I had a failure in a cache drive and decided to replace all drives in the pool.  I was able to mount the drives and pull the data off.  I then replaced the physical drives and rebooted.  Now I have no webgui.  When I SSH into the server, nginx reports as stopped as does PHP-fpm.  I have started both.  This allows me to get to the login screen.  However, after login, I get a blank white screen.  This result is the same in safe mode.  Diagnostics attached.

tower-diagnostics-20231113-1831.zip

Nov 13 18:06:39 Shadesmont kernel: traps: emhttpd[3565] trap divide error ip:55c8f41c299a sp:7fff380d01e0 error:0 in emhttpd[55c8f41b1000+24000]

 

emhttp is crashing, can you post the diags after booting in safe mode?

Here are the new logs.  Error remains the same.  I'm not entirely convinced it booted to safe mode, but that's what I selected on the boot menu.

tower-diagnostics-20231114-0913.zip

 

EDIT:  I removed the three cache drive cages and all is up again.  Now, to figure out why the cache drives are causing these errors....

 

EDIT #2: After regaining gui access, I was able to remove the cache pool entirely, shutdown, reinstall the new cache drives, restart, and recreate the cache pool.  All seems to be working now.

Edited November 14, 20232 yr by newtond

It booted in safe mode, but emhttp is still crashing, could be the pool, looks to me like the config is messed up, try renaming /boot/config/pool/cache.cfg to cache.bak and reboot.