On 02/20/14 at 12:01pm, Vivek Goyal wrote:
On Thu, Feb 20, 2014 at 02:00:54PM +0800, Baoquan He wrote:
kdump now defaults to only dump vmcore to the root partition in the /var/crash directory. This is problematic when /var is a separate file system because this makes saved vmcore hidden since one can not unmount /var even in single user mode. This also has the potential of blindly filling the root file system without out a clue as to why.
Now fix this by failing the loading of kdump kernel if default action is dump_to_rootfs while different disk is mounted on save path.
Hi Bao,
Few things.
- First of all, it does not matter what's the default failure action is. Kdump service should fail irrespective of default action.
So we just need focus the fix on default, no need to care about the default failure action which is specified by "default xxx".
- Our default dump target is root disk and path /var/crash. If we detect some disk mounted on /var/crash we fail.
Agree.
- We should do this checking *only* if user has not specified any disk and path in /etc/kdump.conf. If user has specified root disk in /etc/kdump.conf, then we should continue to dump.
Here user has not specificed any disk, but specified a path like "/home/user/test", and there's another disk mounted on this path, doesn't this need to be cared?
I think this should fail and give out en error message.
- We need to give a very clear message to user that why dump is failing and what user needs to do to avoid problem.
Sure, will think about the error message again.
Thanks Vivek