[PATCH 2/2] makedumpfile: make the incomplete vmcore generated by ENOSPC error analyzable

Wang, Xiao/Wang Xiao wangx.fnst at cn.fujitsu.com
Tue Sep 16 19:41:28 PDT 2014


Hello Dave,

Thanks for your reply.

We can use two methods to indicate that the incomplete vmcore is a 
"fixed" one,
1. Use a flag in elf/kdump dumpfile(like "e_flags" in ELF header and 
"status"
    in disk_dump_header) to indicate it has been "fixed". But actually the
    kdump-compressed file needn't to use such a flag, for we just change the
    data write order of this format.
2. We can just let makedumpfile change the "fixed" dumpfile's filename 
automatically.
    Such as add a "-truncated" flag after those dumpfiles.

On 2014/9/16 21:20, Dave Anderson wrote:
> This patch bothers me -- how would the crash utility know that the "fixed"
> vmcore file was truncated/bogus to begin with?  Are you simply presuming
> that the user (when running makedumpfile) will change the output dumpfile
> name to something that indicates that it's incomplete?
>
> Typically incomplete vmcores fail to initialize due to some missing data,
> and it will confuse the issue if the PT_LOAD segments indicate that everything
> is just fine.  There really should be some kind of indication/flag/note in the
> compressed dumpfile or ELF header that it's a bogus/truncated vmcore.

-- 
Regards
Wang Xiao



More information about the kexec mailing list