[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [cobalt-users] /proc/kcore generating continually?
- Subject: Re: [cobalt-users] /proc/kcore generating continually?
- From: Greg Hewitt-Long <greg@xxxxxxxxxxxxxxxxxxx>
- Date: Fri Nov 23 21:54:04 2001
- List-id: Mailing list for users to share thoughts on Cobalt products. <cobalt-users.list.cobalt.com>
>On Fri, 23 Nov 2001, Greg Hewitt-Long wrote:
>
>> I've just noticed that on our RAQ3, there appears to be a kcore
>generating every minute in /proc....
>>
>> a couple of questions:
>>
>> 1. is this a kernel core dump?
> NO
>(fwiw, linux doesn't even support kernel core dumps, it's not an indy ;)
>> 2. where do I look to find out what's causing it?
>> 3. what are the most likely causes of this?
>
> /proc is a virtual filesystem that maps between things in the kernel and
>the outside world, kcore is a virtual file that represents the memory in
>the machine, leave it alone ;)
>
> > 4. should I be worried?
>
>Nope, it's *supposed* to be there
>
>> 5. it's 201,330,688 bytes big which is running the partitions at:
>>
>> /dev/hda1 743466 527432 216034 71% /
>> /dev/hda3 198601 17124 181477 9% /var
>> /dev/hda4 8597105 7148006 1449099 83% /home
>
>run 'mount' , you will see that /proc is it's own filesystem...
>
>type 'free' note how much free memory you have 'used', note the size of
>kcore...(total not including swap)
>
> /proc is it's own 'magic' filesystem, it doesn't use 'real' disk space
>>
>>
>> Is this likely to grow? If my guess is right, it's a kernel panic -
> and I'm seeing all memory and kernel activity dumped to the machine every
> minute - so I think I need to worry about this, as it's likely to thrash
>the disk to death...
>
>You keep deleting things in /proc and you ARE gonna cause a kernel panic
>;)
>
>No, it won't grow....it's not even really there ;)
>No, it's not a kernel panic ;)
>
>PS: kernel panics, by definition, do NOT write things to disk , the
>machine goes POOF ;)
>
>Unlike windows, unix's generally do NOT trust programs after they have
>made booboo's, when the kernel panics (eg causes an exception) the machine
>halts and won't usually update the disk with probably bad data..
>
thanks - I worked most of it out after posting - I did find the cobalt site completely devoid of useful information (what *HAS* happened to the knowledgebase - it's total garbage now!?!?!)
Thanks again - I should bone up on this stuff - my cobalt and linux knowledge aren't as great as my AIX * & HP/UX and all that's getting older and more out of date by the day.
Greg
--
http://www.webyourbusiness.com/
Providers of E-Commerce Software &
Web Design Consultancy and Services.
PH: (970)266-0195 FAX: (970)266-0158