline-ask.service, after mkinitrd was executed without error: 1. On a UEFI single disk Leap 15.0 Laptop, Btrfs system partition, AMD hardware, no systemd-vconsole-setup errors. 2. On a non-UEFI multiple disk Leap 15.0 Desktop: * /tmp/ and /var/ not in system partition -- in separate ext4 partitions on HDDs; * ext4 system partition on an SSD; * AMD hardware: ** "Setup Virtual Console" boot error message on tty1. ** "systemd-vconsole-setup" boot error messages in the systemd journal. Used the following to pull out the Journal's relevant boot time message entries: journalctl --this-boot | grep -iE ' target | Started | Starting | Mounting | Mounted |vconsole'
OpenSUSE 15 up to date (LTS, NO Tumbleweed), Spanish configuration Exactly the same error "Failed to start setup virstual console". systemctl --full status systemd-vconsole-setup ● systemd-vconsole-setup.service - Setup Virtual Console Loaded: loaded (/usr/lib/systemd/system/systemd-vconsole-setup.service; static; vendor preset: disabled) Active: inactive (dead) since Tue 2018-10-02 09:34:17 CEST; 24h ago Docs: man:systemd-vconsole-setup.service(8) man:vconsole.conf(5) Main PID: 790 (code=exited, status=0/SUCCESS) Any news about it?
*** Bug 1107902 has been marked as a duplicate of this bug. ***
*** Bug 1106848 has been marked as a duplicate of this bug. ***
(In reply to Rafael linux from comment #103) > OpenSUSE 15 up to date (LTS, NO Tumbleweed), Spanish configuration > > Exactly the same error "Failed to start setup virstual console". Did you try exactly the same patch as which worked for Tumbleweed? See above for details.
(In reply to M Fredericks from comment #106) > (In reply to Rafael linux from comment #103) > > OpenSUSE 15 up to date (LTS, NO Tumbleweed), Spanish configuration > > > > Exactly the same error "Failed to start setup virstual console". > > Did you try exactly the same patch as which worked for Tumbleweed? See above > for details. Is there a coming update to solve this bug?
(In reply to M Fredericks from comment #106) > (In reply to Rafael linux from comment #103) > > OpenSUSE 15 up to date (LTS, NO Tumbleweed), Spanish configuration > > > > Exactly the same error "Failed to start setup virstual console". > > Did you try exactly the same patch as which worked for Tumbleweed? See above > for details. That rebuild seems to remove the error. I can't see it in dmesg output (thru Ksyslog). So that patch worked for me, thank you. So I understand is some fail in installation, it is not?
Well, contrary to what I said, the issue is no solved after making "mkinitrd". I just saw the message again AFTER BOOTING FROM HIBERNATION :(
systemd-vconsole-setup.service is forked by systemd-udevd.service. It fails on the first fork, but succeeds on the subsequent ones: -- Logs begin at Sun 2018-08-26 08:48:18 CEST, end at Thu 2018-10-25 07:15:05 CEST. -- Oct 25 07:13:51 erlangen systemd[209]: systemd-vconsole-setup.service: Executing: /usr/lib/systemd/systemd-vconsole-setup Oct 25 07:13:51 erlangen systemd-vconsole-setup[209]: Sysfs UTF-8 flag enabled Oct 25 07:13:51 erlangen systemd-vconsole-setup[209]: UTF-8 kbdmode enabled on /dev/tty1 Oct 25 07:13:51 erlangen systemd-vconsole-setup[209]: Executing "/usr/bin/setfont -C /dev/tty1 -m none eurlatgr.psfu"... Oct 25 07:13:51 erlangen systemd-vconsole-setup[209]: Successfully forked off '(setfont)' as PID 213. Oct 25 07:13:51 erlangen systemd-vconsole-setup[209]: /usr/bin/setfont succeeded. Oct 25 07:13:51 erlangen systemd-vconsole-setup[209]: Executing "/usr/bin/loadkeys -q -C /dev/tty1 -u de-nodeadkeys"... Oct 25 07:13:51 erlangen systemd-vconsole-setup[209]: Successfully forked off '(loadkeys)' as PID 217. Oct 25 07:13:52 erlangen systemd-vconsole-setup[209]: Input/output error ... repeated some 100+ times Oct 25 07:13:52 erlangen systemd-vconsole-setup[209]: Input/output error Oct 25 07:13:52 erlangen systemd-vconsole-setup[209]: KDSKBENT: Input/output error: could not deallocate keymap 128 Oct 25 07:13:52 erlangen systemd-vconsole-setup[209]: /usr/bin/loadkeys failed with exit status 1. Oct 25 07:13:52 erlangen systemd-vconsole-setup[209]: KD_FONT_OP_GET failed while trying to get the font metadata: Input/output error Oct 25 07:13:52 erlangen systemd-vconsole-setup[209]: Fonts will not be copied to remaining consoles Oct 25 07:13:52 erlangen systemd[1]: systemd-vconsole-setup.service: Child 209 belongs to systemd-vconsole-setup.service. Oct 25 07:13:52 erlangen systemd[1]: systemd-vconsole-setup.service: Main process exited, code=exited, status=1/FAILURE Oct 25 07:13:52 erlangen systemd[1]: systemd-vconsole-setup.service: Failed with result 'exit-code'. Oct 25 07:13:52 erlangen systemd[1]: systemd-vconsole-setup.service: Changed start -> failed Oct 25 07:13:52 erlangen systemd[1]: systemd-vconsole-setup.service: Job systemd-vconsole-setup.service/start finished, result=failed Oct 25 07:13:52 erlangen systemd[1]: Failed to start Setup Virtual Console. Oct 25 07:13:52 erlangen systemd[1]: systemd-vconsole-setup.service: Unit entered failed state. Oct 25 07:13:52 erlangen systemd[1]: systemd-vconsole-setup.service: Installed new job systemd-vconsole-setup.service/start as 82 Oct 25 07:13:52 erlangen systemd[1]: systemd-vconsole-setup.service: ConditionPathExists=/dev/tty0 succeeded. Oct 25 07:13:52 erlangen systemd[1]: systemd-vconsole-setup.service: Passing 0 fds to service Oct 25 07:13:52 erlangen systemd[1]: systemd-vconsole-setup.service: About to execute: /usr/lib/systemd/systemd-vconsole-setup Oct 25 07:13:52 erlangen systemd[1]: systemd-vconsole-setup.service: Forked /usr/lib/systemd/systemd-vconsole-setup as 371 Oct 25 07:13:52 erlangen systemd[1]: systemd-vconsole-setup.service: Changed failed -> start Oct 25 07:13:52 erlangen systemd[1]: Starting Setup Virtual Console... Oct 25 07:13:52 erlangen systemd[371]: systemd-vconsole-setup.service: Executing: /usr/lib/systemd/systemd-vconsole-setup Oct 25 07:13:52 erlangen systemd-vconsole-setup[371]: Sysfs UTF-8 flag enabled Oct 25 07:13:52 erlangen systemd-vconsole-setup[371]: UTF-8 kbdmode enabled on /dev/tty1 Oct 25 07:13:52 erlangen systemd-vconsole-setup[371]: Executing "/usr/bin/setfont -C /dev/tty1 -m none eurlatgr.psfu"... Oct 25 07:13:52 erlangen systemd-vconsole-setup[371]: Successfully forked off '(setfont)' as PID 372. Oct 25 07:13:52 erlangen systemd-vconsole-setup[371]: /usr/bin/setfont succeeded. Oct 25 07:13:52 erlangen systemd-vconsole-setup[371]: Executing "/usr/bin/loadkeys -q -C /dev/tty1 -u de-nodeadkeys"... Oct 25 07:13:52 erlangen systemd-vconsole-setup[371]: Successfully forked off '(loadkeys)' as PID 375. Oct 25 07:13:52 erlangen systemd-vconsole-setup[371]: /usr/bin/loadkeys succeeded. Oct 25 07:13:52 erlangen systemd-vconsole-setup[371]: KD_FONT_OP_GET failed while trying to get the font metadata: Function not implemented Oct 25 07:13:52 erlangen systemd-vconsole-setup[371]: Fonts will not be copied to remaining consoles Oct 25 07:13:52 erlangen systemd[1]: systemd-vconsole-setup.service: Child 371 belongs to systemd-vconsole-setup.service. Oct 25 07:13:52 erlangen systemd[1]: systemd-vconsole-setup.service: Main process exited, code=exited, status=0/SUCCESS Oct 25 07:13:52 erlangen systemd[1]: systemd-vconsole-setup.service: Changed start -> dead Oct 25 07:13:52 erlangen systemd[1]: systemd-vconsole-setup.service: Job systemd-vconsole-setup.service/start finished, result=done Oct 25 07:13:52 erlangen systemd[1]: Started Setup Virtual Console. Oct 25 07:13:52 erlangen systemd[1]: systemd-vconsole-setup.service: Failed to send unit remove signal for systemd-vconsole-setup.service: Transport endpoint is not connected Oct 25 07:13:53 erlangen systemd[1]: systemd-vconsole-setup.service: Installed new job systemd-vconsole-setup.service/start as 143 Oct 25 07:13:54 erlangen systemd[416]: systemd-vconsole-setup.service: Executing: /usr/lib/systemd/systemd-vconsole-setup Oct 25 07:13:54 erlangen systemd-vconsole-setup[416]: Sysfs UTF-8 flag enabled Oct 25 07:13:54 erlangen systemd-vconsole-setup[416]: UTF-8 kbdmode enabled on /dev/tty1 Oct 25 07:13:54 erlangen systemd-vconsole-setup[416]: Executing "/usr/bin/setfont -C /dev/tty1 -m none eurlatgr.psfu"... Oct 25 07:13:54 erlangen systemd-vconsole-setup[416]: Successfully forked off '(setfont)' as PID 418. Oct 25 07:13:54 erlangen systemd-vconsole-setup[416]: /usr/bin/setfont succeeded. Oct 25 07:13:54 erlangen systemd-vconsole-setup[416]: Executing "/usr/bin/loadkeys -q -C /dev/tty1 -u de-nodeadkeys"... Oct 25 07:13:54 erlangen systemd-vconsole-setup[416]: Successfully forked off '(loadkeys)' as PID 437. Oct 25 07:13:54 erlangen systemd-vconsole-setup[416]: /usr/bin/loadkeys succeeded. Oct 25 07:13:54 erlangen systemd[1]: systemd-vconsole-setup.service: Child 416 belongs to systemd-vconsole-setup.service. Oct 25 07:13:54 erlangen systemd[1]: systemd-vconsole-setup.service: Main process exited, code=exited, status=0/SUCCESS Oct 25 07:13:54 erlangen systemd[1]: systemd-vconsole-setup.service: Changed start -> dead Oct 25 07:13:54 erlangen systemd[1]: systemd-vconsole-setup.service: Job systemd-vconsole-setup.service/start finished, result=done Oct 25 07:13:54 erlangen systemd[1]: Started Setup Virtual Console. Oct 25 07:13:54 erlangen systemd[1]: systemd-vconsole-setup.service: Failed to send unit change signal for systemd-vconsole-setup.service: Transport endpoint is not connected Oct 25 07:13:55 erlangen systemd[1]: systemd-vconsole-setup.service: Installed new job systemd-vconsole-setup.service/start as 319 Oct 25 07:13:55 erlangen systemd[1]: systemd-vconsole-setup.service: ConditionPathExists=/dev/tty0 succeeded. Oct 25 07:13:55 erlangen systemd[1]: systemd-vconsole-setup.service: Passing 0 fds to service Oct 25 07:13:55 erlangen systemd[1]: systemd-vconsole-setup.service: About to execute: /usr/lib/systemd/systemd-vconsole-setup Oct 25 07:13:55 erlangen systemd[1]: systemd-vconsole-setup.service: Forked /usr/lib/systemd/systemd-vconsole-setup as 709 Oct 25 07:13:55 erlangen systemd[1]: systemd-vconsole-setup.service: Changed dead -> start Oct 25 07:13:55 erlangen systemd[1]: Starting Setup Virtual Console... Oct 25 07:13:55 erlangen systemd[709]: systemd-vconsole-setup.service: Executing: /usr/lib/systemd/systemd-vconsole-setup Oct 25 07:13:55 erlangen systemd-vconsole-setup[709]: Sysfs UTF-8 flag enabled Oct 25 07:13:55 erlangen systemd-vconsole-setup[709]: UTF-8 kbdmode enabled on /dev/tty1 Oct 25 07:13:55 erlangen systemd-vconsole-setup[709]: Executing "/usr/bin/setfont -C /dev/tty1 -m none eurlatgr.psfu"... Oct 25 07:13:55 erlangen systemd-vconsole-setup[709]: Successfully forked off '(setfont)' as PID 713. Oct 25 07:13:55 erlangen systemd-vconsole-setup[709]: /usr/bin/setfont succeeded. Oct 25 07:13:55 erlangen systemd-vconsole-setup[709]: Executing "/usr/bin/loadkeys -q -C /dev/tty1 -u de-nodeadkeys"... Oct 25 07:13:55 erlangen systemd-vconsole-setup[709]: Successfully forked off '(loadkeys)' as PID 723. Oct 25 07:13:55 erlangen systemd-vconsole-setup[709]: /usr/bin/loadkeys succeeded. Oct 25 07:13:55 erlangen systemd[1]: systemd-vconsole-setup.service: Child 709 belongs to systemd-vconsole-setup.service. Oct 25 07:13:55 erlangen systemd[1]: systemd-vconsole-setup.service: Main process exited, code=exited, status=0/SUCCESS Oct 25 07:13:55 erlangen systemd[1]: systemd-vconsole-setup.service: Changed start -> dead Oct 25 07:13:55 erlangen systemd[1]: systemd-vconsole-setup.service: Job systemd-vconsole-setup.service/start finished, result=done Oct 25 07:13:55 erlangen systemd[1]: Started Setup Virtual Console.
(In reply to Karl Mistelberger from comment #110) > systemd-vconsole-setup.service is forked by systemd-udevd.service. It fails > on the first fork, but succeeds on the subsequent ones: That is I think not strange, the problems happen during boot and the first time the files are not present/available yet while later they are. The patch: +After=systemd-vconsole-setup.service +Requires=systemd-vconsole-setup.service is aking sure dracut-cmdline-ask.service is started later in the boot process.
(In reply to Karl Mistelberger from comment #110) > systemd-vconsole-setup.service is forked by systemd-udevd.service. It fails > on the first fork, but succeeds on the subsequent ones: That is I think not strange, the problems happen during boot and the first time the files are not present/available yet while later they are. The patch: +After=systemd-vconsole-setup.service +Requires=systemd-vconsole-setup.service is making sure dracut-cmdline-ask.service is started later in the boot process.
(In reply to M Fredericks from comment #112) > (In reply to Karl Mistelberger from comment #110) > > systemd-vconsole-setup.service is forked by systemd-udevd.service. It fails > > on the first fork, but succeeds on the subsequent ones: > > That is I think not strange, the problems happen during boot and the first > time the files are not present/available yet while later they are. > > The patch: > > +After=systemd-vconsole-setup.service > +Requires=systemd-vconsole-setup.service > > is making sure dracut-cmdline-ask.service is started later in the boot > process. added these lines to section [unit] of /usr/lib/dracut/modules.d/98dracut-systemd/dracut-cmdline-ask.service, ran mkinitrd and rebooted: No more errors from systemd-vconsole-setup.service!
Is this fix going to up released to Factory soon?
Is this fix going to be released to Factory soon?
(In reply to Karl Mistelberger from comment #113) > (In reply to M Fredericks from comment #112) > > (In reply to Karl Mistelberger from comment #110) > > > systemd-vconsole-setup.service is forked by systemd-udevd.service. It fails > > > on the first fork, but succeeds on the subsequent ones: > > > > That is I think not strange, the problems happen during boot and the first > > time the files are not present/available yet while later they are. > > > > The patch: > > > > +After=systemd-vconsole-setup.service > > +Requires=systemd-vconsole-setup.service > > > > is making sure dracut-cmdline-ask.service is started later in the boot > > process. > > added these lines to section [unit] of > /usr/lib/dracut/modules.d/98dracut-systemd/dracut-cmdline-ask.service, ran > mkinitrd and rebooted: No more errors from systemd-vconsole-setup.service! It crucially depends on where you put the additional lines in section [unit], doesn't it?
(In reply to Karl Mistelberger from comment #113) > (In reply to M Fredericks from comment #112) > > (In reply to Karl Mistelberger from comment #110) > > > systemd-vconsole-setup.service is forked by systemd-udevd.service. It fails > > > on the first fork, but succeeds on the subsequent ones: > > > > That is I think not strange, the problems happen during boot and the first > > time the files are not present/available yet while later they are. > > > > The patch: > > > > +After=systemd-vconsole-setup.service > > +Requires=systemd-vconsole-setup.service > > > > is making sure dracut-cmdline-ask.service is started later in the boot > > process. > > added these lines to section [unit] of > /usr/lib/dracut/modules.d/98dracut-systemd/dracut-cmdline-ask.service, ran > mkinitrd and rebooted: No more errors from systemd-vconsole-setup.service! (In reply to M Fredericks from comment #112) > (In reply to Karl Mistelberger from comment #110) > > systemd-vconsole-setup.service is forked by systemd-udevd.service. It fails > > on the first fork, but succeeds on the subsequent ones: > > That is I think not strange, the problems happen during boot and the first > time the files are not present/available yet while later they are. > > The patch: > > +After=systemd-vconsole-setup.service > +Requires=systemd-vconsole-setup.service > > is making sure dracut-cmdline-ask.service is started later in the boot > process. I tried this and locked myself out of my system. I use an encrypted disk setup (LUKS). After applying this changes and recreating the initramfs it was not possible to enter the password for the disk decryption. But without decrypted disks the boot process got stuck.
(In reply to Frank Kruger from comment #116) > It crucially depends on where you put the additional lines in section > [unit], doesn't it? Not sure but better see the attachment "proposed patch" and either apply that or make the changes according to that patch. And .. do not forgot to run mkinitrd after making the change, I forgot that the first time ;-)
(In reply to Hendrik Woltersdorf from comment #117) > I tried this and locked myself out of my system. I use an encrypted disk > setup (LUKS). After applying this changes and recreating the initramfs it > was not possible to enter the password for the disk decryption. But without > decrypted disks the boot process got stuck. Interesting, too bad the patch does have bad side-effects using an encrypted disk setup (LUKS). I guess you landed here because you also see the "Failed to start Setup Virtual Console." message in your log, right? I know too less of the whole process to see what is the problem.
Daniel, when do you plan to release the fix ? AFAICS it's been accepted by upstream almost 2 months ago: https://github.com/dracutdevs/dracut/pull/457 Thanks.
(In reply to Franck Bui from comment #120) > Daniel, when do you plan to release the fix ? > > AFAICS it's been accepted by upstream almost 2 months ago: > https://github.com/dracutdevs/dracut/pull/457 > > Thanks. Applying the above-mentioned changes to dracut-cmdline-ask.service, running mkinitrd and rebooting makes it worse for me (TW20181029). Besides the error message "Failed to start Setup Virtual Console" I get "Dependency failed for dracut ask for additional cmdline parameters". By the way, applying the patch to Leap 15 works, apart from the fact that it seems to affect the trashy plymouth splash screen (the squares moved to the upper left corner).
Leap 15.0 here. Applied the patch above and at time the boot hangs for a few seconds.
(In reply to Frank Kruger from comment #121) > (In reply to Franck Bui from comment #120) > > Daniel, when do you plan to release the fix ? > > > > AFAICS it's been accepted by upstream almost 2 months ago: > > https://github.com/dracutdevs/dracut/pull/457 > > > > Thanks. > > Applying the above-mentioned changes to dracut-cmdline-ask.service, running > mkinitrd and rebooting makes it worse for me (TW20181029). Besides the error > message "Failed to start Setup Virtual Console" I get "Dependency failed for > dracut ask for additional cmdline parameters". By the way, applying the > patch to Leap 15 works, apart from the fact that it seems to affect the > trashy plymouth splash screen (the squares moved to the upper left corner). 20181029-0 killed the changes made. Any idea how to create a drop-in for dracut-cmdline-ask.service? I have one for fstrim.timer: Drop-In: /etc/systemd/system/fstrim.timer.d └─schedule.conf
(In reply to Franck Bui from comment #120) > Daniel, when do you plan to release the fix ? > > AFAICS it's been accepted by upstream almost 2 months ago: > https://github.com/dracutdevs/dracut/pull/457 > > Thanks. Working on it now.
This is an autogenerated message for OBS integration: This bug (1055834) was mentioned in https://build.opensuse.org/request/show/645721 Factory / dracut
Updating to dracut-044.1-49.4.x86_64 from https://download.opensuse.org/repositories/Base:/System/openSUSE_Factory/ with the alleged fix (https://build.opensuse.org/request/show/645917) I am getting the following error messages while rebooting: -- [FAILED] Failed to start Setup Virtual Console See 'systemctl status systemd-vconsole-setup.service' for details [DEPEND] Dependency failed for dracut ask for additional cmdline parameters Starting dracut cmdline hook... -- Anyone else seeing this? Any idea? Thx.
(In reply to Frank Kruger from comment #130) > Updating to dracut-044.1-49.4.x86_64 from > https://download.opensuse.org/repositories/Base:/System/openSUSE_Factory/ > with the alleged fix (https://build.opensuse.org/request/show/645917) I am > getting the following error messages while rebooting: > -- > [FAILED] Failed to start Setup Virtual Console > See 'systemctl status systemd-vconsole-setup.service' for details > [DEPEND] Dependency failed for dracut ask for additional cmdline parameters > Starting dracut cmdline hook... > -- > Anyone else seeing this? Any idea? Thx. With the update to TW20181107 and dracut-044.1-22.1.x86_64 the issue is not solved at all (see above).
version 044.1-22.1 fixed my desktop, but the notebook shows: notebook:~ # journalctl -b -u systemd-vconsole-setup.service -- Logs begin at Tue 2018-11-06 00:51:53 CET, end at Sat 2018-11-10 10:13:33 CET. -- Nov 10 09:54:35 notebook systemd-vconsole-setup[199]: cannot open include file compose.latin1 Nov 10 09:54:35 notebook systemd-vconsole-setup[199]: syntax error, unexpected ERROR Nov 10 09:54:35 notebook systemd-vconsole-setup[199]: /usr/bin/loadkeys failed with exit status 1. Nov 10 09:54:35 notebook systemd-vconsole-setup[199]: KD_FONT_OP_GET failed while trying to get the font metadata: Function not implemented Nov 10 09:54:35 notebook systemd-vconsole-setup[199]: Fonts will not be copied to remaining consoles Nov 10 09:54:37 notebook systemd[1]: Starting Setup Virtual Console... Nov 10 09:54:37 notebook systemd-vconsole-setup[375]: cannot open include file compose.latin1 Nov 10 09:54:37 notebook systemd-vconsole-setup[375]: syntax error, unexpected ERROR Nov 10 09:54:37 notebook systemd-vconsole-setup[375]: /usr/bin/loadkeys failed with exit status 1. Nov 10 09:54:37 notebook systemd[1]: systemd-vconsole-setup.service: Main process exited, code=exited, status=1/FAILURE Nov 10 09:54:37 notebook systemd[1]: systemd-vconsole-setup.service: Failed with result 'exit-code'. Nov 10 09:54:37 notebook systemd[1]: Failed to start Setup Virtual Console. Nov 10 09:54:37 notebook systemd[1]: Started Setup Virtual Console. Nov 10 09:54:38 notebook systemd[1]: Starting Setup Virtual Console... Nov 10 09:54:38 notebook systemd[1]: Started Setup Virtual Console. notebook:~ # Both are running: Betriebssystem: openSUSE Tumbleweed 20181107 KDE-Plasma-Version: 5.14.2 Qt-Version: 5.11.2 KDE-Frameworks-Version: 5.51.0
Looks similar here: sudo journalctl -b -u systemd-vconsole-setup.service -- Logs begin at Sun 2018-02-18 21:00:02 CET, end at Sat 2018-11-10 10:18:29 CET. -- Nov 10 09:59:07 gropius systemd[1]: Starting Setup Virtual Console... Nov 10 09:59:07 gropius systemd-vconsole-setup[332]: cannot open include file compose.latin1 Nov 10 09:59:07 gropius systemd-vconsole-setup[332]: syntax error, unexpected ERROR Nov 10 09:59:07 gropius systemd-vconsole-setup[332]: /usr/bin/loadkeys failed with exit status 1. Nov 10 09:59:07 gropius systemd[1]: systemd-vconsole-setup.service: Main process exited, code=exited, status=1/FAILURE Nov 10 09:59:07 gropius systemd[1]: systemd-vconsole-setup.service: Failed with result 'exit-code'. Nov 10 09:59:07 gropius systemd[1]: Failed to start Setup Virtual Console. Nov 10 09:59:17 gropius systemd[1]: Started Setup Virtual Console. Nov 10 09:59:21 gropius systemd[1]: Starting Setup Virtual Console... Nov 10 09:59:21 gropius systemd[1]: Started Setup Virtual Console.
I am running "openSUSE Tumbleweed 20181107" and do NOT see the lines: systemd-vconsole-setup: cannot open include file compose.latin1 systemd-vconsole-setup: syntax error, unexpected ERROR systemd-vconsole-setup: /usr/bin/loadkeys failed with exit status 1. I see the lines: systemd-vconsole-setup: KD_FONT_OP_GET failed while trying to get the font metadata: Function not implemented systemd-vconsole-setup: Fonts will not be copied to remaining consoles But that seems to be informational only, as systemd-vconsole-setup.service is not reporting an error. Although you see "Failed to start Setup Virtual Console" the underlying error is different.
Is this update going to be pushed to Leap 15.0 as well?
(In reply to Frank Kruger from comment #133) > Looks similar here: > > sudo journalctl -b -u systemd-vconsole-setup.service > -- Logs begin at Sun 2018-02-18 21:00:02 CET, end at Sat 2018-11-10 10:18:29 > CET. -- > Nov 10 09:59:07 gropius systemd[1]: Starting Setup Virtual Console... > Nov 10 09:59:07 gropius systemd-vconsole-setup[332]: cannot open include > file compose.latin1 > Nov 10 09:59:07 gropius systemd-vconsole-setup[332]: syntax error, > unexpected ERROR > Nov 10 09:59:07 gropius systemd-vconsole-setup[332]: /usr/bin/loadkeys > failed with exit status 1. > Nov 10 09:59:07 gropius systemd[1]: systemd-vconsole-setup.service: Main > process exited, code=exited, status=1/FAILURE > Nov 10 09:59:07 gropius systemd[1]: systemd-vconsole-setup.service: Failed > with result 'exit-code'. > Nov 10 09:59:07 gropius systemd[1]: Failed to start Setup Virtual Console. > Nov 10 09:59:17 gropius systemd[1]: Started Setup Virtual Console. > Nov 10 09:59:21 gropius systemd[1]: Starting Setup Virtual Console... > Nov 10 09:59:21 gropius systemd[1]: Started Setup Virtual Console. FYI: I reinstalled kbd and kbd-legacy, rebootet and the issue as well as the above-mentioned error messages are gone.
Fixed: https://forums.opensuse.org/showthread.php/533663-Failed-to-start-Setup-Virtual-Console?p=2885995#post2885995
(In reply to M Fredericks from comment #134) > I am running "openSUSE Tumbleweed 20181107" and do NOT see the lines: > > systemd-vconsole-setup: cannot open include file compose.latin1 > systemd-vconsole-setup: syntax error, unexpected ERROR > systemd-vconsole-setup: /usr/bin/loadkeys failed with exit status 1. > > I see the lines: > > systemd-vconsole-setup: KD_FONT_OP_GET failed while trying to get the font > metadata: Function not implemented > systemd-vconsole-setup: Fonts will not be copied to remaining consoles > > But that seems to be informational only, as systemd-vconsole-setup.service > is not reporting an error. > > Although you see "Failed to start Setup Virtual Console" the underlying > error is different. Agreed. Incidentally, the error message "systemd-vconsole-setup: cannot open include file compose.latin1" is caused by "KEYMAP=de-latin1-nodeadkeys" in /etc/vconsole.conf, which was written by Yast2 Country Settings, but does not exist in /usr/share/kbd/keymaps/xkb/ (only de-nodeadkeys.map.gz exists).
(In reply to Frank Kruger from comment #138) > (In reply to M Fredericks from comment #134) > > I see the lines: > > > > systemd-vconsole-setup: KD_FONT_OP_GET failed while trying to get the font > > metadata: Function not implemented > > systemd-vconsole-setup: Fonts will not be copied to remaining consoles > > > Agreed. Incidentally, the error message "systemd-vconsole-setup: cannot open > include file compose.latin1" is caused by "KEYMAP=de-latin1-nodeadkeys" in > /etc/vconsole.conf, which was written by Yast2 Country Settings, but does > not exist in /usr/share/kbd/keymaps/xkb/ (only de-nodeadkeys.map.gz exists). This morning I booted the computer and did not see the problem anymore. Like written on the forum, the only thing I changed AFAIK is that I fixed another error I saw in the log ("kwin_x11[2297]: kwin_xkbcommon: XKB: couldn't find a Compose file for locale "en_NL.UTF-8") by adding to /usr/share/X11/locale/locale.alias and alias from en_NL.UTF-8 to en_US.UTF-8. Seems like that fixed it.
(In reply to Luigi Baldoni from comment #135) > Is this update going to be pushed to Leap 15.0 as well? Is there any chance to receive an update for Leap 15?
I have submitted a maintenance update (see #150), but this usually takes a while to pass QA.
This bug, or one closely related, causes my system(s) to drop to an emergency console during startup every single time. This does not occur on a vanilla installation of Leap 15, but updating via zypper update results in an unbootable system. I have two machines that his affects, one is Tumbleweed while the other is Leap 15. What works: Leap 15, as shipped, with no updates What doesn't work: Leap 15 after updates, Tumbleweed, Leap 15 network install One system has an ancient nVidia card and an AMD FX-8050 on a gigabyte motherboard and the other has an AMD Vega with AMD 1950X on an Asus X399 motherboard. This conversation may or may not be relevant, it's at least a good read for the snarky comments about SuSE being antiquated and at-fault for this sort of thing: https://github.com/systemd/systemd/issues/6689 Lastly, as none of my computers except my OpenSuSE 42.3 laptop is currently able to boot I may have some trouble getting logs...and I'm not very good so I'll probably have to be coached. Stuff's definitely broken though, so whatever I can do to help. Cheers.
Turns out I was off-base a bit, it's an LVM problem over here. The volumes don't mount properly, which is what stops my boot process. It looks similar to the above, but I guess it's maybe not related. It's still something that's broken with a fully updated system, both Leap and TW, so I'll hunt around for open bugs and if I don't see it listed I'll open a new one. Cheers.
SUSE-RU-2018:3983-1: An update that has two recommended fixes can now be installed. Category: recommended (moderate) Bug References: 1055834,1090884 CVE References: Sources used: SUSE Linux Enterprise Module for Open Buildservice Development Tools 15 (src): dracut-044.1-18.9.1 SUSE Linux Enterprise Module for Basesystem 15 (src): dracut-044.1-18.9.1
SUSE-RU-2018:3990-1: An update that has three recommended fixes can now be installed. Category: recommended (moderate) Bug References: 1055834,1090884,1094603 CVE References: Sources used: SUSE Linux Enterprise Server 12-SP4 (src): dracut-044.1-10.3.1 SUSE Linux Enterprise Desktop 12-SP4 (src): dracut-044.1-10.3.1
openSUSE-RU-2018:4039-1: An update that has two recommended fixes can now be installed. Category: recommended (moderate) Bug References: 1055834,1090884 CVE References: Sources used: openSUSE Leap 15.0 (src): dracut-044.1-lp150.14.9.1
SUSE-RU-2018:4092-1: An update that has 7 recommended fixes can now be installed. Category: recommended (moderate) Bug References: 1055834,1089332,1090884,1094603,1098448,1104090,1104178 CVE References: Sources used: SUSE Linux Enterprise Server 12-SP3 (src): dracut-044.1-114.20.1 SUSE Linux Enterprise Desktop 12-SP3 (src): dracut-044.1-114.20.1 SUSE CaaS Platform ALL (src): dracut-044.1-114.20.1 SUSE CaaS Platform 3.0 (src): dracut-044.1-114.20.1
openSUSE-RU-2018:4139-1: An update that has 7 recommended fixes can now be installed. Category: recommended (moderate) Bug References: 1055834,1089332,1090884,1094603,1098448,1104090,1104178 CVE References: Sources used: openSUSE Leap 42.3 (src): dracut-044.1-32.1
Backport for Leap is on its way. Closing as fixed.
Backport for openSUSE 15.0 is on its way. Closing as fixed.
SUSE-RU-2019:0546-1: An update that has 14 recommended fixes can now be installed. Category: recommended (moderate) Bug References: 1008352,1013573,1053248,1055834,1090884,1098448,1110519,1112327,1113712,1119499,1121251,1124088,1125327,937555 CVE References: Sources used: SUSE Linux Enterprise Server 12-SP3 (src): dracut-044.1-114.25.1 SUSE Linux Enterprise Desktop 12-SP3 (src): dracut-044.1-114.25.1 SUSE CaaS Platform ALL (src): dracut-044.1-114.25.1 SUSE CaaS Platform 3.0 (src): dracut-044.1-114.25.1