libvirt users - Sep 2012 - libvirt failed to respond after quite a few vnet bridges were created

During concurrent 80 VM creation, each VM takes two network connections,
libvirtd failed to respond.  Neither restart libvirtd nor restart CIMOM
helped.

There were 152 vnet bridges found on this RHEL KVM host.  After all the
vnet bridges were manually removed, libvirtd responded normally again.

Is there a known issue about the number limit of vnet bridges regarding
libvirtd?

Thank you.

--Yih
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://listman.redhat.com/archives/libvirt-users/attachments/20120908/b77712c2/attachment.htm>

Hi Yih,

On 09/08/2012 06:06 PM, Yih Chuang wrote:
> During concurrent 80 VM creation, each VM takes two network 
> connections, libvirtd failed to respond.  Neither restart libvirtd nor 
> restart CIMOM helped.
I had a little surprise last week that might just be the thing ruining 
your party too. I found out that [it seems] libvirtd has a global lock 
of some sort that blocks all operations until a VM starts. That includes 
the runtime of startup hook script as well. I have a "prepare" QEMU 
script hook that takes a while - I use it to set up volumes on first run 
and until that hook finishes running the entire libvirtd daemon remains 
totally unresponsive. All queries, no matter if issued through the API 
(I use the python bindings) or virsh get blocked and are unblocked only 
after the hook finishes.

Now, I guess I'm abusing that hook and I should be setting up my volumes 
elsewhere, but still this came to me as a surprise. Anyway, you might 
want to check if some of your VM's run fancy hooks or have some other 
reason that makes their preparation or start up too slow.

Best,
Hristo