|
Bugzilla – Full Text Bug Listing |
| Summary: | Getting 'Invalid MIT-MAGIC-COOKIE-1 key' error if plymouth | ||
|---|---|---|---|
| Product: | [openSUSE] openSUSE Distribution | Reporter: | Forgotten User ngNQ5IyoFI <forgotten_ngNQ5IyoFI> |
| Component: | X.Org | Assignee: | E-mail List <xorg-maintainer-bugs> |
| Status: | RESOLVED WONTFIX | QA Contact: | E-mail List <xorg-maintainer-bugs> |
| Severity: | Normal | ||
| Priority: | P5 - None | CC: | forgotten_ngNQ5IyoFI, msrb, seife |
| Version: | Leap 42.3 | ||
| Target Milestone: | --- | ||
| Hardware: | x86-64 | ||
| OS: | openSUSE 42.3 | ||
| Whiteboard: | |||
| Found By: | --- | Services Priority: | |
| Business Priority: | Blocker: | --- | |
| Marketing QA Status: | --- | IT Deployment: | --- |
|
Description
Forgotten User ngNQ5IyoFI
2018-02-22 14:20:10 UTC
How is this connected to XFCE? My systems are almost all running without plymouth, so XFCE definitely (IMVHO) has nothing to do with this problem. Never said its xfce, but happens while using xfce. Never seen this issue before using openSUSE 42.1 with xfce, but does happen with 42.3 with xfce. So, got no idea. Feel free to modify the bug, if you don't believe its an xfce issue. THe magic cookie is from X, but the behaviour should not be affected by plymouth... Xorg guys any ideas? It seems that your Xauthority file contains an old cookie that matches your display, but is no longer valid. Do the X applications show the error and terminate, or do the actually work? The Xauthority file is generated by your display manager, which one do you use? (GDM, XDM, ...) It should not be affected by plymouth at all. I would guess whatever happened just happened at the same time you uninstalled it. Are you able to reproduce it again? (In reply to Michal Srb from comment #4) > It seems that your Xauthority file contains an old cookie that matches your > display, but is no longer valid. Do the X applications show the error and > terminate, or do the actually work? The Xauthority file is generated by your > display manager, which one do you use? (GDM, XDM, ...) > > It should not be affected by plymouth at all. I would guess whatever > happened just happened at the same time you uninstalled it. Are you able to > reproduce it again? I'm not saying anything, but I reinstalled plymouth and the issue is gone. So, whatever it is, reinstalling plymouth fixed it. Robert, Michal asked for reproduction. So, please give him a favor, uninstall plymouth once more and let us know, whether this breaks Xauthority again. Ok? Thanks. Per definition a bug, which cannot be reproduced is no bug. ;-) Or at least cannot be investigated unless it's obvious what's wrong. And in this case it's not. (In reply to Stefan Dirsch from comment #6) > Robert, Michal asked for reproduction. So, please give him a favor, > uninstall plymouth once more and let us know, whether this breaks Xauthority > again. Ok? Thanks. > > Per definition a bug, which cannot be reproduced is no bug. ;-) Or at least > cannot be investigated unless it's obvious what's wrong. And in this case > it's not. Would you relax? I waited for a response for months, I will try to reproduce, but it will take a bit. (In reply to Robert Milasan from comment #7) > Would you relax? I waited for a response for months, Now I feel sorry. :-( Haven't notice this. > I will try to reproduce, but it will take a bit. Thanks a lot! Alright, so just removed plymouth and libply package and I got the same error: Invalid MIT-MAGIC-COOKIE-1 keyInvalid MIT-MAGIC-COOKIE-1 keyInvalid MIT-MAGIC-COOKIE-1 keyrobert@pandora:~> What exactly you need from me to go forward. I haven't touched anything else besides removing the packages and restarting the laptop. Very interesting that removal of plymouth triggered it again! What display manager do you use? (GDM, XDM, ...) You wrote that you can open any X application and get this error. Does the application print the error and then work normally, or does it fail to start? Can you attach output of `xauth list`? (Note that publishing this file can be considered security risk, but if you trust all users on your machine and did not instruct X to listen on TCP and/or you log out and in again after to generate new cookies, you should be safe.) (In reply to Michal Srb from comment #10) > Very interesting that removal of plymouth triggered it again! > > What display manager do you use? (GDM, XDM, ...) > > You wrote that you can open any X application and get this error. Does the > application print the error and then work normally, or does it fail to start? > > Can you attach output of `xauth list`? (Note that publishing this file can > be considered security risk, but if you trust all users on your machine and > did not instruct X to listen on TCP and/or you log out and in again after to > generate new cookies, you should be safe.) I'm using lightdm with lightdm slick greeter. The applications seems to work fine, just the errors are constant for any X app. robert@pandora:~> rpm -qa|grep -i lightdm lightdm-lang-1.22.0-5.3.1.noarch lightdm-slick-greeter-1.0.8-2.1.x86_64 liblightdm-gobject-1-0-1.22.0-5.3.1.x86_64 lightdm-slick-greeter-branding-openSUSE-1.0-1.4.noarch lightdm-slick-greeter-lang-1.0.8-2.1.noarch lightdm-1.22.0-5.3.1.x86_64 robert@pandora:~> xauth list pandora/unix:0 MIT-MAGIC-COOKIE-1 c8a2d3ed44e47762188faac90a2634a9 pandora.concurtech.net/unix:0 MIT-MAGIC-COOKIE-1 abfebb3247d4d22d127c8f7873c18504 Thank you. Lightdm uses the user's $HOME/.Xauthority, so it is possible that some previous entry remains in there. But in your case there are only two and they differ in the hostname. Is it possible that your computer starts with hostname "pandora" and later is given the full name from DHCP to become "pandora.concurtech.net"? Can you check the output of `hostname -f` and content of XAUTHLOCALHOSTNAME environmental variable (inside graphical session)? Plymouth may be causing different timing during boot so that the hostname is changed at different point relative to the start of X. (In reply to Michal Srb from comment #12) > Thank you. Lightdm uses the user's $HOME/.Xauthority, so it is possible that > some previous entry remains in there. But in your case there are only two > and they differ in the hostname. > > Is it possible that your computer starts with hostname "pandora" and later > is given the full name from DHCP to become "pandora.concurtech.net"? > > Can you check the output of `hostname -f` and content of XAUTHLOCALHOSTNAME > environmental variable (inside graphical session)? > > Plymouth may be causing different timing during boot so that the hostname is > changed at different point relative to the start of X. There is no DHCP, this seems to be related to some config issue. robert@pandora:~> hostname -f pandora.concurtech.net robert@pandora:~> cat /etc/hostname pandora robert@pandora:~> cat /etc/HOSTNAME pandora robert@pandora:~> cat /etc/hosts # # hosts This file describes a number of hostname-to-address # mappings for the TCP/IP subsystem. It is mostly # used at boot time, when no name servers are running. # On small systems, this file can be used instead of a # "named" name server. # Syntax: # # IP-Address Full-Qualified-Hostname Short-Hostname # 127.0.0.1 localhost 127.0.0.1 pandora.concurtech.net pandora # special IPv6 addresses ::1 localhost ipv6-localhost ipv6-loopback fe00::0 ipv6-localnet ff00::0 ipv6-mcastprefix ff02::1 ipv6-allnodes ff02::2 ipv6-allrouters ff02::3 ipv6-allhosts robert@pandora:~> echo $XAUTHLOCALHOSTNAME pandora robert@pandora:~> cat .Xauthority pandora.concurtech.net0MIT-MAGIC-COOKIE-1���2G��-|�xs��pandora0MIT-MAGIC-COOKIE-1Q�/w�PcS�(J�azrobert@pandora:~> Can I drop .Xauthority (I mean delete), will that get regenerated correctly? Yes, you can. But you'll need a new session login to create a new one. Also it would be better to just rename it, so we still have a chance to investigate the issue. After fixing the hostname, the error disappeared. Not sure if thats a good thing or not. I'm afraid we won't get here to anywhere. :-( Therefore closing this bug eventually. In case you figure out something, please let me know and/or open the bug. Thanks! |