Hi,
while benchmarking my kvm machines using unixbench i see the following
reproducable behaviour.
Host has 24 cores.
KVM Machine set to 2 cores: Unixbench Value 2514
KVM Machine set to 16 cores: Unixbench Value 6714
KVM Machine set to 24 cores: Unixbench Value 1742
Also i see a max of 105%
What are the physical processors ? (Is it an numa architecture ? if yes, could
be a memory bandwith problem)
Do you have tried to launch unixbench on the host to compare ?
Maybe can you try to add +x2apic cpu flag ? It should reduce interrupts. (It's
already in current qemu-server git)
the client_migrate_info host= value (proxyticket), seem to crash the client
if the value is 247 characters.
Do you think it's possible to reduce the proxyticket size ?
I have thinked about it, as the proxyticket in sent through tls, maybe it's
possible to create a ticket like vnc ticket. (I
the client_migrate_info host= value (proxyticket), seem to crash the
client if the value is 247 characters.
Do you think it's possible to reduce the proxyticket size ?
I have thinked about it, as the proxyticket in sent through tls, maybe it's
possible to create a ticket like vnc ticket.
I'm reading the spice code to see why it don't take bigger string in
client_info_migrate. (As it's working for initial connection, this is strange)
I guess it is easy to create a shorter ticket - I will take a look at this now.
___
pve-devel mailing
I'm reading the spice code to see why it don't take bigger string in
client_info_migrate. (As it's working for initial connection, this is strange)
Sigh, I am unable to make the ticket smaller - sorry.
___
pve-devel mailing list
Hi,
Am 19.07.2013 09:22, schrieb Alexandre DERUMIER:
What are the physical processors ? (Is it an numa architecture ? if yes,
could be a memory bandwith problem)
Do you have tried to launch unixbench on the host to compare ?
Maybe can you try to add +x2apic cpu flag ? It should reduce
Sigh, I am unable to make the ticket smaller - sorry.
Does it need to be in plain text?
What?
___
pve-devel mailing list
pve-devel@pve.proxmox.com
http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
Hi guys,
I'm interested in contributing code/features for Proxmox, and want to know
where to begin.
(I've already contributed with Danish translation... small thing, I know)
It's important to me to know how my code/features will be included in the
official release, and not only as my private
Begin forwarded message:
Date: Fri, 19 Jul 2013 11:28:37 +0200
From: Michael Rasmussen m...@datanom.net
To: Dietmar Maurer diet...@proxmox.com
Subject: Re: [pve-devel] spice tls + proxy:
ssl_verify.c:484:openssl_verify: ssl: hostname
On Fri, 19 Jul 2013 09:11:48 +
Dietmar
Hi all,
I have just tested the newest released version of icedtea-plugin
(1.4.3) which include the patch for the bug we have reported
(http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=707729).
Works flawlessly and no 100% usage here;-) Good bye Oracle Java and
welcome OpenJDK!
--
Hilsen/Regards
On Fri, 19 Jul 2013 09:30:56 +
Dietmar Maurer diet...@proxmox.com wrote:
What?
I was thinking of making it smaller by using compression.
we are limited to use character [a-z0-9].
compression + quoted-printable or compression + base64 encoded ?
--
Hilsen/Regards
Michael
It’s not a “shame” that we do not integrate this, it’s a feature that we
include only code fits into the product. The way it should work is defined,
but this guy does not followed this idea. This is ok for me but it’s clear
that in this case we will not integrate it.
My two cents ;)
Yeah.
On Fri, 19 Jul 2013 09:30:56 +
Dietmar Maurer diet...@proxmox.com wrote:
What?
I was thinking of making it smaller by using compression.
we are limited to use character [a-z0-9].
compression + quoted-printable or compression + base64 encoded ?
The result may only
On Fri, 19 Jul 2013 09:45:40 +
Dietmar Maurer diet...@proxmox.com wrote:
The result may only contain character [a-z0-9] (no binary data).
That is the point of quoted printable and base64?
--
Hilsen/Regards
Michael Rasmussen
Get my public GnuPG keys:
michael at rasmussen dot cc
I'm trying to build last pve-qemu-kvm git, but spice-socket.patch is missing
Applying patch spice-socket.patch
patching file qemPatch spice-tests.patch does not exist
u-options.hx
Hunk #1 succeeded at 899 (offset 12 lines).
Hunk #2 succeeded at 923 (offset 12 lines).
patching file
fixed
-Original Message-
From: pve-devel-boun...@pve.proxmox.com [mailto:pve-devel-
boun...@pve.proxmox.com] On Behalf Of Alexandre DERUMIER
Sent: Freitag, 19. Juli 2013 11:54
To: pve-devel@pve.proxmox.com
Subject: [pve-devel] pve-qemu-kvm git, missing spice-socket.patch
I'm
The result may only contain character [a-z0-9] (no binary data).
That is the point of quoted printable and base64?
no. Any why does that make the string smaller?
___
pve-devel mailing list
pve-devel@pve.proxmox.com
-Original Message-
From: pve-devel-boun...@pve.proxmox.com [mailto:pve-devel-
boun...@pve.proxmox.com] On Behalf Of Michael Rasmussen
Sent: Freitag, 19. Juli 2013 11:33
To: pve-devel@pve.proxmox.com
Subject: [pve-devel] icedtea-plugin
Hi all,
I have just tested the newest
Maybe can we try to compress the key with gzip, then convert it with base32 ?
- Mail original -
De: Dietmar Maurer diet...@proxmox.com
À: Michael Rasmussen m...@datanom.net, pve-devel@pve.proxmox.com
Envoyé: Vendredi 19 Juillet 2013 12:06:23
Objet: Re: [pve-devel] spice tls + proxy:
I'm reading the spice code to see why it don't take bigger string in
client_info_migrate. (As it's working for initial connection, this is
strange)
Sigh, I am unable to make the ticket smaller - sorry.
I guess I can make it smaller if I use the less secure pve-www.key - will try
that
Please wait a few minutes - I will try a more reasonable approach.
-Original Message-
From: Alexandre DERUMIER [mailto:aderum...@odiso.com]
Sent: Freitag, 19. Juli 2013 12:13
To: Dietmar Maurer
Cc: Michael Rasmussen; pve-devel@pve.proxmox.com
Subject: Re: [pve-devel] spice tls +
22 matches
Mail list logo