Re: > 15,000 Simultaneous Connections

Jesse Pollard (pollard@tomcat.admin.navo.hpc.mil)
Wed, 15 Sep 1999 17:13:28 -0500 (CDT)

jordy@wserv.com wrote:
>Mike Jagdis wrote:
>>
>> On Wed, 15 Sep 1999, Stephen C. Tweedie wrote:
>>
>> > The cost per fd is miniscule.
>>
>> Taking the 15000 fds in the subject select would need to do a
>> 4 page kmalloc and the current poll would do a 32 page kmalloc.
>> Then you have to loop over all the descriptors calling their poll
>> handlers.
>>
>> Incidentally, it looks like there is a possible DoS with queued
>> signals. If you flood the queue with signals but keep them blocked
>> no one else will get a look in.
>
>Yeah, per-user/per-process limits need to be implemented. In fact, there are a
>lot of resources which should have per-system, per-user, per-process, and
>per-group limits put in. Sounds like a good project for a someone.

change the per-group to per-project so that group access permissions don't
overlap resource limits (disk/cpu/memory).

-------------------------------------------------------------------------
Jesse I Pollard, II
Email: pollard@navo.hpc.mil

Any opinions expressed are solely my own.

-
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/