Skip to main content ↑ ↓
Topic: Login issue after upgrade above linux-5.15.5-artix1-1 (Read 1903 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Login issue after upgrade above linux-5.15.5-artix1-1

After updating the system above linux-5.15.5-artix1-1 I can not log in. After entering the username and password, the same window appear. Initially, the system was set to enter without requiring a login and password. But after the update, this request appears.

Re: Login issue after upgrade above linux-5.15.5-artix1-1

Reply #1 –
You should boot into a live session and mount your system to query more information about what is going on in your logs

Re: Login issue after upgrade above linux-5.15.5-artix1-1

Reply #3 –
Search for authentication issues

Re: Login issue after upgrade above linux-5.15.5-artix1-1

Reply #4 –
Quote from: qontinuum – on
Search for authentication issues

As I understand it, I have to search in this folder "/var/log". But in which file and which line should I pay attention to? There are a lot of files: auth.log , errors.log , messages.log , everything.log , and others

Re: Login issue after upgrade above linux-5.15.5-artix1-1

Reply #5 –
Quote from: MrBrooks – on
As I understand it, I have to search in this folder "/var/log". But in which file and which line should I pay attention to? There are a lot of files: auth.log , errors.log , messages.log , everything.log , and others
How about:
Code: [Select]
$ grep -E '(auth|login)' /var/log/* 2>/dev/null | less
or even
Code: [Select]
$ grep -E '(auth|login)' /var/log/* 2>/dev/null >interesting.log.lines
then you get those lines neatly in a file.

Re: Login issue after upgrade above linux-5.15.5-artix1-1

Reply #6 –
Quote from: strajder – on
How about:
Code: [Select]
$ grep -E '(auth|login)' /var/log/* 2>/dev/null | less
or even
Code: [Select]
$ grep -E '(auth|login)' /var/log/* 2>/dev/null >interesting.log.lines
then you get those lines neatly in a file.

Thanks, this makes searching much faster.

For now I have found these suspicious lines. I tried to login three times.
Code: [Select]
/log/errors.log:Jan 26 17:00:39 linux lightdm[1701]: pam_elogind(lightdm-greeter:session): Failed to release session: No session 'c1' known
/log/errors.log:Jan 26 17:01:04 linux lightdm[1884]: pam_elogind(lightdm-greeter:session): Failed to release session: No session 'c2' known
/log/errors.log:Jan 26 17:01:12 linux lightdm[2014]: pam_elogind(lightdm-greeter:session): Failed to release session: No session 'c3' known
/log/errors.log:Jan 26 17:01:24 linux lightdm[2140]: pam_elogind(lightdm-greeter:session): Failed to connect to system bus: No such file or dire
ctory

and
Code: [Select]
log/auth.log:Jan 26 17:00:22 linux lightdm[1605]: pam_succeed_if(lightdm-autologin:auth): requirement "user ingroup autologin" was met by user 
"artix"
/log/auth.log:Jan 26 17:00:22 linux lightdm[1605]: pam_unix(lightdm-autologin:session): session opened for user artix(uid=1000) by (uid=0)
/log/auth.log:Jan 26 17:00:22 linux kernel: elogind[1510]: New session 1 of user artix.
/log/auth.log:Jan 26 17:00:22 linux lightdm[1605]: pam_env(lightdm-autologin:session): deprecated reading of user environment enabled
/log/auth.log:Jan 26 17:00:23 linux lightdm[1605]: pam_unix(lightdm-autologin:session): session closed for user artix
/log/auth.log:Jan 26 17:00:23 linux kernel: elogind[1510]: Removed session 1.
/log/auth.log:Jan 26 17:00:25 linux kernel: elogind[1510]: New session c1 of user lightdm.
/log/auth.log:Jan 26 17:00:39 linux kernel: elogind[1510]: Removed session c1.
/log/auth.log:Jan 26 17:00:39 linux lightdm[1701]: pam_elogind(lightdm-greeter:session): Failed to release session: No session 'c1' known
/log/auth.log:Jan 26 17:00:39 linux kernel: elogind[1510]: New session 2 of user artix.
/log/auth.log:Jan 26 17:00:40 linux kernel: elogind[1510]: Removed session 2.
/log/auth.log:Jan 26 17:00:41 linux kernel: elogind[1510]: New session c2 of user lightdm.
/log/auth.log:Jan 26 17:01:04 linux kernel: elogind[1510]: Removed session c2.
/log/auth.log:Jan 26 17:01:04 linux lightdm[1884]: pam_elogind(lightdm-greeter:session): Failed to release session: No session 'c2' known
/log/auth.log:Jan 26 17:01:04 linux kernel: elogind[1510]: New session 3 of user artix.
/log/auth.log:Jan 26 17:01:05 linux kernel: elogind[1510]: Removed session 3.
/log/auth.log:Jan 26 17:01:06 linux kernel: elogind[1510]: New session c3 of user lightdm.
/log/auth.log:Jan 26 17:01:12 linux kernel: elogind[1510]: Removed session c3.
/log/auth.log:Jan 26 17:01:12 linux lightdm[2014]: pam_elogind(lightdm-greeter:session): Failed to release session: No session 'c3' known
/log/auth.log:Jan 26 17:01:12 linux kernel: elogind[1510]: New session 4 of user artix.
/log/auth.log:Jan 26 17:01:13 linux kernel: elogind[1510]: Removed session 4.
/log/auth.log:Jan 26 17:01:15 linux kernel: elogind[1510]: New session c4 of user lightdm.
/log/auth.log:Jan 26 17:01:23 linux kernel: elogind[1510]: System is rebooting..
/log/auth.log:Jan 26 17:01:24 linux kernel: elogind[1510]: Received signal 15 [TERM]
/log/auth.log:Jan 26 17:01:24 linux lightdm[2140]: pam_elogind(lightdm-greeter:session): Failed to connect to system bus: No such file or directory

Re: Login issue after upgrade above linux-5.15.5-artix1-1

Reply #11 –
Code: [Select]
Jan 24 04:57:35 linux lightdm[1600]: pam_unix(lightdm-autologin:session): session closed for user artix
This looks like a log from live environment's /var/log. If your user was named "artix", disregard the rest of this paragraph. Otherwise, please upload the corresponding logs from your hard disk. You can do that by mounting the root partition (or the partition holding the /var/log on the hard disk) to /mnt in the live environment, like described in the "Troubleshooting" article. Then find the logs under /mnt/var/log instead of /var/log.

Update: Looking at errors.log, this looks suspicious:
Code: [Select]
Jan 23 18:03:19 linux kernel: blk_update_request: I/O error, dev sdb, sector 2043972240 op 0x0:(READ) flags 0x80700 phys_seg 4 prio class 0
Jan 23 18:03:19 linux kernel: blk_update_request: I/O error, dev sdb, sector 2043972240 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
Jan 23 18:03:19 linux kernel: Buffer I/O error on dev sdb5, logical block 175121746, async page read
Jan 23 18:03:19 linux kernel: blk_update_request: I/O error, dev sdb, sector 2043972368 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
[...]
Jan 23 18:24:05 linux kernel: blk_update_request: I/O error, dev sdb, sector 1687508488 op 0x1:(WRITE) flags 0x4800 phys_seg 256 prio class 0
Jan 23 18:24:05 linux kernel: Buffer I/O error on dev sdb5, logical block 130563777, lost async page write
Jan 23 18:24:05 linux kernel: Buffer I/O error on dev sdb5, logical block 130563778, lost async page write
web search gives these links as some of the results:
https://serverfault.com/a/866150
https://www.linuxquestions.org/questions/linux-hardware-18/buffer-i-o-error-on-dev-sdb1-async-page-read-4175600715/
https://old.reddit.com/r/debian/comments/omic2r/buffer_io_on_dev_sdb_logical_block_0_async_page/h5lcovs/

Next,
Code: [Select]
Jan 24 04:57:35 linux pulseaudio[1678]: [pulseaudio] stdin-util.c: Lost I/O connection in module "module-gsettings"
Jan 24 04:57:35 linux pulseaudio[1678]: [pulseaudio] core-util.c: Failed to create secure directory (/run/user/1000/pulse): Нет такого файла или каталога
this is weird. For some reason (maybe related to disk errors above), a directory required for pulseaudio to function is missing. I'd check hard disk cables and hardware first.

Re: Login issue after upgrade above linux-5.15.5-artix1-1

Reply #12 –
Quote from: strajder – on
Code: [Select]
Jan 24 04:57:35 linux lightdm[1600]: pam_unix(lightdm-autologin:session): session closed for user artix
This looks like a log from live environment's /var/log. If your user was named "artix", disregard the rest of this paragraph. Otherwise, please upload the corresponding logs from your hard disk. You can do that by mounting the root partition (or the partition holding the /var/log on the hard disk) to /mnt in the live environment, like described in the "Troubleshooting" article. Then find the logs under /mnt/var/log instead of /var/log.

Update: Looking at errors.log, this looks suspicious:
Code: [Select]
Jan 23 18:03:19 linux kernel: blk_update_request: I/O error, dev sdb, sector 2043972240 op 0x0:(READ) flags 0x80700 phys_seg 4 prio class 0
Jan 23 18:03:19 linux kernel: blk_update_request: I/O error, dev sdb, sector 2043972240 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
Jan 23 18:03:19 linux kernel: Buffer I/O error on dev sdb5, logical block 175121746, async page read
Jan 23 18:03:19 linux kernel: blk_update_request: I/O error, dev sdb, sector 2043972368 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
[...]
Jan 23 18:24:05 linux kernel: blk_update_request: I/O error, dev sdb, sector 1687508488 op 0x1:(WRITE) flags 0x4800 phys_seg 256 prio class 0
Jan 23 18:24:05 linux kernel: Buffer I/O error on dev sdb5, logical block 130563777, lost async page write
Jan 23 18:24:05 linux kernel: Buffer I/O error on dev sdb5, logical block 130563778, lost async page write
web search gives these links as some of the results:
https://serverfault.com/a/866150
https://www.linuxquestions.org/questions/linux-hardware-18/buffer-i-o-error-on-dev-sdb1-async-page-read-4175600715/
https://old.reddit.com/r/debian/comments/omic2r/buffer_io_on_dev_sdb_logical_block_0_async_page/h5lcovs/

Next,
Code: [Select]
Jan 24 04:57:35 linux pulseaudio[1678]: [pulseaudio] stdin-util.c: Lost I/O connection in module "module-gsettings"
Jan 24 04:57:35 linux pulseaudio[1678]: [pulseaudio] core-util.c: Failed to create secure directory (/run/user/1000/pulse): Нет такого файла или каталога
this is weird. For some reason (maybe related to disk errors above), a directory required for pulseaudio to function is missing. I'd check hard disk cables and hardware first.

1. That's right, user was named "artix"
2. dev sdb - it's a card reader. Didn't notice any problems with it.
3. The system is installed on Nvme and there are no problems before the update. Updated system after Jan 26 17:00.
Perhaps a pulseaudio bug is causing the login to fail after an update?

Re: Login issue after upgrade above linux-5.15.5-artix1-1

Reply #13 –
Quote from: MrBrooks – on
2. dev sdb - it's a card reader. Didn't notice any problems with it.
Are you sure? Devices could be assigned different pseudo-files under /dev; what is once represented by /dev/sda could under some circumstances be under /dev/sdb and vice versa. That's why it is recommended to use UUIDs in /etc/fstab instead of device filenames.

What is the output of
Code: [Select]
lsblk -f

Quote from: MrBrooks – on
Perhaps a pulseaudio bug is causing the login to fail after an update?
It could be. You said you had autologin enabled in lightdm? What is the contents of /etc/lightdm/lightdm.conf? And the output of
Code: [Select]
groups artix
(if the user is named "artix").

If the issue is related to a pacman update, also supply part of /var/log/pacman.log around the date when you experienced the issue first.

Re: Login issue after upgrade above linux-5.15.5-artix1-1

Reply #14 –
strajder
Perhaps it would be faster to reinstall Linux? There will be no problems with installing the necessary programs, but the visual configuration of all panels, shortcuts and other things will take a lot of time. Is it possible to save all these settings with the possibility of transferring to a new system?