Re: [Vserver] cpu limits clone vservers
On Wed, Nov 24, 2004 at 05:01:47PM +1300, Sam Vilain wrote: Jörn Engel wrote: ...and the big challenge is - how do you apply this to memory usage? Oh, you could. But the general concept behind my unquoted list is a renewing resource. Network throughput is renewing. Network bandwidth usually isn't. With swapping, you can turn memory into cache and locality to the cpu is a renewable resource. Yep, that was my thought too. Memory seems like a static resource, so consider RSS used per second the renewable resource. Then you could charge tokens as normal. However, there are some tricky questions: 1) who do you charge shared memory (binaries etc) to ? 2) do you count mmap()'d regions in the buffercache? 3) if a process is sitting idle, but there is no VM contention, then they are using that memory more, so maybe they are using more fast memory tokens - but they might not really be occupying it, because it is not active. Maybe the thing with memory is that it's not important about how much is used per second, but more about how much active memory you are *displacing* per second into other places. We can find out from the VM subsystem how much RAM is displaced into swap by a context / process. It might also be possible for the MMU to pages to be swapped out can not easily be assigned to a context, this is different for pages getting paged in ... report how much L2/L3 cache is displaced during a given slice. I have a hunch that the best solution to the memory usage problem will have to take into account the multi-tiered nature of memory. So, I think it would be excellent to be able to penalise contexts that thrash the L3 cache. Systems with megabytes of L3 cache were designed to keep the most essential parts of most of the run queue hot - programs that thwart this by being bulky and excessively using pointers waste that cache. And then, it needs to all be done with no more than a few hundred cycles every reschedule. Hmm. Here's a thought about an algorithm that might work. This is all speculation without much regard to the existing implementations out there, of course. Season with grains of salt to taste. Each context is assigned a target RSS and VM size. Usage is counted a la disklimits (Herbert - is this already done?), but all complex yep, not relative as with disklimits, but absolute and in identical way the kernel accounts RSS and VM recalculation happens when somethings tries to swap something else out. As well as memory totals, each context also has a score that tracks how good or bad they've been with memory. Let's call that the Jabba value. When swap displacement occurs, it is first taken from disproportionately fat jabbas that are running on nearby CPUs (for NUMA). Displacing other's memory makes your context a fatter jabba too, but taking from jabbas that are already fat is not as bad as taking it from a hungry jabba. When someone takes your memory, that makes you a thinner jabba. This is not the same as simply a ratio of your context's memory usage to the allocated amount. Depending on the functions used to alter the jabba value, it should hopefully end up measuring something more akin to the amount of system memory turnover a context is inducing. It might also need something to act as a damper to pull a context's jabba nearer towards the zero point during lulls of VM activity. Then, if you are a fat jabba, maybe you might end up getting rescheduled instead of getting more memory whenever you want it! thought about a simpler approach, with a TB for the actual page-ins, so that every page-in will consume a token, and you get a number per interval, as usual ... best, Herbert -- Sam Vilain, sam /\T vilain |T net, PGP key ID: 0x05B52F13 (include my PGP key ID in personal replies to avoid spam filtering) ___ Vserver mailing list [EMAIL PROTECTED] http://list.linux-vserver.org/mailman/listinfo/vserver ___ Vserver mailing list [EMAIL PROTECTED] http://list.linux-vserver.org/mailman/listinfo/vserver
Re: [Vserver] cpu limits clone vservers
On Wed, 24 November 2004 14:02:07 +0100, Herbert Poetzl wrote: On Wed, Nov 24, 2004 at 05:01:47PM +1300, Sam Vilain wrote: pages to be swapped out can not easily be assigned to a context, this is different for pages getting paged in ... Or any page-fault done for the context, for that matter. There is no fundamental difference between swapping in, faulting in the recently-started openoffice, mallocmemset, ... Here's a thought about an algorithm that might work. This is all speculation without much regard to the existing implementations out there, of course. Season with grains of salt to taste. Each context is assigned a target RSS and VM size. Usage is counted a la disklimits (Herbert - is this already done?), but all complex yep, not relative as with disklimits, but absolute and in identical way the kernel accounts RSS and VM recalculation happens when somethings tries to swap something else out. As well as memory totals, each context also has a score that tracks how good or bad they've been with memory. Let's call that the Jabba value. When swap displacement occurs, it is first taken from disproportionately fat jabbas that are running on nearby CPUs (for NUMA). Displacing other's memory makes your context a fatter jabba too, but taking from jabbas that are already fat is not as bad as taking it from a hungry jabba. When someone takes your memory, that makes you a thinner jabba. This is not the same as simply a ratio of your context's memory usage to the allocated amount. Depending on the functions used to alter the jabba value, it should hopefully end up measuring something more akin to the amount of system memory turnover a context is inducing. It might also need something to act as a damper to pull a context's jabba nearer towards the zero point during lulls of VM activity. Then, if you are a fat jabba, maybe you might end up getting rescheduled instead of getting more memory whenever you want it! thought about a simpler approach, with a TB for the actual page-ins, so that every page-in will consume a token, and you get a number per interval, as usual ... It misses a few corner-cases, but I cannot think of anything better. More complicated approaches would miss different corner-cases, but that's not a real advantage. With the simple TB approach, any IO caused on behalf of a process gets accounted. Updatedb would be a huge offender, but that looks more like a feature than a bug. Jörn -- Victory in war is not repetitious. -- Sun Tzu ___ Vserver mailing list [EMAIL PROTECTED] http://list.linux-vserver.org/mailman/listinfo/vserver
Re: [Vserver] cpu limits clone vservers
On Wed, 24 November 2004 15:58:01 +1300, Sam Vilain wrote: If you increase the priority of the administrative daemons to -19, then you will get what people *actually* want ;-) which is CPU time for administration to come before all normal activity on the system. Even with a load over 100 you should still be able to log in via ssh if it is niced that high. Sounds interesting. Last time I tried my forkbomb on 2.6, it took me a full hour before I finally gave up and rebootet the system. But that was without any ulimit enforcement. Jörn -- Fantasy is more important than knowledge. Knowledge is limited, while fantasy embraces the whole world. -- Albert Einstein ___ Vserver mailing list [EMAIL PROTECTED] http://list.linux-vserver.org/mailman/listinfo/vserver
Re: [Vserver] cpu limits clone vservers
On Wed, 24 Nov 2004, Herbert Poetzl wrote:
Then, if you are a fat jabba, maybe you might end up getting rescheduled
instead of getting more memory whenever you want it!
thought about a simpler approach, with a TB for the
actual page-ins, so that every page-in will consume
a token, and you get a number per interval, as usual ...
There probably still needs to be a target size, which if exceeded, your
bucket is refilled slower. This way small contexts would not be suffering
because of a large and very active context. The sysadmins would need to
make sure that the sum of all targets does not exceed physical RAM.
So you'd have two additional parameters - target size and fill-interval
multiplier.
if (is_exceeded(target)) {
interval *= multplier;
}
Also - at which point does a malloc actually fail? It seems like context 0
should have a priority over other contexts - a non-0 context should under
no circumstances be able to exhaust the system memory.
May be there should be an additional level in the bucket - reschedule
level. If I actually empty the bucket, the malloc fails?
my $0.02
Grisha
___
Vserver mailing list
[EMAIL PROTECTED]
http://list.linux-vserver.org/mailman/listinfo/vserver
Re: [Vserver] cpu limits clone vservers
On Wed, Nov 24, 2004 at 11:00:40AM -0500, Gregory (Grisha) Trubetskoy wrote:
On Wed, 24 Nov 2004, Herbert Poetzl wrote:
Then, if you are a fat jabba, maybe you might end up getting rescheduled
instead of getting more memory whenever you want it!
thought about a simpler approach, with a TB for the
actual page-ins, so that every page-in will consume
a token, and you get a number per interval, as usual ...
There probably still needs to be a target size, which if exceeded, your
bucket is refilled slower. This way small contexts would not be suffering
because of a large and very active context. The sysadmins would need to
make sure that the sum of all targets does not exceed physical RAM.
So you'd have two additional parameters - target size and fill-interval
multiplier.
well, actually I was more thinking of using something
similar to the scheduler TB (see the vserver paper for
details) and just do the check/dec for every 'page'
if (is_exceeded(target)) {
interval *= multplier;
}
Also - at which point does a malloc actually fail? It seems like context 0
should have a priority over other contexts - a non-0 context should under
no circumstances be able to exhaust the system memory.
well, malloc is something which will fail if RSS or
VM limits are reached (actually it's brk() and mmap())
but in the typical case the page TB would just cause
the context to be not scheduled (i.e. a page in would
take a _lot_ longer than usual ;)
best,
Herbert
May be there should be an additional level in the bucket - reschedule
level. If I actually empty the bucket, the malloc fails?
my $0.02
Grisha
___
Vserver mailing list
[EMAIL PROTECTED]
http://list.linux-vserver.org/mailman/listinfo/vserver
___
Vserver mailing list
[EMAIL PROTECTED]
http://list.linux-vserver.org/mailman/listinfo/vserver
[Vserver] cpu limits clone vservers
Hi,
I am testing the linux vservers development release 1.9.3.
I have 2 problems:
1. When I check the limits with vlimit I can't see any cpu resource:
[EMAIL PROTECTED] linux-2.6.9]# vlimit -c 49167 --all -d
RSS N/A N/A 10
NPROC N/A N/A 70
NOFILE N/A N/A inf
MEMLOCK N/A N/A inf
AS N/A N/A inf
Though :
[EMAIL PROTECTED] util-vserver]# vlimit -c 49168 --cpu 30
vc_set_rlimit(): Success
If I run a process that does only while(1){} inside the vserver, the
cpu is used only 25%-30%. So the limitations work but not how I want
them to. No matter how much I set the cpu limit (10,20,50,100,200...)
and no matter how many processes I have in the vserver the cpu is used
25-30%.
I think I missed a piece of the puzzle.
2. How can I clone a vserver? On 2.4.x I had the newvserver utility.
On 2.6 I have vbuild and vcopy but neither of them works.
[EMAIL PROTECTED] util-vserver]# ./vbuild --debug --test andreea andreea_clone
sh: line 1: /usr/lib/util-vserver/distrib-info: No such file or
directory
Can't lstat file andreea/.vdir (No such file or directory)
[EMAIL PROTECTED] util-vserver]# ./vcopy andreea_clone andreea
unification not configured for source vserver
Reading the error I get at vcopy I understand that vcopy creates vserver
using unification. I don't want unification. I want every vserver to
have it's own logical volume, thus I can limit the space for every
vserver very easy.
I think vbuild is what I want but it's not working. Is there another
utility I don't know about? Or how can I make vbuild work?
Thanks a lot.
--
Andreea Gansac
Web Engineer
iNES Group
tel: 021-232.21.12
fax: 021-232.34.61
www.ines.ro
___
Vserver mailing list
[EMAIL PROTECTED]
http://list.linux-vserver.org/mailman/listinfo/vserver
Re: [Vserver] cpu limits clone vservers
On Tue, 23 Nov 2004, Andreea Gansac wrote:
[EMAIL PROTECTED] util-vserver]# vlimit -c 49168 --cpu 30
vc_set_rlimit(): Success
If I run a process that does only while(1){} inside the vserver, the
cpu is used only 25%-30%.
If I'm not mistaken, this simply sets the cpu time to 30 seconds, so after
30 seconds of cpu time is used, processes in your context will be killed.
Take a look at this thread, it descibes what you want. (Read the whole
thread, because the first message from me has some ommissions):
http://list.linux-vserver.org/archive/vserver/msg08134.html
Reading the error I get at vcopy I understand that vcopy creates vserver
using unification. I don't want unification. I want every vserver to
have it's own logical volume, thus I can limit the space for every
vserver very easy.
I think vbuild is what I want but it's not working. Is there another
utility I don't know about? Or how can I make vbuild work?
You can limit the space much easier using the VServer disk limits. google
for vserver vdlimit. Basically you need xid tagging enabled in the kernel
(under VServer menu option in kernel config, off by default), need to
compile the vdlimit tool, then the partition on which vservers reside
needs to be mounted with the tagxid option, then you can set a limit like
this:
/usr/local/vdlimit-0.01/vdlimit -a -x 1 \
-S 0,10,0,1,5 /vservers
This means that for context , 0 space is presently used, 10 is
maximum allowed, 0 inodes presently used, 1 inodes maximum allowed, 5%
of disk space is reserved for root. Note that these limits exist only
while the serer is up and therefore need to be saved on shutdown and
restored on startup. The list archives have example scripts of how people
do this.
Grisha
___
Vserver mailing list
[EMAIL PROTECTED]
http://list.linux-vserver.org/mailman/listinfo/vserver
Re: [Vserver] cpu limits clone vservers
On Tue, Nov 23, 2004 at 03:00:06PM +0200, Andreea Gansac wrote:
Hi,
I am testing the linux vservers development release 1.9.3.
I have 2 problems:
1. When I check the limits with vlimit I can't see any cpu resource:
[EMAIL PROTECTED] linux-2.6.9]# vlimit -c 49167 --all -d
RSS N/A N/A 10
NPROC N/A N/A 70
NOFILE N/A N/A inf
MEMLOCK N/A N/A inf
AS N/A N/A inf
Though :
[EMAIL PROTECTED] util-vserver]# vlimit -c 49168 --cpu 30
vc_set_rlimit(): Success
we could do CPU limits (similar to ulimit) but would
you really want to limit a vserver to, let's say 1minute
of CPU usage in total?
If I run a process that does only while(1){} inside the vserver, the
cpu is used only 25%-30%. So the limitations work but not how I want
them to. No matter how much I set the cpu limit (10,20,50,100,200...)
and no matter how many processes I have in the vserver the cpu is used
25-30%.
I think I missed a piece of the puzzle.
yep, probably the hard scheduler, the token bucket
and the vsched tool ...
2. How can I clone a vserver? On 2.4.x I had the newvserver utility.
On 2.6 I have vbuild and vcopy but neither of them works.
[EMAIL PROTECTED] util-vserver]# ./vbuild --debug --test andreea andreea_clone
sh: line 1: /usr/lib/util-vserver/distrib-info: No such file or
directory
Can't lstat file andreea/.vdir (No such file or directory)
[EMAIL PROTECTED] util-vserver]# ./vcopy andreea_clone andreea
unification not configured for source vserver
Reading the error I get at vcopy I understand that vcopy creates vserver
using unification. I don't want unification. I want every vserver to
have it's own logical volume, thus I can limit the space for every
vserver very easy.
I think vbuild is what I want but it's not working. Is there another
utility I don't know about? Or how can I make vbuild work?
I think vserver name build ... is what you want ...
HTH,
Herbert
Thanks a lot.
--
Andreea Gansac
Web Engineer
iNES Group
tel: 021-232.21.12
fax: 021-232.34.61
www.ines.ro
___
Vserver mailing list
[EMAIL PROTECTED]
http://list.linux-vserver.org/mailman/listinfo/vserver
___
Vserver mailing list
[EMAIL PROTECTED]
http://list.linux-vserver.org/mailman/listinfo/vserver
Re: [Vserver] cpu limits clone vservers
On Tue, 23 November 2004 16:44:22 +0100, Herbert Poetzl wrote: we could do CPU limits (similar to ulimit) but would you really want to limit a vserver to, let's say 1minute of CPU usage in total? That's basically the same problem as with any shared resource consumption. For networking, HTB is relatively close to what most people want and I don't see how CPU is a much different resource. What most people want in plain English: o Every user gets some guaranteed lower bound. o Sum of lower bounds doesn't exceed total resources. o Most of the time, not all resources get consumed. Add them to the 'leftover' pool. o Users that demand more resources than their lower bound get serviced from the leftover pool. o Users that, on average, use less resources get a higher priority when accessing the leftover pool. List could be longer, but everything else is details. Most controversy will be over the question of how exactly to prioritize the nicer users. But in the end, CPU-hogs will be limited to something close to their lower bounds and nice users operate well below but can get a lot more power in a burst, as least sometimes. Yeah, code doesn't exist. The usual. Jörn -- He who knows others is wise. He who knows himself is enlightened. -- Lao Tsu ___ Vserver mailing list [EMAIL PROTECTED] http://list.linux-vserver.org/mailman/listinfo/vserver
Re: [Vserver] cpu limits clone vservers
On Tue, Nov 23, 2004 at 06:18:54PM +0100, Jörn Engel wrote: On Tue, 23 November 2004 16:44:22 +0100, Herbert Poetzl wrote: we could do CPU limits (similar to ulimit) but would you really want to limit a vserver to, let's say 1minute of CPU usage in total? That's basically the same problem as with any shared resource consumption. For networking, HTB is relatively close to what most people want and I don't see how CPU is a much different resource. What most people want in plain English: o Every user gets some guaranteed lower bound. o Sum of lower bounds doesn't exceed total resources. o Most of the time, not all resources get consumed. Add them to the 'leftover' pool. o Users that demand more resources than their lower bound get serviced from the leftover pool. o Users that, on average, use less resources get a higher priority when accessing the leftover pool. List could be longer, but everything else is details. Most controversy will be over the question of how exactly to prioritize the nicer users. But in the end, CPU-hogs will be limited to something close to their lower bounds and nice users operate well below but can get a lot more power in a burst, as least sometimes. Yeah, code doesn't exist. The usual. ahem, maybe you should read up on the TokenBucket stuff for CPU usage in linux-vserver ... http://linux-vserver.org/Linux-VServer-Paper-06 06.3. Token Bucket Extensions or do you mean something different? if not, then it's already implemented ;) best, Herbert PS: what about linux-vserver CoW? Jörn -- He who knows others is wise. He who knows himself is enlightened. -- Lao Tsu ___ Vserver mailing list [EMAIL PROTECTED] http://list.linux-vserver.org/mailman/listinfo/vserver ___ Vserver mailing list [EMAIL PROTECTED] http://list.linux-vserver.org/mailman/listinfo/vserver
Re: [Vserver] cpu limits clone vservers
On Tue, 23 Nov 2004, [iso-8859-1] J?rn Engel wrote: What most people want in plain English: o Every user gets some guaranteed lower bound. o Sum of lower bounds doesn't exceed total resources. o Most of the time, not all resources get consumed. Add them to the 'leftover' pool. o Users that demand more resources than their lower bound get serviced from the leftover pool. o Users that, on average, use less resources get a higher priority when accessing the leftover pool. ...and the big challenge is - how do you apply this to memory usage? Grisha ___ Vserver mailing list [EMAIL PROTECTED] http://list.linux-vserver.org/mailman/listinfo/vserver
Re: [Vserver] cpu limits clone vservers
...and the big challenge is - how do you apply this to memory usage? oooh! I know! swap out those that are over limit to tape! -- Key fingerprint = 40D0 9FFB 9939 7320 8294 05E0 BCC7 02C4 75CC 50D9 ___ Vserver mailing list [EMAIL PROTECTED] http://list.linux-vserver.org/mailman/listinfo/vserver
Re: [Vserver] cpu limits clone vservers
On Tue, 2004-11-23 at 12:47 -0500, Gregory (Grisha) Trubetskoy wrote: On Tue, 23 Nov 2004, [iso-8859-1] J?rn Engel wrote: What most people want in plain English: o Every user gets some guaranteed lower bound. o Sum of lower bounds doesn't exceed total resources. o Most of the time, not all resources get consumed. Add them to the 'leftover' pool. o Users that demand more resources than their lower bound get serviced from the leftover pool. o Users that, on average, use less resources get a higher priority when accessing the leftover pool. ...and the big challenge is - how do you apply this to memory usage? Grisha This would be a cool thing. We could squabble over the details, but if it did what you said above and left some room for tweaking I'll bet people would be [even more] pleased [though we are already ecstatic now] with the vserver project. Of course, I'm still using CTX 17, so I'm pretty easy to please I guess. I'd be curious to know what happens when there is contention for that pool of RAM. I've got a nightly batch job that lasts 15 minutes but uses most of the server's ram during the process. Right now, everything works OK, but I suspect under this vserver panacea edition I would have problems because idle vservers will be allocated their minimum ram even though they don't need it. I guess I could just allocate 4MB of RAM as the minimum or some other small number to get the effect of what I do now... still, a bit thought provoking. Keep up the interesting conversation and work, -- Matthew Nuzum | Makers of Elite Content Management System www.followers.net | View samples of Elite CMS in action [EMAIL PROTECTED] | http://www.followers.net/portfolio/ ___ Vserver mailing list [EMAIL PROTECTED] http://list.linux-vserver.org/mailman/listinfo/vserver
Re: [Vserver] cpu limits clone vservers
On Tue, 23 November 2004 19:08:50 +0100, Jörn Engel wrote: I love it when someone else already did the work. ;) Except when it's only partial. If implementation matches documentation, the fixed lower bound is 0 (zero). That's pretty low. Most people want to say something like Ssh will always get 5% of cpu, no matter how many forkbombs explode. And the administrator's shell will inherit those 5%. Ok, not many people know they want to say it, but some may learn the hard way over time. ;) Jörn PS: [EMAIL PROTECTED]:/tmp cat _ head __ . _. _ [EMAIL PROTECTED]:/tmp . _ Have fun! -- Fancy algorithms are slow when n is small, and n is usually small. Fancy algorithms have big constants. Until you know that n is frequently going to be big, don't get fancy. -- Rob Pike ___ Vserver mailing list [EMAIL PROTECTED] http://list.linux-vserver.org/mailman/listinfo/vserver
Re: [Vserver] cpu limits clone vservers
On Tue, Nov 23, 2004 at 07:39:26PM +0100, Jörn Engel wrote: On Tue, 23 November 2004 19:08:50 +0100, Jörn Engel wrote: I love it when someone else already did the work. ;) Except when it's only partial. If implementation matches documentation, the fixed lower bound is 0 (zero). That's pretty low. Most people want to say something like Ssh will always get 5% of cpu, no matter how many forkbombs explode. And the administrator's shell will inherit those 5%. Ok, not many people know they want to say it, but some may learn the hard way over time. ;) yep, but taking care that overbooking doesn't happen can be done in userspace, literally ... so a 'minimum' of available resources can be guaranteeed only if you limit all other contexts to 1.0 - Sum[max], which in turn, is sufficient unless you have 'better' suggestions to solve this ... best, Herbert Jörn PS: [EMAIL PROTECTED]:/tmp cat _ head __ . _. _ [EMAIL PROTECTED]:/tmp . _ Have fun! -- Fancy algorithms are slow when n is small, and n is usually small. Fancy algorithms have big constants. Until you know that n is frequently going to be big, don't get fancy. -- Rob Pike ___ Vserver mailing list [EMAIL PROTECTED] http://list.linux-vserver.org/mailman/listinfo/vserver ___ Vserver mailing list [EMAIL PROTECTED] http://list.linux-vserver.org/mailman/listinfo/vserver
Re: [Vserver] cpu limits clone vservers
Jörn Engel wrote: ...and the big challenge is - how do you apply this to memory usage? Oh, you could. But the general concept behind my unquoted list is a renewing resource. Network throughput is renewing. Network bandwidth usually isn't. With swapping, you can turn memory into cache and locality to the cpu is a renewable resource. Yep, that was my thought too. Memory seems like a static resource, so consider RSS used per second the renewable resource. Then you could charge tokens as normal. However, there are some tricky questions: 1) who do you charge shared memory (binaries etc) to ? 2) do you count mmap()'d regions in the buffercache? 3) if a process is sitting idle, but there is no VM contention, then they are using that memory more, so maybe they are using more fast memory tokens - but they might not really be occupying it, because it is not active. Maybe the thing with memory is that it's not important about how much is used per second, but more about how much active memory you are *displacing* per second into other places. We can find out from the VM subsystem how much RAM is displaced into swap by a context / process. It might also be possible for the MMU to report how much L2/L3 cache is displaced during a given slice. I have a hunch that the best solution to the memory usage problem will have to take into account the multi-tiered nature of memory. So, I think it would be excellent to be able to penalise contexts that thrash the L3 cache. Systems with megabytes of L3 cache were designed to keep the most essential parts of most of the run queue hot - programs that thwart this by being bulky and excessively using pointers waste that cache. And then, it needs to all be done with no more than a few hundred cycles every reschedule. Hmm. Here's a thought about an algorithm that might work. This is all speculation without much regard to the existing implementations out there, of course. Season with grains of salt to taste. Each context is assigned a target RSS and VM size. Usage is counted a la disklimits (Herbert - is this already done?), but all complex recalculation happens when somethings tries to swap something else out. As well as memory totals, each context also has a score that tracks how good or bad they've been with memory. Let's call that the Jabba value. When swap displacement occurs, it is first taken from disproportionately fat jabbas that are running on nearby CPUs (for NUMA). Displacing other's memory makes your context a fatter jabba too, but taking from jabbas that are already fat is not as bad as taking it from a hungry jabba. When someone takes your memory, that makes you a thinner jabba. This is not the same as simply a ratio of your context's memory usage to the allocated amount. Depending on the functions used to alter the jabba value, it should hopefully end up measuring something more akin to the amount of system memory turnover a context is inducing. It might also need something to act as a damper to pull a context's jabba nearer towards the zero point during lulls of VM activity. Then, if you are a fat jabba, maybe you might end up getting rescheduled instead of getting more memory whenever you want it! -- Sam Vilain, sam /\T vilain |T net, PGP key ID: 0x05B52F13 (include my PGP key ID in personal replies to avoid spam filtering) ___ Vserver mailing list [EMAIL PROTECTED] http://list.linux-vserver.org/mailman/listinfo/vserver
