Solved Boot process stuck after reboot

It's a pity joshbsd you don't follow up completely. The size of the loader is the same, it makes sense to do a cksum on it too. Then check the same under FreeBSD directory as it may take precedence during boot.

But now please do comment out that kern.vty line from your loader.conf. It's very likely this is causing a problem (that's why I actually asked if you can confirm that device is up maybe but you see nothing on screen).

W.hâ/t Let's forget about that argument. We misunderstood each other.
However using loader from 12 on 13 caused some issues before.
 
_martin said:
It's a pity joshbsd you don't follow up completely. The size of the loader is the same, it makes sense to do a cksum on it too. Then check the same under FreeBSD directory as it may take precedence during boot.

But now please do comment out that kern.vty line from your loader.conf. It's very likely this is causing a problem (that's why I actually asked if you can confirm that device is up maybe but you see nothing on screen).

W.hâ/t Let's forget about that argument. We misunderstood each other.
However using loader from 12 on 13 caused some issues before.
Click to expand...
I did comment and put =vt but the login loop is still there
 
_martin We do misunderstand each other haha :)
I was just making the point that we've already tried with every line commented in loader.conf by loading the old kernel.
 
It's a joy to read stuff when it occurs later as you posted it. :)

You started the thread as if you did nothing but you should have come clean from the start - you did a lot. You changed handful of config files which means you did something more.
You can't login as your login shell is zsh which is not a default shell for users. You probably installed this shell from ports. Right now /bin/zsh is missing which by default one would expect it to be in /usr/local/bin/zsh as it would come from ports. You maybe created a symlink in /bin and now it doesn't exist or your /usr/ is not a proper one.

Login as root and change the shell of your user: pw user mod joshua -s /bin/sh.

You need to check what BE are you in with bectl show and quite possibly check what is actually mounted where (df -m)..
 
_martin said:
It's a joy to read stuff when it occurs later as you posted it. :)

You started the thread as if you did nothing but you should have come clean from the start - you did a lot. You changed handful of config files which means you did something more.
You can't login as your login shell is zsh which is not a default shell for users. You probably installed this shell from ports. Right now /bin/zsh is missing which by default one would expect it to be in /usr/local/bin/zsh as it would come from ports. You maybe created a symlink in /bin and now it doesn't exist or your /usr/ is not a proper one.

Login as root and change the shell of your user: pw user mod joshua -s /bin/sh.

You need to check what BE are you in with bectl show and quite possibly check what is actually mounted where (df -m)..
Click to expand...
I did not change any setting, I already told you that, I came clean, I was reading and installed dhcp client and yes now I remember I changed the shell from csh to zsh, is that suposed to break the all system, tell me please what setting did I. Change cause you say that I did alot, which one and for what purpose?

Is there a journal of activity in the system?

For the rest asked:

# cksum /tmp/mydata/boot/loader.conf
38991491763 171 /tmp/mydata/boot/loader.conf
# cksum /tmp/efi/efi/boot/bootx64.efi
2411978603 892928 /tmp/efi/efi/boot/bootx64.efi
 
Changes in loader.conf are not consistent with what you did (installed dhcp client). I personally legacy boot everything I own/support but it seems older syscons and EFI boot don't like each other (tested it in VM now). So somebody had to change it since your last boot as you didn't have this problem before. Same goes with missing zsh.

As a generic rule of thumb you should not change root's shell (unless you know what you're doing). Users that use non-standard shells (e.g. installed from ports) need to be aware that default installation of such shells is under /usr/local, most likely /usr/local/bin. Shell has to exist and be in /etc/shells for users to successfully login. Depending on the setup you may introduce complication with BE if you cross mounts and lose this shell (overwrite where certain dataset should be,etc.). Users without valid shell won't login as you experienced now.

Don't get me wrong, it's just what you mentioned you did and what you presented just doesn't add up. Anyway point is not to blame you but to navigate you through it.
There's no journal activity per say, it's a bit of investigation work to do. I'd do with these checks:

a) verify which BE are you in: bectl show
b) check what /usr is now: df -m /usr
c) check if state of the packages is what you expect: pkg info (check if zsh is installed)

Did you change your shell already? Are you able to login to WM (window manager) as before ?
 
_martin said:
Changes in loader.conf are not consistent with what you did (installed dhcp client). I personally legacy boot everything I own/support but it seems older syscons and EFI boot don't like each other (tested it in VM now). So somebody had to change it since your last boot as you didn't have this problem before. Same goes with missing zsh.

As a generic rule of thumb you should not change root's shell (unless you know what you're doing). Users that use non-standard shells (e.g. installed from ports) need to be aware that default installation of such shells is under /usr/local, most likely /usr/local/bin. Shell has to exist and be in /etc/shells for users to successfully login. Depending on the setup you may introduce complication with BE if you cross mounts and lose this shell (overwrite where certain dataset should be,etc.). Users without valid shell won't login as you experienced now.

Don't get me wrong, it's just what you mentioned you did and what you presented just doesn't add up. Anyway point is not to blame you but to navigate you through it.
There's no journal activity per say, it's a bit of investigation work to do. I'd do with these checks:

a) verify which BE are you in: bectl show
b) check what /usr is now: df -m /usr
c) check if state of the packages is what you expect: pkg info (check if zsh is installed)

Did you change your shell already? Are you able to login to WM (window manager) as before ?
Click to expand...
No, I didn't change anything yet, I was not aware that changing the shell was so complex, so it's this that break everything and changed the setting tty in loader.conf for sc that I swear I did not touch it once? Or may be have been hacked?

Now I wonder since I'm in the live CD all mounted is it possible to bring back csh as my default shell properly and be done with all that from there cause I doubt I will be able to login at all if I reboot in the login loop?
 
W.hâ/t said:
_martin We do misunderstand each other haha :)
I was just making the point that we've already tried with every line commented in loader.conf by loading the old kernel.
Click to expand...
Wrong... It reads the systemctl tunables. I did not know...
 
While everything is possible I don't think you got hacked just to get few tunables changed. More likely you don't remember changing it or you did copy-paste more stuff you remember.
You confused me a bit again with the live CD -- I thought you were using the other BE to check system. Anyway, yes, you can. There are more ways of doing this but probably most convinient way would be to mount the BE to a location, i.e. /mnt from life CD and do chroot /mnt. From there you can use the pw command I shared above.
 
Problem solved it was that line in loader.conf : Kern.vtty=sc which I can't know where it come from.. And for the login loop I chrooted my imported pool in live cd and restored the csh shell whit the help of _martin
 
To be honest joshbsd you are lucky the FreeBSD community is so friendly, you should really try to follow up properly (meaning: document every step you have taken between your posts when you come here asking for more help). Otherwise it makes it very difficult and confusing to help you, in most other communities people would have stopped trying to help you a long time ago.

It's totally understandable to not realize the impact of things such as changing the default shell (even though the handbook warns of the risks) but once people point out in a friendly manner that there is more going on than you have disclosed, you should make it your duty to try to remember everything you've done. This is also a way to respect the time people are investing to help you.

This thread is painful to watch because in all honestly you are abusing the kindness of those who are investing plenty of time to help you and explain you how FreeBSD works: if others invest time to help you, invest time to come out clean and follow up properly.

You increasingly look like someone who didn't read the handbook, copy-and-pasted a bunch of linux commands you didn't understand, broke a bunch of things in the process but is uninterested in understanding how FreeBSD works, and do not care to follow up properly. I feel you are crossing the line of disrespecting/abusing the time people are investing to help you.
 
reddy said:
To be honest joshbsd you are lucky the FreeBSD community is so friendly, you should really try to follow up properly (meaning: document every step you have taken between your posts when you come here asking for more help). Otherwise it makes it very difficult and confusing to help you, in most other communities people would have stopped trying to help you a long time ago.

It's totally understandable to not realize the impact of things such as changing the default shell (even though the handbook warns of the risks) but once people point out in a friendly manner that there is more going on than you have disclosed, you should make it your duty to try to remember everything you've done. This is also a way to respect the time people are investing to help you.

This thread is painful to watch because in all honestly you are abusing the kindness of those who are investing plenty of time to help you and explain you how FreeBSD works: if others invest time to help you, invest time to come out clean and follow up properly.

You increasingly look like someone who didn't read the handbook, copy-and-pasted a bunch of linux commands you didn't understand, broke a bunch of things in the process but is uninterested in understanding how FreeBSD works, and do not care to follow up properly. I feel you are crossing the line of disrespecting/abusing the time people are investing to help you.
Click to expand...
Have a nice day?
 
You must log in or register to reply here.
Back
Top