SliTaz Forum

RAM corrupted on kernel panic?

(5 posts) (3 voices)

• Started 1 year ago by obeliksz
• Latest reply from Filou
• This topic is not a support question

Tags:

No tags yet.

1. 

   obeliksz
   

   offline

   Member

   Hello,

   I had a Slitaz5 run from memory on 2 GB of RAM.

   After adding 2-3 packages, some locales and created a new image I tried to boot that and the system entered into kernel panic. I've done this procedure before, adding new packages to the image several times before but never encountered this problem. The rootfs.img.gz file was around 120Mb before and I dont think that it could have gotten larger than 300.

   I have seen kernel panic before so just rebooted thinking maybe the image file got corrupted, will have to redo what I've done. But after reboot could not enter even the bios settings, removed a ram slot and it booted again... into kernel panic. Now I guess both my rams got corrupted somehow from the faulty booting.

   Did anyone else also experienced something like this? I'm curious, does this happen if the image gets too large for the RAM, should I have been more cautious with this? This old laptop was falling apart anyway so it is not a big loss. Maybe it was about to fail anyway but this successive fall for both rams after kernel panic makes me doubt that.

   Posted 1 year ago #
2. 

   Ravenslock
   

   offline

   Member

   Hi, I've experience something similar with kernel panic on my Acer
   Aspire One Netbook when I tried out different Linux distros using
   the 'dd' command for copying the ISO file to my USB stick.

   This results in a black screen when I switch on the netbook or
   reboot it. Then I have to flash the BIOS to fix it.

   Posted 1 year ago #
3. 

   Filou
   

   offline

   Moderator

   Hi obeliksz,

   I experienced similar corruption events and identified two possible causes:
   1.) compression with lzma / xz in an older version caused corrupted files due to wrong length. I guess this should have been solved now, because recent problems related to 2.).
   2.) I remember to having had problems with packages that replaced /init (or /init.sh).
   While my working rootfs had a symbolic link pointing to /bin/busybox the replacing (in this case outdated) slitaz scripts messed up and lead to a kernel panic.

   Maybe you could investigate in this direction?

   Posted 1 year ago #
4. 

   obeliksz
   

   offline

   Member

   Thanks guys, for coming up with your experiences.

   After this kernel panic I'm not even able to get into the BIOS screen. I guess if I would put in a new RAM it would boot again, but I'm afraid it would also damage the newly inserted RAM slot. After my kernel panic it wasnt booting anymore, no bios screen, after I remove one of the two RAM cards, it booted, but after it reached the kernel panic again, it wouldn't boot anymore.

   Before this experience I didnt even thought that it is possible that the RAM to be damaged by just booting bad/corrupted stuff or too large image. I'm thinking of the RAM like that is always cleared empty after a shutdown but apparently the software, kernel loading can corrupt the RAM?

   Posted 1 year ago #
5. 

   Filou
   

   offline

   Moderator

   Frankly, I really doubt that!

   I'd rather believe that in that case the kernel panic is the result from - rather than the cause for - corrupt RAM modules. Maybe your RAM controller has been damaged due to some reason but I'd doubt that this could be caused by a kernel panic.

   The only ways to physically damage hardware by software is indirectly (by too much heat, for example by putting high computing strain on a processor and disabling the cooling fan // or by a short-circuit although I don't have an idea how to do that software-wise). None of which would indicate to a kernel panic, that stops any further computing since the root filesystem cannot be booted...

   Posted 1 year ago #

RSS feed for this topic

Reply

You must log in to post.