Continuous Reboot

Hi

My workstation reboots on a loop endlessly(looks like a memory dump). I am able to boot it to single user but not multi-user. I am attaching a video of the error with this.

Please let me know what to do?

Thanks
Ravi
 
Don't post videos to the forums please.

[Video deleted]
 
Members who want to help are not near your 'server'. How do you know it has dumped the memory ? To solve the problem you need to provide any error massages collected. If it is fresh install then may reinstall the same version (?) of OS first and then check for multi user login.Check for compatibility of machine (?).
 
Members who want to help are not near your 'server'. How do you know it has dumped the memory ? To solve the problem you need to provide any error massages collected. If it is fresh install then may reinstall the same version (?) of OS first and then check for multi user login.Check for compatibility of machine (?).

Hi

I see a message (attaching a photograph of laptop screen) before restart that - core dumped @ /var/crash/vmcore.0, but there is no /crash folder under /var, when viewed, logged in as single boot user.

It is not a fresh install.

Thanks
Ravi
 

Attachments

  • IMG_0191__1594653936_43.230.65.116.jpg
    IMG_0191__1594653936_43.230.65.116.jpg
    58.2 KB · Views: 205
Boot into single user and disable all non-essential tweaks. Then by manually invoking single one's sequentially, you will find the misbehaving one. Looks like your UFS filesystem is not clean, which obviously happens on crash, but it could also be the root cause. Boot single user and repair it fsck -y /. Consider to enable soft update journaling or gjournal(8) as described in an article in the docs, if you do not already have this enabled.
 
Boot into single user and disable all non-essential tweaks. Then by manually invoking single one's sequentially, you will find the misbehaving one. Looks like your UFS filesystem is not clean, which obviously happens on crash, but it could also be the root cause. Boot single user and repair it fsck -y /. Consider to enable soft update journaling or gjournal(8) as described in an article in the docs, if you do not already have this enabled.

Hi

Ran commands 1) fsck -y, 2) mount -u /, 3) mount -a -t ufs, from single user boot. All the commands complete without any error! Issue still the same.

Not sure about gjournal.

Thanks
Ravi
 
According to the screenshot your filesystem is corrupt (several mangled entries).

All the commands complete without any error!
How old is this system? More specifically, how old is the harddrive?
 
tunefs -p will tell you if journaling is enabled. Disable all non-essential tweaks, invoke them manually one after the other. Then you'll find the bad one.
EDIT If you already have a corrupted HDD, some otherwise good service script, executable, library, kernel module file etc. may contain a bad sector. I.e. the filesystem looks clean, but a specific file does not contain the data it should. have a look at recoverdisk(1)
 
According to the screenshot your filesystem is corrupt (several mangled entries).


How old is this system? More specifically, how old is the harddrive?
Hi

The system, as well as hard drive is slightly more than 5 yrs old!

Thanks
Ravi
 
Your system is a running one. You may have some idea after what modification this error appeared just to guess. From the picture there may be some empty file or module. Try to use kernel from base if it is there.
 
Hi

Running fsck without options, 2 times, healed the error. Although the error still appears on the screen - I am able to login through multi-user.
Everything seems to run same as before!
Thanks
Ravi
 
Although the error still appears on the screen
Now provide the specific error which is still there. If it is serious or require further action gurus here may help you. Any error need to be attended first before going spiral. Enjoy FreeBSD.
 
Back
Top