Login screen frozen keybord/mouse not responsive
Re: Login screen frozen keybord/mouse not responsive
You can find package history by using the 'yum history' command (see the man page, it has lots of subcommands and options) or by reading the various /var/log/dnf.rpm-log* files or even by using `rpm -qi centos-stream-release` which will show a build and install date/time.
You can see what is running by using the `ps fax` command. That shows you the processes as a tree so you can see what started what. On my CentOS 7 system that shows stuff running as a child of /usr/sbin/gdm and those would be the processes started when you logged in and will have a bunch of child processes themselves. I would imagine that if you killed those then gdm would pop back in and prompt for a login. Or you could `systemctl list-units` and find ones in there that sound like they have to do with the GUI and systemctl restart them one by one. Or perhaps `systemctl isolate multi-user.target` which will switch from what used to be called runlevel 5 (GUI with network) to runlevel 3, command line only logins. That should kill the entire GUI off then you can `systemctl isolate graphical.target` to fire it all back up again. Or just `shutdown -r now` from a root command line and reboot the entire box. If sshd is not running I'd suggest starting it up from webmin.
You can see what is running by using the `ps fax` command. That shows you the processes as a tree so you can see what started what. On my CentOS 7 system that shows stuff running as a child of /usr/sbin/gdm and those would be the processes started when you logged in and will have a bunch of child processes themselves. I would imagine that if you killed those then gdm would pop back in and prompt for a login. Or you could `systemctl list-units` and find ones in there that sound like they have to do with the GUI and systemctl restart them one by one. Or perhaps `systemctl isolate multi-user.target` which will switch from what used to be called runlevel 5 (GUI with network) to runlevel 3, command line only logins. That should kill the entire GUI off then you can `systemctl isolate graphical.target` to fire it all back up again. Or just `shutdown -r now` from a root command line and reboot the entire box. If sshd is not running I'd suggest starting it up from webmin.
The future appears to be RHEL or Debian. I think I'm going Debian.
Info for USB installs on http://wiki.centos.org/HowTos/InstallFromUSBkey
CentOS 5 and 6 are deadest, do not use them.
Use the FAQ Luke
Info for USB installs on http://wiki.centos.org/HowTos/InstallFromUSBkey
CentOS 5 and 6 are deadest, do not use them.
Use the FAQ Luke
Re: Login screen frozen keybord/mouse not responsive
Thanks Trevor!
I really want to avoid to shut it down as it is my running web- and mailserver. The risk that it just might just become completely unreachable , and non-functional, is too big. I think I will try the first part of your reply and see what I can find that looks like gdm.
Logwatch (yes this works too) also reports a cron initiated process that is apparently started at midnight and at noon and is delayed every minute for one hour after it is initiated at midnight and noon. It seems not to be working. Could perhaps be the renewal of the SSL certificate through Let'sencrypt. These are renewed every 3 months. Just checked the padlock of the website and this shows it's OK until 19 March.
The only cron command that starts every 12 hours is :
/usr/local/bin/certbot-auto renew >/dev/null 2>&1 --renew-hook "systemctl restart httpd.service" (when checking with webmin).
and doesn't seem to work properly. Why is this cron process delayed every 12 hours and then not executed?
Can this be related to the login issue??
Any comments/suggestions?
I would'nt be able to that using webmin from another computer, would I? Leaving "Runlevel 5" would disable the network connection wouldn't it? So I would not be able to put it back to runlevel 5.Or perhaps `systemctl isolate multi-user.target` which will switch from what used to be called runlevel 5 (GUI with network) to runlevel 3, command line only logins. That should kill the entire GUI off then you can `systemctl isolate graphical.target` to fire it all back up again.
I really want to avoid to shut it down as it is my running web- and mailserver. The risk that it just might just become completely unreachable , and non-functional, is too big. I think I will try the first part of your reply and see what I can find that looks like gdm.
Logwatch (yes this works too) also reports a cron initiated process that is apparently started at midnight and at noon and is delayed every minute for one hour after it is initiated at midnight and noon. It seems not to be working. Could perhaps be the renewal of the SSL certificate through Let'sencrypt. These are renewed every 3 months. Just checked the padlock of the website and this shows it's OK until 19 March.
The only cron command that starts every 12 hours is :
/usr/local/bin/certbot-auto renew >/dev/null 2>&1 --renew-hook "systemctl restart httpd.service" (when checking with webmin).
and doesn't seem to work properly. Why is this cron process delayed every 12 hours and then not executed?
Can this be related to the login issue??
Any comments/suggestions?
Re: Login screen frozen keybord/mouse not responsive
From systemctl show graphical.target:
Code: Select all
Requires=multi-user.target
Wants=gdm.service udisks2.service accounts-daemon.service rtkit-daemon.service systemd-update-utmp-runlevel.service
Code: Select all
Requires=basic.target
Wants=cups.service firewalld.service ModemManager.service irqbalance.service crond.service chronyd.service netcf-transaction.service rpcbind.service systemd-logind.service mdmonitor.service dnf-makecache.timer libstoragemgmt.service rsyslog.service sssd.service plymouth-quit.service dbus.service vmtoolsd.service lm_sensors.service plymouth-quit-wait.service sshd.service cups.path systemd-user-sessions.service atd.service auditd.service kdump.service remote-fs.target smartd.service ksmtuned.service vdo.service tuned.service nfs-client.target systemd-ask-password-wall.path rhsmcertd.service avahi-daemon.service mcelog.service getty.target systemd-update-utmp-runlevel.service gpm.service ksm.service NetworkManager.service
I have switched between multi-user and graphical targets many times via ssh connection. The network/session/connection is not lost.
I don't (want to) know what webmin is, but systemctl show httpd.service reveals:
Code: Select all
Requires=-.mount system.slice basic.target
WantedBy=multi-user.target
Conflicts=shutdown.target
Before=shutdown.target multi-user.target
After=basic.target system.slice -.mount network.target systemd-journald.socket tmp.mount nss-lookup.target remote-fs.target
Even systemctl restart NetworkManager appears to keep existing connections, unlike nmcli con down xxx ; nmcli con up xxx.
Re: Login screen frozen keybord/mouse not responsive
Hi jlehtone,
Thx for your reply. OK. I see now that it should work as you propose.
However I seem to have problems in executing commands through the WEBMIN interface, see for instance:
and:
So I don't seem to be able to run such commands. I can kill processes, which I have done. I killed all processes owned by a std user I use on the server, as I don't want to be logged in as root all the time. By killing those processes, among which some were gdm related processes, the GUI to my monitor dissapeared complety and now only shows something like std output with some warnings from the system. I don't have any GUI anymore. Keyboard/mouse are dead.
And according to the codes above, I cannot start the GUI either, nor start sshd. All systemctl like commands don't seem to work through Webmin.
So I guess this is it. EIther reboot and risk that it may not reboot properly (and my main server is down), OR launch my old centos 7 backup server and make it the main server (requires some work with mail-backups), OR build a new server from scratch with Rocky Linux or Alma Linux.
Or do I gave any other options? to turn my server into normal operation again?
Thx for any feedback!
Thx for your reply. OK. I see now that it should work as you propose.
However I seem to have problems in executing commands through the WEBMIN interface, see for instance:
Code: Select all
> systemctl isolate graphical.target
Failed to start graphical.target: Connection timed out
See system logs and 'systemctl status graphical.target' for details.
> systemctl start sshd*
Failed to list units: Connection timed out
Failed to expand names: Connection timed out
> systemctl start sshd.service
Failed to start sshd.service: Connection timed out
See system logs and 'systemctl status sshd.service' for details.
> systemctl show httpd.service
Failed to get properties: Connection timed out
Code: Select all
> usr/bin/startx
sh: usr/bin/startx: No such file or directory
> /usr/bin/startx
xauth: file /root/.serverauth.3470139 does not exist
xauth: file /root/.Xauthority does not exist
xauth: file /root/.Xauthority does not exist
X.Org X Server 1.20.11
X Protocol Version 11, Revision 0
Build Operating System: 4.19.34-300.el7.x86_64
Current Operating System: Linux server1.hartings.se 4.18.0-338.el8.x86_64 #1 SMP Fri Aug 27 17:32:14 UTC 2021 x86_64
Kernel command line: BOOT_IMAGE=(mduuid/7e7d733e07b3602321ad24f663d35fcf)/vmlinuz-4.18.0-338.el8.x86_64 root=UUID=ddc9f682-ba50-4132-9314-72d1dc86c0ce ro crashkernel=auto resume=UUID=ec5afab0-2237-4868-ac4c-2f8431867cf4 rd.md.uuid=a930eea7:c7665434:955edc48:0ed252b0 rd.md.uuid=7e7d733e:07b36023:21ad24f6:63d35fcf rd.md.uuid=ed5f06b7:6846d371:47d22bf5:abdc2f34 rhgb quiet
Build Date: 10 June 2021 11:58:07PM
Build ID: xorg-x11-server 1.20.11-2.el8
Current version of pixman: 0.38.4
Before reporting problems, check http://wiki.x.org
to make sure that you have the latest version.
Markers: (--) probed, (**) from config file, (==) default setting,
(++) from command line, (!!) notice, (II) informational,
(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
(==) Log file: "/var/log/Xorg.0.log", Time: Tue Jan 25 13:22:01 2022
(==) Using config directory: "/etc/X11/xorg.conf.d"
(==) Using system config directory "/usr/share/X11/xorg.conf.d"
xf86EnableIOPorts: failed to set IOPL for I/O (Operation not permitted)
(EE)
Fatal server error:
(EE) AddScreen/ScreenInit failed for driver 0
(EE)
(EE)
Please consult the The X.Org Foundation support
at http://wiki.x.org
for help.
(EE) Please also check the log file at "/var/log/Xorg.0.log" for additional information.
(EE)
(EE) Server terminated with error (1). Closing log file.
xinit: giving up
xinit: unable to connect to X server: Connection refused
xinit: server error
> systemctl isolate graphical.target
Failed to start graphical.target: Connection timed out
See system logs and 'systemctl status graphical.target' for details.
And according to the codes above, I cannot start the GUI either, nor start sshd. All systemctl like commands don't seem to work through Webmin.
So I guess this is it. EIther reboot and risk that it may not reboot properly (and my main server is down), OR launch my old centos 7 backup server and make it the main server (requires some work with mail-backups), OR build a new server from scratch with Rocky Linux or Alma Linux.
Or do I gave any other options? to turn my server into normal operation again?
Thx for any feedback!
Re: Login screen frozen keybord/mouse not responsive
Like I said, I don't know what webmin is, what you can do with it, nor how you have to do things with it. Logically, if webmin is an interface to manage a server, then it should have some means to manage services. "Kill" should definitely not be the only options.
Are you sure that the sshd.service is not running?
Are you sure that the sshd.service is not running?
Re: Login screen frozen keybord/mouse not responsive
Yes. Can't see anything like it in the process list (that I can see through webmin). Sshd is started at a reboot, but I always stop it manually when I am not planning to be away for some time. Which hasn't happened since the last update and reboot in December.Are you sure that the sshd.service is not running?
Webmin is just a "universal" package and unfortunately not as powerful as I thought it was, or it's just not working properly now.
I found a setting in webmin though which can change the "run level". Don't know if this is relevant for centos as centos doesn't use this anymore and everything is controlled by systemctl. But could be worth a try?
Re: Login screen frozen keybord/mouse not responsive
RunlevelX are aliases for systemd targets:
Either webmin does use systemd properly as backend, calls telinit like in SysV-based systems, or something.
ssh ... I do all management via ssh. I might restrict who can access via sshd, but never disable sshd.
Code: Select all
Id=rescue.target
Names=runlevel1.target rescue.target
Id=multi-user.target
Names=runlevel2.target multi-user.target runlevel3.target runlevel4.target
Id=graphical.target
Names=runlevel5.target graphical.target
Id=reboot.target
Names=runlevel6.target reboot.target
man telinit wrote:telinit may be used to change the SysV system runlevel. Since the concept of SysV runlevels is obsolete the runlevel requests will be transparently translated into systemd unit activation requests.
ssh ... I do all management via ssh. I might restrict who can access via sshd, but never disable sshd.
Re: Login screen frozen keybord/mouse not responsive
Thx jlehtone for the good info on Runlevels! I will try it tomorrow.
Btw, is there a way to activate ssh without using the systemctl command, which perhaps could work through Webmin?
Yes, this is one of the lessons learned for me... I will no longer deactivate it after a reboot!ssh ... I do all management via ssh. I might restrict who can access via sshd, but never disable sshd.
Btw, is there a way to activate ssh without using the systemctl command, which perhaps could work through Webmin?
Re: Login screen frozen keybord/mouse not responsive
https://doxfer.webmin.com/Webmin/SSH_Server says you can start/stop it from webmin.
The future appears to be RHEL or Debian. I think I'm going Debian.
Info for USB installs on http://wiki.centos.org/HowTos/InstallFromUSBkey
CentOS 5 and 6 are deadest, do not use them.
Use the FAQ Luke
Info for USB installs on http://wiki.centos.org/HowTos/InstallFromUSBkey
CentOS 5 and 6 are deadest, do not use them.
Use the FAQ Luke
Re: Login screen frozen keybord/mouse not responsive
Thx Trevor for your quick reply!
I tried to start sshd through Webmin (there is a dedicated SSH Server tab in Webmin), but it seems it just uses the systemctl command and the output is this:
There are now 441 new (stream) packages waiting to be updated (I only updated the security related updates yesterday through Webmin). I noticed also there is a Webmin update. I will try to update the Webmin package and try it again.
As I now seem to be running Stream, should I update all packages? What would you recommend in my situation?
When everything is under controll again I plan anyhow to change to Rocky or Alma linux. I noticed there is an update script to move from Stream 8 to Rocky/Alma linux.
I tried to start sshd through Webmin (there is a dedicated SSH Server tab in Webmin), but it seems it just uses the systemctl command and the output is this:
Code: Select all
Failed to start SSH server :
Redirecting to /bin/systemctl start sshd.service
Failed to start sshd.service: Connection timed out
See system logs and 'systemctl status sshd.service' for details.
As I now seem to be running Stream, should I update all packages? What would you recommend in my situation?
When everything is under controll again I plan anyhow to change to Rocky or Alma linux. I noticed there is an update script to move from Stream 8 to Rocky/Alma linux.