Hi,
Am 08.02.2013 08:15, schrieb Alexandre DERUMIER:
>>> Like uPNP on firewalls? Now I really gets scared!
>
> Hopefully, it's the switch which do the gvrp request ;)
> So the security is on the switch side and not proxmox.
>
> It's just that currently, proxmox doesn't respond to gvrp request.
>
Am 11.02.2013 07:06, schrieb Dietmar Maurer:
>> Patrick told me this is generally the way to go. He told something about mac
>> mismatches if you go the current way. But I'm not sure if I understood him
>> correctly.
>
> Can you point me to that discussion?
>
Sure:
https://lkml.org/lkml/2013/2/7
Hi list,
i really would like to set ulimit -c unlimited for the kvm process. From
time to time i've seen kvm segfaulting but without a coredump i'm unable
to get a stacktrace.
What would be the best way to archieve this with PVE?
Stefan
___
pve-devel m
Hi,
Am 11.02.2013 09:57, schrieb Dietmar Maurer:
Would you like to have a patchset to make pve-qemu-kvm compatible
with 1.4?
>>>
>>> Yes, but I will start that in 2 weeks (not earlier). I will let you know
>>> when I
>> need that patches.
>>
>> OK just send a mail and i'll send you the p
Hi,
Am 11.02.2013 09:43, schrieb Dietmar Maurer:
>> i really would like to set ulimit -c unlimited for the kvm process. From
>> time to
>> time i've seen kvm segfaulting but without a coredump i'm unable to get a
>> stacktrace.
>>
>> What would be the best way to archieve this with PVE?
>
> no id
One patch got this one:
Original-Nachricht
Betreff: Your message to pve-devel awaits moderator approval
Datum: Mon, 11 Feb 2013 10:32:54 +0100
Von: pve-devel-boun...@pve.proxmox.com
An: s.pri...@profihost.ag
Your mail to 'pve-devel' with the subject
[PATCH 4/9] cleanups 1
Hello,
please wait a bit i'll contact Patrick in a few minutes as i wanted to
switch to bonding today and it stops working again.
Let's see how a real solution would look like. Right now i've the same
problem as alexandre that the VM is not reachable at all when using bond.
Alexandre maybe you c
bond-lacp_rate fast
>bond-xmit-hash-policy layer2+3
> slaves eth0 eth1
>
>
> - Mail original -
>
> De: "Stefan Priebe - Profihost AG"
> À: "Dietmar Maurer"
> Cc: "Alexandre DERUMIER" , pve-devel@pve.proxm
Hi,
as this problem has nothing todo with my "new" vlan code i open up a new
thread.
The VMs work fine if the bridge is on top of a normal ethernet device.
They do not work if the bridge is on top of a bond.
Strangely i see ALL Host Traffic on the TAP device if the bridge is on a
bond which can'
-devel@pve.proxmox.com, "Dietmar Maurer"
> Envoyé: Lundi 11 Février 2013 20:44:28
> Objet: Re: [pve-devel] new bridge code doesn't work with redhat kernel
>
> HI,
>
> right now i'm talking about bridge on top of a bond NO VLAN involved.
> My commit
ge-
>> From: Stefan Priebe - Profihost AG [mailto:s.pri...@profihost.ag]
>> Sent: Dienstag, 12. Februar 2013 08:49
>> To: Alexandre DERUMIER
>> Cc: pve-devel@pve.proxmox.com; Dietmar Maurer
>> Subject: Re: [pve-devel] new bridge code doesn't work with redhat kern
Hi,
might you please also try to set your ethX devices in the bond to
promisious mode?
ip link set promisc on dev eth0
ip link set promisc on dev eth1
To me then everything works fine - like it is filtering to much when in
bond mode...
Stefan
Am 11.02.2013 15:06, schrieb Alexandre DERUMIER:
>>
Hi Dietmar,
thanks for pointing me to the idea the ethernet driver could be the problem.
I now updated to latest igb driver and it starts to work fine *GR*... i
never imagined that this could be the problem.
Stefan
Am 12.02.2013 09:15, schrieb Stefan Priebe - Profihost AG:
> Hi,
>
>
*arg* ignore me it still doesn't work... i accidently started the VM on
another host not using bonding... *sorry*
Stefan
Am 12.02.2013 09:34, schrieb Stefan Priebe - Profihost AG:
> Hi Dietmar,
>
> thanks for pointing me to the idea the ethernet driver could be the problem.
>
&
Am 13.02.2013 11:53, schrieb Dietmar Maurer:
>> I don't have too much time to test it,
>> So I rollback to old code for now, works 100% for me.
>
> I just reverted the patch. Please test if it works again.
>
OK i'll keep my patch then local.
Stefan
___
Hi,
but i think it's still crazy for the users - for me too. An example is
RBD. Default is writeback but rbd still behaves differently if you
select default or if you select writeback.
see for example here:
http://www.spinics.net/lists/ceph-devel/msg06920.html
Stefan
Am 14.02.2013 07:49, schrieb
Hi,
for cache=none librbd disables the rbd cache internal. Only for
cache=writeback it enables the internal rbd cache.
At least that's what the devs of ceph told me. I also encouraged data
loss when using cache=writeback with rbd and crashing VMs and i haven#t
seen this since using cache=none.
G
=8669142704b9d71c459c9740559c26d91ae772a2;hb=1f212b9d3edd8679bafd3bcf0301795206438724;hpb=9315cbfd8d7074eca44fbc5f93902e97b27d5240)
>
>
> so, rbd cache should work with default qemu 1.3 cache.
>
>
> - Mail original -
>
> De: "Alexandre DERUMIER"
>
Hi,
why is e1000 now the default network card? Doesn't virtio provide the
best performance?
Stefan
___
pve-devel mailing list
pve-devel@pve.proxmox.com
http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
Hello,
right now the network limit is in MB/s shouldn't it be Mbit/s like all
network stuff in the world?
Stefan
___
pve-devel mailing list
pve-devel@pve.proxmox.com
http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
At least to me it didn't made any difference.
Stefan
Am 16.02.2013 um 10:56 schrieb Dietmar Maurer :
> And should we support stripes somehow?
>
> –stripe-unit and –stripe-count
>
> What are reasonable defaults for that?
>
> From: pve-devel-boun...@pve.proxmox.com
> [mailto:pve-devel-boun.
Is the backup code ready for testing? Should it work with rbd? Anything special
I should focus on?
Stefan
___
pve-devel mailing list
pve-devel@pve.proxmox.com
http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
I can Look too should it generally work through vzdump and mode snapshot
Am 16.02.2013 um 15:57 schrieb Dietmar Maurer :
>> A forum user have the same problem. I think it shouldn't use "vzdump --
>> mode snapshot", maybe the new backup system is not yet switched for all
>> storages ?
>
> Will ta
Right but people are used to use Mbit/s for everything network related. So I
think a lot of people who want to use it will use it incorrect.
Stefan
Am 17.02.2013 um 07:43 schrieb Dietmar Maurer :
>> nobody an opinion about that? To me it had happened several times that i
>> putted in 100MB/s an
So right now the only way to get a consistent backup is to use stop mode?
Regarding vzdump the backup was finished and I wanted to start a 2nd backup for
the same server.
Greets
Stefan
Am 17.02.2013 um 07:45 schrieb Dietmar Maurer :
>> 1.) does the new Backup code snapshot mode mean the snapsh
Hi,
Am 17.02.2013 15:25, schrieb Dietmar Maurer:
>> All networking measurements are in bits/s Megabits, Kilobits
>>
>> Having Proxmox's in MB/s is inconsistent with the way the rest of the
>> industry measures network bandwidth.
>
> But 'tc' limits transferred bytes, not physically transfer
Hi,
Am 17.02.2013 15:39, schrieb Dietmar Maurer:
>>> All networking measurements are in bits/s Megabits, Kilobits
>>>
>>> Having Proxmox's in MB/s is inconsistent with the way the rest of the
>>> industry measures network bandwidth.
>
> Besides, is there any other reason why 'Bits' are bette
Hi,
Am 17.02.2013 15:50, schrieb Dietmar Maurer:
>> Right but people are used to use Mbit/s for everything network related. So I
>> think a lot of people who want to use it will use it incorrect.
>
> Maybe It would be nice to have an extjs widget where we can select the unit?
Could be a solution
Hi,
8 Bit = 1 Byte
Stefan
Am 18.02.2013 09:08, schrieb Dietmar Maurer:
>>> But 'tc' limits transferred bytes, not physically transferred bits.
>>
>> i think multiply or divide by 8 shouldn't be a problem ;-)
>
> Why 8? Usually a byte uses 9 or 10bits one the network?
>
_
Am 18.02.2013 09:11, schrieb Dietmar Maurer:
>>> Besides, is there any other reason why 'Bits' are better that 'Bytes'?
>>
>> No the users are just used to it. Even my grandfather knows that his network
>> card is 100Mbit/s so even he would always put in 100 or 1000 or
>> 10 and thinks in Mbit/s.
Am 18.02.2013 09:13, schrieb Dietmar Maurer:
>> 8 Bit = 1 Byte
>
> But the ethernet speed limit refers to physically transferable bits, and
> usually
>
> 9Bit = 1 Byte
No idea what you mean but 8 bit is 1 byte. Also the whole internet
refers to 8 bit.
http://www.unitjuggler.com/memory-umwandel
Hi,
Am 18.02.2013 09:27, schrieb Dietmar Maurer:
>>> But the ethernet speed limit refers to physically transferable bits,
>>> and usually
>>>
>>> 9Bit = 1 Byte
>>
>> No idea what you mean but 8 bit is 1 byte. Also the whole internet refers to
>> 8
>> bit.
>
> We talk about network card speed limi
Am 18.02.2013 09:39, schrieb Dietmar Maurer:
>>> We talk about network card speed limit.
>>> At network level, you need to transfer parity and stop bits, ethernet
>> headers, ...
>>>
>>> So you need more the 8 bits to transfer one byte!
>>
>> Yes but you also have the same overhead using MB/s so i
Same for migrate a running vm. I think it will break both.
Am 18.02.2013 um 10:40 schrieb Dietmar Maurer :
> I we change the HW structure (attach usb table to ehci), does that break
> all snapshots?
> ___
> pve-devel mailing list
> pve-devel@pve.proxmox
Am 18.02.2013 09:17, schrieb Dietmar Maurer:
>> Yes, sure but they know networking stuff is always mbit/s even wlan es
>> mbit/s ;-)
>>
>> Maybe a poll in the PVE Forum?
>
> I already pointed out how you can implement this (writing a extjs widget),
> so there is no need to poll.
I searched googl
Hello,
restore from snapshot does not work with current git code.
Message:
Unknown savevm section or instance ':00:01.2/uhci' 0
kvm: Error -22 while loading VM state
Greets,
Stefan
___
pve-devel mailing list
pve-devel@pve.proxmox.com
http://pve.pro
e the test 5 or 6 times, restoring from old
> snapshot, and I didn't have any problem.
>
> I'll redo tests
>
>
> ----- Mail original -
>
> De: "Stefan Priebe - Profihost AG"
> À: pve-devel@pve.proxmox.com
> Envoyé: Jeudi 21 Février 2013 14:51:
l be stable, this will not break the old snapshot with
> usb1/2 controllers.
>
>
> - Mail original -
>
> De: "Alexandre DERUMIER"
> À: "Stefan Priebe - Profihost AG"
> Cc: pve-devel@pve.proxmox.com
> Envoyé: Jeudi 21 Février 2013 15:03:
ent or
> console ?
>
>
> - Mail original -
>
> De: "Alexandre DERUMIER"
> À: "Stefan Priebe - Profihost AG"
> Cc: pve-devel@pve.proxmox.com
> Envoyé: Jeudi 21 Février 2013 15:15:44
> Objet: Re: [pve-devel] restore from snapshot doe
Hi funny,
this time it worked...
Stefan
Am 21.02.2013 15:43, schrieb Alexandre DERUMIER:
> ok. And you can reproduce it ? or do you have had the bug 1 time ?
>
> - Mail original -
>
> De: "Stefan Priebe - Profihost AG"
> À: "Alexandre DERUMIER&q
Right now if I change the network speed the card is unplugged and plugged in
again. The vm is offline after that in linux.
Stefan
Am 21.02.2013 um 18:44 schrieb Alexandre Derumier :
> Here my patch queues for qemu-server
>
> ___
> pve-devel mailing l
Hi,
yes current git code isn't usable for me anymore.
The dmesg output looks like this and then the network is and stays offline.
[61836.877654] pci :00:12.0: [1af4:1000] type 00 class 0x02
[61836.877732] pci :00:12.0: reg 10: [io 0x-0x001f]
[61836.85] pci :00:12.0: reg
Hi,
yes it's always this snapshot.
Stefan
Am 21.02.2013 15:58, schrieb Alexandre DERUMIER:
> Do you mean with same snapshot ?
>
>
> - Mail original -
>
> De: "Stefan Priebe - Profihost AG"
> À: "Alexandre DERUMIER"
> Cc: pve-devel
Seems i missed you patch. Which one was it?
Greets,
Stefan
Am 21.02.2013 16:16, schrieb Alexandre DERUMIER:
> I have just sent a patch on the mailing, could you try it to see if it's help
> ?
>
> - Mail original -
>
> De: "Alexandre DERUMIER"
&
Hi,
Am 22.02.2013 09:33, schrieb Alexandre DERUMIER:
> Do you have my today patch ? ("don't unplug nic on update")
>
> It should fix it.
>
> The problem was that on update, I tried to unplug then plug network card.
> As your guest don't have hotplug acpi modules, the unplug fail.
Strange it sho
Hi,
Am 13.02.2013 18:56, schrieb Dietmar Maurer:
>> -Original Message-
>> From: Stefan Priebe - Profihost AG [mailto:s.pri...@profihost.ag]
>> Sent: Mittwoch, 13. Februar 2013 13:38
>> To: Dietmar Maurer
>> Cc: Alexandre DERUMIER; pve-devel@pve.proxmox.com
&
Hello,
I've seen this sometimes. Is there any way to see how the output of the
ssh command was?
Feb 22 14:48:05 migration speed: 819.20 MB/s - downtime 49 ms
Feb 22 14:48:05 migration status: completed
Feb 22 14:48:06 ERROR: command '/usr/bin/ssh -o 'BatchMode=yes'
root@10.255.0.20 qm resume 129
it is because we want to be sure to
> resume it after move the config file.
>
>
>
> - Mail original -
>
> De: "Stefan Priebe - Profihost AG"
> À: pve-devel@pve.proxmox.com
> Envoyé: Vendredi 22 Février 2013 15:01:25
> Objet: [pve-devel] succes
Mhm but in cases like Mine we have no running vm on both sides. So are you sure
that when migrating there could be a reason to have two vms running?
Stefan
Am 22.02.2013 um 18:51 schrieb Dietmar Maurer :
>> Dietmar why do we pause?
>
> For safety reasons. We want to avoid the same VM running t
e.
>
> Ideas are welcome to improve this ;)
>
>
>
>
>
> - Mail original -
>
> De: "Dietmar Maurer"
> À: "Stefan Priebe - Profihost AG"
> Cc: "Alexandre DERUMIER" , pve-devel@pve.proxmox.com
> Envoyé: Vendredi 2
sorry sent new fix.
Stefan
Am 28.02.2013 12:53, schrieb Dietmar Maurer:
> I will revert that patch. This only works if the VM is on local machine. Else
> it breaks the console.
>
> I guess you should add the code into 'qm vncproxy' instead (file 'qm').
>
>> -Original Message-
>> From:
this is too complex
+ return if (!$rate || ($rate && $rate == 0));
this is the same:
return if !$rate;
$rate can be 0 or undef
Stefan
Am 05.03.2013 12:20, schrieb Alexandre DERUMIER:
>>> This expression makes no sense to me? Maybe we just need
>>>
>>> return if !$rate;
>
> pve-manager can s
Is there a way to get monthly traffic per nic per vm?
Stefan
___
pve-devel mailing list
pve-devel@pve.proxmox.com
http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
..@pve.proxmox.com [mailto:pve-devel-
>> boun...@pve.proxmox.com] On Behalf Of Dietmar Maurer
>> Sent: Samstag, 09. März 2013 14:55
>> To: Alexandre DERUMIER; Stefan Priebe - Profihost AG
>> Cc: pve-devel@pve.proxmox.com
>> Subject: Re: [pve-devel] Count monthly traffic
&g
Am 09.03.2013 um 15:08 schrieb Dietmar Maurer :
>> Maybe can we improve stats with more detailled values for each nics, each
>> disk
>> && each cpu cores ?
>
> You will end up developing a monitoring application. And the amount of data
> you need to
> store will also grow considerable.
>
> @S
Hi,
i'm still struggling.
1.) first i need to update the status.c file and introduce:
pve3.0-vm folder.
2.) What are these skip steps? in status.c?
3.) Then i need to update PVE/API2Tools.pm to pass the additional values.
4.) bin/pvestatd needs to be updated to add the new values
But then we hav
Am 12.03.2013 16:28, schrieb Dietmar Maurer:
>> Yes, I think it make sense. Like this we can avoid to write null values for
>> non
>> existant network card.
>> I don't know if Dietmar have an opinion about this.
>
> The suggestion was to create one new file (to store 64 new values).
> Another op
Why does a single file creates less load? Both contain the same amount of data.
Am 12.03.2013 um 16:49 schrieb Dietmar Maurer :
>> Yes - just let me know what you prefer. Internally we create one .rrd for
>> every
>> interface, disk, ... for our normal physikal servers. This is very flexible
>>
Am 12.03.2013 um 19:04 schrieb Dietmar Maurer :
>
>
>> -Original Message-----
>> From: Stefan Priebe - Profihost AG [mailto:s.pri...@profihost.ag]
>> Sent: Dienstag, 12. März 2013 16:32
>> To: Dietmar Maurer
>> Cc: Alexandre DERUMIER; pve-devel@pve.pr
Hi,
Am 13.03.2013 07:17, schrieb Dietmar Maurer:
>>> How large is a file to store 32 netin/netout values with our setup?
>>>
>> No idea.I can check this if you want.
>
> Please do
>
>> but this means we always have a bigger file
>> cause most vms have just one or two network cards.
>
> File si
ur Systèmes et Réseaux*
>
> *Fixe : *03 20 68 88 85
>
> *Fax : *03 20 68 90 88
>
> 45 Bvd du Général Leclerc 59100 Roubaix
> 12 rue Marivaux 75002 Paris
>
>
>
>
> *De: *"Stefan Priebe - Profihost AG"
> *À: *&quo
Hello,
after adding three new nodes to my proxmox cluster the daemon log of all
new nodes looks like this:
Mar 22 11:30:47 cloud1-1312 pmxcfs[2292]: [dcdb] notice: data
verification successful
Mar 22 11:30:47 cloud1-1312 pmxcfs[2292]: [status] notice: received log
Mar 22 11:30:49 cloud1-1312 pmxc
Hi,
Am 22.03.2013 12:15, schrieb Dietmar Maurer:
>> after adding three new nodes to my proxmox cluster the daemon log of all new
>> nodes looks like this:
>
> cluster connection works? What is the output of
>
> # pvecm nodes
>
> and
>
> # pvecm status
When the nodes are freshly bootet everythi
it always seems to happen, when 1312 comes up... is there any way to
debug this?
Stefan
Am 22.03.2013 12:38, schrieb Stefan Priebe - Profihost AG:
> Hi,
> Am 22.03.2013 12:15, schrieb Dietmar Maurer:
>>> after adding three new nodes to my proxmox cluster the daemon log of all new
OK it has nothing todo with 1312 - it just happens randomly - no idea why.
Could it make sense to update the Corosync Cluster Engine. i've 1.4.4
currently.
Any other ideas how to find out why it fails or what fails.
Stefan
Am 22.03.2013 13:56, schrieb Stefan Priebe - Profihost AG:
> i
> Am 22.03.2013 13:56, schrieb Stefan Priebe - Profihost AG:
>> it always seems to happen, when 1312 comes up... is there any way to
>> debug this?
>>
>> Stefan
>>
>> Am 22.03.2013 12:38, schrieb Stefan Priebe - Profihost AG:
>>> Hi,
>>> Am 22.03.20
Hi,
Am 22.03.2013 15:13, schrieb Dietmar Maurer:
>> But nodes are suddenly disconnected... corosync process does not run
>> anymore.But nothing in the corosync log and or daemon log.
>
> If that is reproducible, I would enable core dumps and debug corosync.
Yes will do this. Is the corosync 1.4.
Important for pve?
Anfang der weitergeleiteten E‑Mail:
> Von: Paolo Bonzini
> Datum: 22. März 2013 17:57:48 MEZ
> An: qemu-de...@nongnu.org
> Kopie: cor...@linux.vnet.ibm.com, qemu-sta...@nongnu.org
> Betreff: [Qemu-stable] [PATCH] qemu-bridge-helper: force usage of a very high
> MAC address f
I find this one easier to read otherwise I think it's fine.
Int($size/1024+1)
Stefab
Am 25.03.2013 um 06:35 schrieb Dietmar Maurer :
> User reports problems on restore:
>
> http://forum.proxmox.com/threads/13296-Ceph-Restore-Failure
>
> We do:
>
> my $cmd = &$rbd_cmd($scfg, $storeid,
Am 25.03.2013 08:18, schrieb Dietmar Maurer:
> But your solution grow the image by 1MB by default!
*urg* meant:
int($size/1024+0.9)
Stefan
___
pve-devel mailing list
pve-devel@pve.proxmox.com
http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
This works best with a vanilla kernel Bit does not with the rhel6 one. I use
the same patch locally.
Stefan
Am 02.04.2013 um 00:52 schrieb Andrew Thrift :
> FYI patch is:
>
>
> *** Network.pm.orig 2013-04-02 10:19:42.0 +1300
> --- Network.pm 2013-04-02 11:41:44.0 +1300
>
Hi,
is it planned to have predefined Hardware / VM Profiles in Proxmox?
Stefan
___
pve-devel mailing list
pve-devel@pve.proxmox.com
http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
Am 16.04.2013 10:42, schrieb Dietmar Maurer:
>> is it planned to have predefined Hardware / VM Profiles in Proxmox?
>
> VM templates will provide a way to do that (patches from Alexandre).
Yes and no as the disk will be also cloned. I'm just speaking about the
settings like in OpenStack, OpenNebu
Hi,
Am 16.04.2013 10:58, schrieb Dietmar Maurer:
>>> VM templates will provide a way to do that (patches from Alexandre).
>>
>> Yes and no as the disk will be also cloned. I'm just speaking about the
>> settings
>> like in OpenStack, OpenNebula and others.
>
> We can also allow to create template
you don't try to use qemu-img on a running vm, because it
> don't work)
>
> - Mail original -
>
> De: "Stefan Priebe - Profihost AG"
> À: "Dietmar Maurer"
> Cc: pve-devel@pve.proxmox.com
> Envoyé: Mardi 16 Avril 2013 11:04:39
&
a running vm, because it
>> don't work)
>>
>> - Mail original -
>>
>> De: "Stefan Priebe - Profihost AG"
>> À: "Dietmar Maurer"
>> Cc: pve-devel@pve.proxmox.com
>> Envoyé: Mardi 16 Avril 2013 11:04:39
>> O
ndard proxmox disk name...
>
> (should be base-166-disk-1)
>
>
> - Mail original -
>
> De: "Stefan Priebe - Profihost AG"
> À: "Alexandre DERUMIER"
> Cc: pve-devel@pve.proxmox.com, "Dietmar Maurer"
> Envoyé: Mardi 16 Avri
muServer::update_config_nolock($vmid, $conf, 1); get's never
executed here in Qemu.pm
As we have a return before.
Stefan
Am 16.04.2013 13:28, schrieb Stefan Priebe - Profihost AG:
> This template has no disk. And the Name base- was chosen by me while
> creating the VM. I didn't k
Hello,
i was trying to debug my template creation problem.
qm template 166
calls POST 166/template
which then spawns a worker doing:
PVE::QemuServer::template_create($vmid, $conf, $disk);
but sub template_create in QemuServer.pm does nothing with the VM itself
it just changes drives.
But i wa
/)
> {
>
> also
>
>> parse_volname()
> if ($volname =~ m/^((base-(\d+)-\S+)\/)?((base)?(vm)?-(\d+)-\S+)$/) {
>
>
>
> If you have edited manually the vm config file with custom name, I think qemu
> can start,
> but I'm not sure of the behaviour
d=none''
failed: got signal 1
Stefan
Am 16.04.2013 13:17, schrieb Stefan Priebe - Profihost AG:
> Hi,
>
> will recheck that.
>
> Right now i want to create a template. But nothing happens.If i convert
> a Machine into a template i just see a task qmtemplate => OK. That
so a segfault.
I'm using: pve-qemu-kvm 1.4-6
Stefan
> - Mail original -
>
> De: "Stefan Priebe - Profihost AG"
> À: "Alexandre DERUMIER"
> Cc: pve-devel@pve.proxmox.com, "Dietmar Maurer"
> Envoyé: Mardi 16 Avril 2013 14:36:50
Yes the idea was to have a template so nobody can touch it ;-)
Stefan
> - Mail original -
>
> De: "Stefan Priebe - Profihost AG"
> À: "Alexandre DERUMIER"
> Cc: pve-devel@pve.proxmox.com, "Dietmar Maurer"
> Envoyé: Mardi 16 A
l by example ?
> Also a segfault.
>
> I'm using: pve-qemu-kvm 1.4-6
>
> Stefan
>
>
>> - Mail original -
>>
>> De: "Stefan Priebe - Profihost AG"
>> À: "Alexandre DERUMIER"
>> Cc: pve-devel@pve.proxmo
Am 16.04.2013 15:42, schrieb Dietmar Maurer:
but it seems i can't create a template without a drive ;-) I just
wanted to have a template with settings for VMs. Like predefined
configuration without a disk.
>>
>> ok, I understand now ;)
>>
>> Indeed, currently, a template is a vm with
Am 12.03.2013 18:53, schrieb Dietmar Maurer:
>> Why does a single file creates less load? Both contain the same amount of
>> data.
>
> A VM can use up to 32 network devices. Updating 32 files create more load
> than updating a single file?
I want to proceed with this one. First i need to know h
.255.0.102\:6789\;:auth_supported=none''
>
>
> - Mail original -
>
> De: "Alexandre DERUMIER"
> À: "Stefan Priebe - Profihost AG"
> Cc: pve-devel@pve.proxmox.com, "Dietmar Maurer"
> Envoyé: Mardi 16 Avril 2013 15:44:1
ah it works if i set source AND dst format to raw. So we have to
overwrite it for rbd to raw.
Stefan
Am 16.04.2013 16:07, schrieb Stefan Priebe - Profihost AG:
> Hi,
>
> does not work for me ;-(
>
> activate volume vmstorssd1:vm-170-disk-1
> copy vmstorssd1:vm-166-disk-1 to
ieb Alexandre DERUMIER:
> oh, I forgot, also -f !
>
> -f raw -O raw
>
>
> (I have tested it's work fine for me, rbd to rbd)
>
> - Mail original -
>
> De: "Stefan Priebe - Profihost AG"
> À: "Alexandre DERUMIER"
> Cc: pve-
Am 17.04.2013 07:24, schrieb Dietmar Maurer:
>
>> Am 12.03.2013 18:53, schrieb Dietmar Maurer:
Why does a single file creates less load? Both contain the same amount of
>> data.
>>>
>>> A VM can use up to 32 network devices. Updating 32 files create more load
>> than updating a single file?
>
Am 17.04.2013 09:08, schrieb Dietmar Maurer:
>>> It is still not clear if we want a single file, or multiple files. I
>>> guess this requires careful evaluation. How does such system behave
>>> when we have 3000VMs? We need to test that before we go further.
>>
>> Oh i thought a single file contain
Am 17.04.2013 09:28, schrieb Dietmar Maurer:
>> Sure but how to benchmark? Just create a file with 32 values and create
>> 2 files (this but be the avg of network interfaces) may be it is just 1.5.
>>
>> And then update both each second and look at the disk i/o? Or how did you
>> imagine this. I th
Am 17.04.2013 09:44, schrieb Dietmar Maurer:
>>> I am particularly interested in IO load (on normal ide disks) and memory
>> requirements of rrdcached.
>>
>> Don't have IDE Disks at all. Can only provide SSD or SAS or SATA2 Disks.
>
> I know that you have quite fast hardware. But we need to test o
Am 17.04.2013 13:52, schrieb Dietmar Maurer:
>> But on IDE with 3000 Files it doesn't work every 10s. At least without
>> rrdcache.
>> But that sounds in general a way to much for such a HW.
>
> So maybe it is a bad idea to use rrd for that purpose - any other suggestions?
How have you calculate
Am 17.04.2013 14:09, schrieb Dietmar Maurer:
>> How have you calculated your 3000 Files?
>>
>> 3000 VMs all with 32 nics seems unrealistic to me.
>
> I simply want to show you that this can overload all! servers inside a
> cluster by
> just writing rrd files. I am not keen to spend any resources
Am 17.04.2013 14:11, schrieb Dietmar Maurer:
>> How have you calculated your 3000 Files?
>>
>> 3000 VMs all with 32 nics seems unrealistic to me.
>
> BTW, we have several users with >500VMs, so I guess we can easily reach that
> limit.
Sure but those won't have 32 NICs. So if we go the way one f
Am 17.04.2013 14:17, schrieb Dietmar Maurer:
>> Nowhere ;-) how about just return the counter values for the correct tap
>> device
>> through API?
>>
>> So it is basically:
>> 1.) a wrapper from netX to correct tap
>> 2.) query tap counter inout / output values
>> 3.) allow to query this through A
Hi,
Am 17.04.2013 14:42, schrieb Adrian Costin:
> I'm sorry to dropping in, but isn't Proxmox already counting traffic in
> it's internal RRD? Couldn't the "rrddata" API call be used to retrieve
> the data in the RRD externally and process it to count the total
> bandwidth used in 1 month?
>
> May
*urg* sorry yes correct path is PVE/API2/Nodes.pm.
Am 18.04.2013 11:17, schrieb Dietmar Maurer:
> I am unable to apply that patch, I get
>
> error: API2/Nodes.pm: does not exist in index
>
> correct path is PVE/API2/Nodes.pm?
>
>> -Original Message-
>> From: pve-devel-boun...@pve.proxmo
Hi,
download.proxmox.com seems to be down.
Stefan
___
pve-devel mailing list
pve-devel@pve.proxmox.com
http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
401 - 500 of 1027 matches
Mail list logo
