[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-API] question about check VM host cpu and ram


  • To: Olivier LAMBERT <lambert.olivier@xxxxxxxxx>
  • From: Christopher Winslett <christopherwinslett@xxxxxxxxx>
  • Date: Fri, 18 Sep 2009 09:11:17 -0500
  • Cc: xen-api@xxxxxxxxxxxxxxxxxxx
  • Delivery-date: Fri, 18 Sep 2009 07:11:41 -0700
  • Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; b=Mtd55VHx7Jf/EmIYlzUvIUwvKTaMZIoIEjC701YcrsIkM4cs45CaHzIdNProIMJIIS PPIVhkAQxiMsAGfdowR+VGbg5YmUbcmP4jFdsrPyR7GgtPjGesAKRXheD+L/1i5/hv0R hvzHh7zDgXdwxGyJ5LSFlGUnhTLCQS2ApSsZI=
  • List-id: Discussion of API issues surrounding Xen <xen-api.lists.xensource.com>

It is "normal" in that mine does the same thing.

We use Nagios for these metrics.

cw

On Fri, Sep 18, 2009 at 7:58 AM, Olivier LAMBERT <lambert.olivier@xxxxxxxxx> wrote:
Many thanks !

It's more clear for me now.
I've forgot the mid (what a shame :o )

But despite VM_metrics.get_record works, the current used memory is
"missing" (or not implemented). Damned :/
I've got :

VCPUs_params cpumap01,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31
cap 0
weight 256
last_updated 20090918T12:46:32
uuid 6af1f0db-ee19-bf94-efa4-ca021080b1e3
VCPUs_flags 0 blocked online
start_time 20090918T09:58:38
memory_actual 268435456
state blocked
VCPUs_utilisation 0 0.00016007062223236292
VCPUs_number 1
VCPUs_CPU 0 2

Is it "normal" ?

Thanks.

On Fri, Sep 18, 2009 at 2:01 PM, Christopher Winslett
<christopherwinslett@xxxxxxxxx> wrote:
> Olivier,
> Use "VM_metrics" to see the the hardware reported metrics.
> Use "VM_guest_metrics" to see the VM reported metrics.  Probably need to
> install the XENSource kernel drivers on the guest OS to get these figures;
> all mine were null too.
> What type of figures are you looking for?
> The following RPC calls work for me:
>
> sid = Session ID
> mid = Metrics ID Associated with VM
>
> VM_metrics.get_record(sid, mid)
>
> It would return a value for memory_actual, VCPUs_number.
> In Ruby it would be the following RPC calls:
>
> require "XMLRPC/Client"
>
> s = XMLRPC::Client.new2("https://localhost")
> sid = s.call("session.login_with_password", "uname", "pass")["Value"]
>
> mid = "OpaqueRef:xxxxx"
>
> memory = s.call("VM_metrics.get_record", sid, mid)["Value"]["actual_memory"]
>
> That being said, to get utilization numbers, we use a network monitor system
> like Nagios.
> cw
>
> On Fri, Sep 18, 2009 at 5:33 AM, Olivier LAMBERT <lambert.olivier@xxxxxxxxx>
> wrote:
>>
>> I'm reading the Xen Management API pdf (Version: API Revision 1.0.6,
>> Date: 24th July 2008, Stable Release).
>> It talks about class "VM metrics" (§2.9) and  "VM guest metrics"
>> (§2.10). But all method call with these classes in RPC send me NULL
>> response.
>>
>> Despite it works fine for other classes, like VM, or network etc. All
>> classes with metrics seems to be not recognized by the API.
>>
>> Configuration : Dom0 on Debian Lenny, with xen-hypervisor-3.2-1-amd64.
>>
>> How can I do to gather informations about CPU, ram and network on
>> domU's with the API ?
>>
>>
>> Thank you,
>>
>>
>>
>> Olivier.
>>
>> _______________________________________________
>> xen-api mailing list
>> xen-api@xxxxxxxxxxxxxxxxxxx
>> http://lists.xensource.com/mailman/listinfo/xen-api
>
>
>
> --
> Christopher Winslett
> christopherwinslett@xxxxxxxxx
>



--
Christopher Winslett
christopherwinslett@xxxxxxxxx
_______________________________________________
xen-api mailing list
xen-api@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/mailman/listinfo/xen-api

 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.