Björn Steinbrink napsal(a):
On 2006.10.30 14:54:36 +0100, Jaroslav Tomecek wrote:
Hi,
1) I tried '/usr/sbin/vlimit -c 1000 --cpu 30'. It returned:
'vc_set_rlimit(): Invalid argument'
Any idea?
It's probably just not implemented/supported. I didn't check though. But
do you really _want_ that? It would limit your vserver to 30 seconds of
cpu time and kill it after it has used them up.
If you want to limit cpu usage you should rather take a look at vsched.
I'm testing 'vserver' for some organisation. They want me to test
everything ;-). Btw. I tried cpu scheduling. I set 1/5 of cpu for the
first and 4/5 of cpu for the second 'vserver'. Then I ran two identical
programs, each in one 'vserver'. It computed some floating stuff. After
each e.g. 1000 cycles it printed dot. I expected that there would be
five times more dots in the second vserver after some time. But the
numbers of dots were identical.
# cat /etc/vservers/test1/schedule
1
5
100
200
10
dummy
# cat /etc/vservers/test1/schedule
4
5
100
200
10
dummy
Is it the right way?
2) I compiled a program on FC6 using g++ (GCC) 4.1.1 20061011 (Red Hat
4.1.1-30) on host server:
#include <iostream>
using namespace std;
int main(void)
{
return 0;
}
But in FC5-based vserver 'test' it returned:
'[EMAIL PROTECTED] /]# ./test-programme
Floating point exception'
Maybe a glibc/libstdc++ conflict? No real idea about it... :(
Any idea?
/usr/sbin/vserver --version
vserver 0.30.210 -- manages the state of vservers
This program is part of util-vserver 0.30.210
An upgrade to .211 won't hurt, but it's probably not affecting the above
issues at all.
HTH
Björn
_______________________________________________
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
