[Devel] Re: first stable release of OpenVZ kernel virtualization solution

[Kirill Korotaev]

> * Kirill Korotaev <dev at sw.ru> wrote:
>>>but what you have right now is an in essence swapless system, correct?  
>>>Do you support swapping at all in OVZ instances?
> 
>>Yes, swap is supported and processes are swapped in/out as in usual 
>>kernel. The only difference in comparison with std kernel is that UBC 
>>limits the amount of swappable memory VPS can have.
> 
> i mean, the only way to protect a high-prio instance against a low-prio 
> instance doing heavy swapout is by making the low-prio instance 
> swapless, correct? (either by not enabling it to swap at all, or by 
> tweaking the UBC limits in a way that can never lead to swapping).
correct. But not active/big swapping is ok, as it usually leads system 
to some equlibrium... only swap hog is bad.

> how about the 'dirty data creator' scenario: an instance filling up all 
> of the RAM with dirty data, at which point a highprio instance is 
> significantly impacted.
yes, this can be a problem which should be solved yet.
This can also be limited by UBC settings, but in general your point is 
valid.

> my point is, that such a swap or writeout related slowdown of a highprio 
> instance can be just as bad as a real DoS - and it brings us essentially 
> back to where we started with vserver. (and writeout related slowdowns 
> of unrelated instances cannot be avoided even with the most conservative 
> UBC settings, correct?)
We plan to use CFQv2 in some near future, but currently writeout is not 
controlled by UBC anyhow.
The only note is that currently used disk I/O scheduler (anticipatory) 
behaves quite well when one VPS is doing massive writes...
Disk I/O is a kind of problem for any of existing virtualization 
solutions and OpenVZ is not different here...

Kirill