On Fri, 7 Jul 2000, Jesse Pollard wrote:
>> Therefore, since some people WANT OOM killing to be done, and
>> others such as myself do NOT want it to be done, could someone in
>> the know of doing so, please make it a compile time or run time
>> tunable option? I'd like to tell my kernel "If an OOM condition
>> occurs, under absolutely *NO* circumstances are you to EVER kill
>> a running process".
>
>So you want your system to hang in a deadlock... forcing you to
>press the reset button to restart....
No, of course not. I don't want to lose my data though
either. How can I tell the kernel "never kill the programs
listed in this text file"? I can't. Not to my knowledge.
I'd like to be able to decide certain apps that never die, or
perhaps instead of certain apps.. certain 'conditions' that an
app can not be killed under.
For example.. if netscape is loaded, hasn't alloced memory in a
long time, or forced overcommited pages to be actually given,
then if OOM occurs, DONT KILL NETSCAPE.
I guess this problem is extremely complex though, so what I want
might very well be impossible.
>> I am *NOT* asking for this to be the default option, nor am I
>> asking that everyone else use it. I *FULLY* understand the need
>> to have a system like we have right now FOR SOME SYSTEMS, however
>> my system does not need it, and I suspect many other desktop
>> systems do not either. I'd rather have the application that is
>> hogging memory DIE than everything on my system by some
>> "smart" algorithm.
>>
>> Why not have the kernel just write a couple hundred megs of
>> random data to the root filesystem and reboot when detecting
>> OOM? Sounds about as logical to me..
>
>thats what happens, sort of, if the kernel isn't allowed to kill
>off some processes. Just fsck every filesystem that was mounted...
Sorry... I was just giving off some steam with sarcasm
there.. Wasn't aimed at anyone.. more like yelling out loud in a
room by yourself when Windows 95 bluescreens for no reason and
kills 4 hours work.
>> Then again, I just lost a bunch of work, so my current pissed off
>> opinion likely reflects that. Please keep that in mind while
>> flaming me with the wisdom of the current situation..
>>
>> GRRR!!!!
>>
>> Not trying to irk anyone, just letting off some steam.
>
>First time it happened to me I wanted global user resource limits on memory
>so that I would get notified when I was going over available memory. The
>quota should be about 5-10MB less than physical maximum (say 1-5% of total
>reserved for root). That way support daemons don't have to be killed
>(the X server for instance) and I know (and processes know) when the
>resources are no longer available.
Someone has mentioned bean-counters being a solution. Perhaps it
will... but it'll likely be 2 years before we find out. 2.4.0
isn't out yet, and could be 2 months before it is. Then it'll be
at least 18 months before 2.6.0 is out... ;o(
What might be nice instead is a daemon that sets off an alarm at
when a predetermined amount of VM is used up - say 80% or
so.. configureable of course. That might work good enough for me
for now.
-- Mike A. Harris Linux advocate Computer Consultant GNU advocate Capslock Consulting Open Source advocateI've overclocked my keyboard interface. It's quite messy dipping my hands into the mineral oil, but *MAN* is my keyboard ever fast now! - Anonymous Coward
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sat Jul 15 2000 - 21:00:08 EST