fedora 16 with xen balloning.

[itamar]

hi.

I have a fedora 16 with all updates.

when I run xm mem-set 0 1024 the machines crashes after some seconds.

[Konrad Rzeszutek Wilk]

On Wed, Sep 28, 2011 at 04:16:52PM -0300, Itamar Reis Peixoto wrote:
> hi.
>
> I have a fedora 16 with all updates.
>
> when I run xm mem-set 0 1024 the machines crashes after some seconds.

What is the crash (serial log please)? How much memory do you have initially?
Does it crash if you have 'dom0_mem=max:2G' on your Xen hypervisor line?

Does it crash if you use 'xl mem-set 0 1024'?

[itamar]

On Wed, Sep 28, 2011 at 4:21 PM, Konrad Rzeszutek Wilk
 wrote:
> On Wed, Sep 28, 2011 at 04:16:52PM -0300, Itamar Reis Peixoto wrote:
> > hi.
> >
> > I have a fedora 16 with all updates.
> >
> > when I run xm mem-set 0 1024 the machines crashes after some seconds.
>
> What is the crash (serial log please)? How much memory do you have initially?
> Does it crash if you have 'dom0_mem=max:2G' on your Xen hypervisor line?
>
> Does it crash if you use 'xl mem-set 0 1024'?

[root@localhost ~]# uname -a
Linux localhost.localdomain 3.1.0-0.rc6.git0.3.fc16.x86_64 #1 SMP Fri
Sep 16 12:26:22 UTC 2011 x86_64 x86_64 x86_64 GNU/Linux

[root@localhost ~]# xm list
Name                                        ID   Mem VCPUs      State   Time(s)
Domain-0                                     0  3584     4     r-----     40.4
[root@localhost ~]#

[root@localhost ~]# xl mem-set 0 1024
libxl: error: libxl.c:2114:libxl_set_memory_target cannot get memory
info from /local/domain/0/memory/static-max
: No such file or directory
[root@localhost ~]#

had to reboot after xm mem-set 0 1024

[itamar]

On Wed, Sep 28, 2011 at 4:32 PM, Itamar Reis Peixoto
 wrote:
> On Wed, Sep 28, 2011 at 4:21 PM, Konrad Rzeszutek Wilk
>  wrote:
> > On Wed, Sep 28, 2011 at 04:16:52PM -0300, Itamar Reis Peixoto wrote:
> > > hi.
> > >
> > > I have a fedora 16 with all updates.
> > >
> > > when I run xm mem-set 0 1024 the machines crashes after some seconds.
> >
> > What is the crash (serial log please)? How much memory do you have initially?
> > Does it crash if you have 'dom0_mem=max:2G' on your Xen hypervisor line?
> >
> > Does it crash if you use 'xl mem-set 0 1024'?
>
> [root@localhost ~]# uname -a
> Linux localhost.localdomain 3.1.0-0.rc6.git0.3.fc16.x86_64 #1 SMP Fri
> Sep 16 12:26:22 UTC 2011 x86_64 x86_64 x86_64 GNU/Linux
>
> [root@localhost ~]# xm list
> Name                                        ID   Mem VCPUs      State   Time(s)
> Domain-0                                     0  3584     4     r-----     40.4
> [root@localhost ~]#
>
> [root@localhost ~]# xl mem-set 0 1024
> libxl: error: libxl.c:2114:libxl_set_memory_target cannot get memory
> info from /local/domain/0/memory/static-max
> : No such file or directory
> [root@localhost ~]#
>
> had to reboot after xm mem-set 0 1024

also virt-manager doesn't recognize it as dom0, I see some messages in log

[   22.531841] XENBUS: Unable to read cpu state
[   22.531981] XENBUS: Unable to read cpu state
[   22.532213] XENBUS: Unable to read cpu state
[   22.532350] XENBUS: Unable to read cpu state

[Jerry Amundson]

On Wed, Sep 28, 2011 at 3:11 PM, Itamar Reis Peixoto <
[e-mail]> wrote:
> On Wed, Sep 28, 2011 at 4:32 PM, Itamar Reis Peixoto
>  wrote:
> > On Wed, Sep 28, 2011 at 4:21 PM, Konrad Rzeszutek Wilk
> >  wrote:
> > > On Wed, Sep 28, 2011 at 04:16:52PM -0300, Itamar Reis Peixoto wrote:
> > > > hi.
> > > >
> > > > I have a fedora 16 with all updates.
> > > >
> > > > when I run xm mem-set 0 1024 the machines crashes after some seconds.

xm mem-set 0 1536
worked fine for me.
(My KDE workstation, so 1024 is too little RAM to do day-job :-)
> > > What is the crash (serial log please)? How much memory do you have
>
> initially?
> > > Does it crash if you have 'dom0_mem=max:2G' on your Xen hypervisor line?
> > >
> > > Does it crash if you use 'xl mem-set 0 1024'?
> >
> > [root@localhost ~]# uname -a
> > Linux localhost.localdomain 3.1.0-0.rc6.git0.3.fc16.x86_64 #1 SMP Fri
> > Sep 16 12:26:22 UTC 2011 x86_64 x86_64 x86_64 GNU/Linux
> >
> > [root@localhost ~]# xm list
> > Name                                        ID   Mem VCPUs      State
>
> Time(s)
> > Domain-0                                     0  3584     4     r-----
>
> 40.4
> > [root@localhost ~]#
> >
> > [root@localhost ~]# xl mem-set 0 1024
> > libxl: error: libxl.c:2114:libxl_set_memory_target cannot get memory
> > info from /local/domain/0/memory/static-max
> > : No such file or directory
> > [root@localhost ~]#
> >
> > had to reboot after xm mem-set 0 1024

You previously said crash - which is it? They are two different things.
Did you wait a fair amount of time?
Check for disk activity? It may just need some swapping...
> also virt-manager doesn't recognize it as dom0, I see some messages in log
>
> [   22.531841] XENBUS: Unable to read cpu state
> [   22.531981] XENBUS: Unable to read cpu state
> [   22.532213] XENBUS: Unable to read cpu state
> [   22.532350] XENBUS: Unable to read cpu state

virt-manager does not log to messages, so those two things are unrelated.
What happens when you try virt-manager?
What is in /root/.virt-manager/virt-manager.log?

jerry

[Konrad Rzeszutek Wilk]

> > > What is the crash (serial log please)? How much memory do you have initially?
> > > Does it crash if you have 'dom0_mem=max:2G' on your Xen hypervisor line?

^^^^^^^^^^^^^^^
Did that make a change?
> > > Does it crash if you use 'xl mem-set 0 1024'?
> >
> > [root@localhost ~]# uname -a
> > Linux localhost.localdomain 3.1.0-0.rc6.git0.3.fc16.x86_64 #1 SMP Fri
> > Sep 16 12:26:22 UTC 2011 x86_64 x86_64 x86_64 GNU/Linux
> >
> > [root@localhost ~]# xm list
> > Name                                        ID   Mem VCPUs      State   Time(s)
> > Domain-0                                     0  3584     4     r-----     40.4
> > [root@localhost ~]#
> >
> > [root@localhost ~]# xl mem-set 0 1024
> > libxl: error: libxl.c:2114:libxl_set_memory_target cannot get memory
> > info from /local/domain/0/memory/static-max
> > : No such file or directory

Ok, did 'xl list' show a new value?  What does this command give you:

find /sys -name target_kb | xargs cat
> > [root@localhost ~]#
> >
> > had to reboot after xm mem-set 0 1024

Why?Did it crash? Was it running to slow? If you ran:
while (true)
do
 xl list | grep Domain-0
done

Did you see the numbers go up/down? Can you be more specific please?
> >
>
> also virt-manager doesn't recognize it as dom0, I see some messages in log

Hm, as in.. you can't see Xen in it? If so, we need to rope in the
Xen libvirt expert from Novell. Can you send another email with details
and I will (once I recall his name) CC him.
> [   22.531841] XENBUS: Unable to read cpu state
> [   22.531981] XENBUS: Unable to read cpu state
> [   22.532213] XENBUS: Unable to read cpu state
> [   22.532350] XENBUS: Unable to read cpu state

Ok, those are a different issue - and you can ignore that.
> --
>
> Itamar Reis Peixoto
> msn, google talk: [e-mail]
> [phone] (FIXO SP)
> [phone] (TIM)
> [phone] (OI)
> [phone] (FIXO MG)

[itamar]

On Wed, Sep 28, 2011 at 6:18 PM, Konrad Rzeszutek Wilk
 wrote:
> Ok, did 'xl list' show a new value?  What does this command give you:
>
> find /sys -name target_kb | xargs cat

[root@localhost ~]# find /sys -name target_kb | xargs cat
3670016

while (true)
 do
  xl list | grep Domain-0
 done

I am able to see the numbers going down, and after some time stop.(freeze)
> Did you see the numbers go up/down? Can you be more specific please?
> > >

I have ordered a serial adapter to have the logs next time.

[Konrad Rzeszutek Wilk]

On Wed, Sep 28, 2011 at 06:33:52PM -0300, Itamar Reis Peixoto wrote:
> On Wed, Sep 28, 2011 at 6:18 PM, Konrad Rzeszutek Wilk
>  wrote:
> > Ok, did 'xl list' show a new value?  What does this command give you:
> >
> > find /sys -name target_kb | xargs cat
>
> [root@localhost ~]# find /sys -name target_kb | xargs cat
> 3670016

Ok, so 3.6GB at start.
> while (true)
>  do
>   xl list | grep Domain-0
>  done
>
> I am able to see the numbers going down, and after some time stop.(freeze)

And ... how low?
> > Did you see the numbers go up/down? Can you be more specific please?
> > > >
>
> I have ordered a serial adapter to have the logs next time.

[itamar]

On Wed, Sep 28, 2011 at 7:59 PM, Konrad Rzeszutek Wilk
 wrote:
> On Wed, Sep 28, 2011 at 06:33:52PM -0300, Itamar Reis Peixoto wrote:
> > On Wed, Sep 28, 2011 at 6:18 PM, Konrad Rzeszutek Wilk
> >  wrote:
> > > Ok, did 'xl list' show a new value?  What does this command give you:
> > >
> > > find /sys -name target_kb | xargs cat
> >
> > [root@localhost ~]# find /sys -name target_kb | xargs cat
> > 3670016
>
> Ok, so 3.6GB at start.
> > while (true)
> >  do
> >   xl list | grep Domain-0
> >  done
> >
> > I am able to see the numbers going down, and after some time stop.(freeze)
>
> And ... how low?

last message I was able to see is at  2260

the hard disk led keeps always on and I have waited 10 minutes and no
luck, freezed.
> > > Did you see the numbers go up/down? Can you be more specific please?
> > > > >
> >
> > I have ordered a serial adapter to have the logs next time.

[Konrad Rzeszutek Wilk]

On Wed, Sep 28, 2011 at 08:03:37PM -0300, Itamar Reis Peixoto wrote:
> On Wed, Sep 28, 2011 at 7:59 PM, Konrad Rzeszutek Wilk
>  wrote:
> > On Wed, Sep 28, 2011 at 06:33:52PM -0300, Itamar Reis Peixoto wrote:
> > > On Wed, Sep 28, 2011 at 6:18 PM, Konrad Rzeszutek Wilk
> > >  wrote:
> > > > Ok, did 'xl list' show a new value?  What does this command give you:
> > > >
> > > > find /sys -name target_kb | xargs cat
> > >
> > > [root@localhost ~]# find /sys -name target_kb | xargs cat
> > > 3670016
> >
> > Ok, so 3.6GB at start.
> > > while (true)
> > >  do
> > >   xl list | grep Domain-0
> > >  done
> > >
> > > I am able to see the numbers going down, and after some time stop.(freeze)
> >
> > And ... how low?
>
> last message I was able to see is at  2260

So I did the same test on my 4GB box and found out that 'xm mem-set 0 1024'
xend helpfully (<sarcasm>) decreased the memory to 214MB. Which will definitly cause
head-ache for programs.

[Virgil]

AAAaaaahhhhhgggghhh

Noun: BALLOON!
Verb: BALLOONING!

V

On Fri, 30 Sep 2011 08:56:28 AM Konrad Rzeszutek Wilk wrote:
> On Wed, Sep 28, 2011 at 08:03:37PM -0300, Itamar Reis Peixoto wrote:
> > On Wed, Sep 28, 2011 at 7:59 PM, Konrad Rzeszutek Wilk
> >
> >  wrote:
> > > On Wed, Sep 28, 2011 at 06:33:52PM -0300, Itamar Reis Peixoto wrote:
> > > > On Wed, Sep 28, 2011 at 6:18 PM, Konrad Rzeszutek Wilk
> > > >
> > > >  wrote:
> > > > > Ok, did 'xl list' show a new value?  What does this command give
> > > > > you:
> > > > >
> > > > > find /sys -name target_kb | xargs cat
> > > >
> > > > [root@localhost ~]# find /sys -name target_kb | xargs cat
> > > > 3670016
> > >
> > > Ok, so 3.6GB at start.
> > > > while (true)
> > > >  do
> > > >   xl list | grep Domain-0
> > > >  done
> > > >
> > > > I am able to see the numbers going down, and after some time
> > > > stop.(freeze)
> > >
> > > And ... how low?
> >
> > last message I was able to see is at  2260
>
> So I did the same test on my 4GB box and found out that 'xm mem-set 0 1024'
> xend helpfully (<sarcasm>) decreased the memory to 214MB. Which will
> definitly cause head-ache for programs.
>
> --
> xen mailing list
> [e-mail]
> https://admin.fedoraproject.org/mailman/listinfo/xen