Re: Issues with Rocky 9.1 qcow2 image

[Nux]

What hardware profile are you using? If you boot a system rescue cd or a rocky 
8 one, what is the output of "lspci"? 

On 6 December 2022 18:41:25 GMT, Jeremy Hansen  
wrote:
>Unfortunately I’m getting a kernel panic upon boot. I tried both the 
>production grub option and the rescue and I get the same result. Maybe 
>something with the size of the root disk? I have it just at 8 gig right now? 
>Something with the initrd perhaps?
>
>Thanks!
>-jeremy
>
>> On Tuesday, Dec 06, 2022 at 12:53 PM, Jeremy Hansen > (mailto:jer...@skidrow.la)> wrote:
>> Wow. Thank you! I appreciate this. I’ll report back after trying.
>>
>> -jeremy
>>
>>
>>
>> > On Tuesday, Dec 06, 2022 at 11:39 AM, Nux > > (mailto:n...@li.nux.ro)> wrote:
>> >
>> >
>> > God knows, I've built you one with the Cloudstack bits enabled, you can
>> > grab it from below. It's got the root login enabled, so that's your
>> > default user. Let me know if it works.
>> >
>> > http://dl.openvm.eu/cloudstack/rockylinux/9/rockylinux-9-kvm-rootuser.qcow2.bz2
>> >
>> > I'll build more later on with non-priviliged users.
>> >
>> > ---
>> > Nux
>> > www.nux.ro [1]
>> >
>> > On 2022-12-06 15:26, Jeremy Hansen wrote:
>> >
>> > > Unfortunately I think it gets stuck before ever reaching grub. I tried
>> > > to catch it on the console after a reboot and this is all I see. I
>> > > never get to the grub menu.
>> > >
>> > > Is there a way I can alter the qcow2 image before I create a template?
>> > >
>> > > I thought perhaps I could see the IP it obtains via dhcp, but I see
>> > > nothing coming from that VM which tells me it's not getting to the
>> > > point of bringing up its interface.
>> > >
>> > > Thanks
>> > > -jeremy
>> > >
>> > > On Tuesday, Dec 06, 2022 at 4:59 AM, Nux  wrote:
>> > >
>> > > You can't alter the kernel parameters from libvirt as the image loads
>> > > its own kernel, but you can alter the parameters at their grub screen.
>> > > Try to open the console proxy as soon as possible and hit up or down
>> > > arrow so grub doesn't proceed, then hit "e" key (I think) to edit the
>> > > appropriate vmlinuz entry.
>> > >
>> > > My hunch is that this being a generic image (made mostly with EC2 and
>> > > possibly Openstack in mind) they send all output to a serial console
>> > > instead of vga, so you would need to remove any parameters such as
>> > > console=ttyS0 and then boot it.
>> > >
>> > > ---
>> > > Nux
>> > > www.nux.ro [1]
>> > >
>> > > On 2022-12-06 04:26, Jeremy Hansen wrote:
>> > >
>> > > I'm trying to boot a Rocky cloud image for Rocky 9.1 using their qcow2
>> > > image.
>> > >
>> > > As soon as I start the VM, I get:
>> > >
>> > > "Probing EDD (edd=off to disable)... ok"
>> > >
>> > > and then nothing. I don't see a DHCP lease being pulled, so I assume
>> > > the VM isn't actually proceeding with boot.
>> > >
>> > > Anyone else see this? This is on Cloudstack 4.17.1.0. I posted a
>> > > similar message on the Rocky mail list. Is there an easy way via virsh
>> > > or cloudstack to alter the kernel command line parameters in an
>> > > existing image before it boots?
>> > >
>> > > Thanks
>> > > -jeremy
>> > >
>> > > Links:
>> > > --
>> > > [1] http://www.nux.ro
>> >
>> >
>> > Links:
>> > --
>> > [1] http://www.nux.ro

Re: Issues with Rocky 9.1 qcow2 image

[Jeremy Hansen]

Strange, I tried launching the Rocky 9.1 ISO to build a new image and it also 
kernel panics upon boot. Not sure what the issue is. My Rocky 8 images seem 
fine.

Hmmm…

-jeremy

> On Tuesday, Dec 06, 2022 at 1:41 PM, Jeremy Hansen  (mailto:jer...@skidrow.la)> wrote:
> Unfortunately I’m getting a kernel panic upon boot. I tried both the 
> production grub option and the rescue and I get the same result. Maybe 
> something with the size of the root disk? I have it just at 8 gig right now? 
> Something with the initrd perhaps?
>
> Thanks!
> -jeremy
>
>
>
>
>
> > On Tuesday, Dec 06, 2022 at 12:53 PM, Jeremy Hansen  > (mailto:jer...@skidrow.la)> wrote:
> > Wow. Thank you! I appreciate this. I’ll report back after trying.
> >
> > -jeremy
> >
> >
> >
> > > On Tuesday, Dec 06, 2022 at 11:39 AM, Nux  > > (mailto:n...@li.nux.ro)> wrote:
> > >
> > >
> > > God knows, I've built you one with the Cloudstack bits enabled, you can
> > > grab it from below. It's got the root login enabled, so that's your
> > > default user. Let me know if it works.
> > >
> > > http://dl.openvm.eu/cloudstack/rockylinux/9/rockylinux-9-kvm-rootuser.qcow2.bz2
> > >
> > > I'll build more later on with non-priviliged users.
> > >
> > > ---
> > > Nux
> > > www.nux.ro [1]
> > >
> > > On 2022-12-06 15:26, Jeremy Hansen wrote:
> > >
> > > > Unfortunately I think it gets stuck before ever reaching grub. I tried
> > > > to catch it on the console after a reboot and this is all I see. I
> > > > never get to the grub menu.
> > > >
> > > > Is there a way I can alter the qcow2 image before I create a template?
> > > >
> > > > I thought perhaps I could see the IP it obtains via dhcp, but I see
> > > > nothing coming from that VM which tells me it's not getting to the
> > > > point of bringing up its interface.
> > > >
> > > > Thanks
> > > > -jeremy
> > > >
> > > > On Tuesday, Dec 06, 2022 at 4:59 AM, Nux  wrote:
> > > >
> > > > You can't alter the kernel parameters from libvirt as the image loads
> > > > its own kernel, but you can alter the parameters at their grub screen.
> > > > Try to open the console proxy as soon as possible and hit up or down
> > > > arrow so grub doesn't proceed, then hit "e" key (I think) to edit the
> > > > appropriate vmlinuz entry.
> > > >
> > > > My hunch is that this being a generic image (made mostly with EC2 and
> > > > possibly Openstack in mind) they send all output to a serial console
> > > > instead of vga, so you would need to remove any parameters such as
> > > > console=ttyS0 and then boot it.
> > > >
> > > > ---
> > > > Nux
> > > > www.nux.ro [1]
> > > >
> > > > On 2022-12-06 04:26, Jeremy Hansen wrote:
> > > >
> > > > I'm trying to boot a Rocky cloud image for Rocky 9.1 using their qcow2
> > > > image.
> > > >
> > > > As soon as I start the VM, I get:
> > > >
> > > > "Probing EDD (edd=off to disable)... ok"
> > > >
> > > > and then nothing. I don't see a DHCP lease being pulled, so I assume
> > > > the VM isn't actually proceeding with boot.
> > > >
> > > > Anyone else see this? This is on Cloudstack 4.17.1.0. I posted a
> > > > similar message on the Rocky mail list. Is there an easy way via virsh
> > > > or cloudstack to alter the kernel command line parameters in an
> > > > existing image before it boots?
> > > >
> > > > Thanks
> > > > -jeremy
> > > >
> > > > Links:
> > > > --
> > > > [1] http://www.nux.ro
> > >
> > >
> > > Links:
> > > --
> > > [1] http://www.nux.ro


signature.asc
Description: PGP signature

Re: Redirect HTTP 8080 to HTTPS 8443 on Cloudstack 4.17.1.0

[Darren Cole]

At one point I tried using Cloudstack for terminating the SSL, but have since 
switched to Apache because that is easier. 
Apache handles SSL termination and proxies the 443 connection to Cloudstack 
localhost 8080. 
I have Cloudstack only binding 8080 to localhost, so it is not accessible 
except over SSL. 

I have not gotten around to configuring the console access to work. 
Both clusters I've run are very much testing only side projects. 

Darren 
-- 
This e-mail is confidential. Any distribution, use or copying of this e-mail or 
the information it contains other than by the intended recipient is forbidden. 
If you are not the intended recipient, please advise the sender (by return 
e-mail or otherwise) immediately and delete this e-mail. 


From: "Nux"  
To: "users"  
Sent: Monday, December 5, 2022 5:36:09 AM 
Subject: Re: Redirect HTTP 8080 to HTTPS 8443 on Cloudstack 4.17.1.0 



I never bothered to do this in Cloudstack, I always used Apache to terminate 
SSL and do proxy requests, you can of course use any other web servers for this 
(nginx, lighttpd). 
--- 
Nux 
[ http://www.nux.ro/ | www.nux.ro ] 





On 2022-12-02 13:58, Granwille Strauss wrote: 




Hi Guys 

Would like to follow up on this any ideas? 
On 11/18/22 16:47, David Larsen wrote: 

BQ_BEGIN

Hi

We are using Cloudstack 4.17.1.0. [ https://host.domain.com:8443/ | 
Https://host.domain.com:8443 ] works fine.

I followed to steps below to redirect http 8080 to https 8443 with no luck
Is there another way to do this in 4.17.1.0?
Hope so... :-)



For auto redirection from 8080 to 8443, Add below content in 
'/usr/share/cloudstack-management/webapp/WEB-INF/web.xml' file at line 22




Everything in the webapp

/*





CONFIDENTIAL












...

8443






Restart the management service
systemctl restart cloudstack-management


Best regards
David Larsen 



-- 
Regards / Groete 


[ https://www.namhost.com/ ]Granwille Strauss // Senior Systems Admin 

e: [ mailto:granwi...@namhost.com | granwi...@namhost.com ] 
m: +264 81 323 1260 
w: [ https://www.namhost.com/ | www.namhost.com ] 

[ https://www.facebook.com/namhost ] [ https://twitter.com/namhost ] [ 
https://www.instagram.com/namhostinternetservices/ ] [ 
https://www.linkedin.com/company/namhos ] [ 
https://www.youtube.com/channel/UCTd5v-kVPaic_dguGur15AA ] 

[ 
https://www.adsigner.com/v1/l/631091998d4670001fe43ec2/621c9b76c140bb001ed0f818/banner
 ] 


Namhost Internet Services (Pty) Ltd , 
24 Black Eagle Rd, Hermanus, 7210, RSA 


The content of this message is confidential. If you have received it by 
mistake, please inform us by email reply and then delete the message. It is 
forbidden to copy, forward, or in any way reveal the contents of this message 
to anyone without our explicit consent. The integrity and security of this 
email cannot be guaranteed over the Internet. Therefore, the sender will not be 
held liable for any damage caused by the message. For our full privacy policy 
and disclaimers, please go to [ https://www.namhost.com/privacy-policy | 
https://www.namhost.com/privacy-policy ] 
[ 
https://www.adsigner.com/v1/c/631091998d4670001fe43ec2/621c9b76c140bb001ed0f818 
] 

BQ_END

Re: Issues with Rocky 9.1 qcow2 image

[Jeremy Hansen]

Wow. Thank you! I appreciate this. I’ll report back after trying.

-jeremy

> On Tuesday, Dec 06, 2022 at 11:39 AM, Nux  (mailto:n...@li.nux.ro)> wrote:
>
>
> God knows, I've built you one with the Cloudstack bits enabled, you can
> grab it from below. It's got the root login enabled, so that's your
> default user. Let me know if it works.
>
> http://dl.openvm.eu/cloudstack/rockylinux/9/rockylinux-9-kvm-rootuser.qcow2.bz2
>
> I'll build more later on with non-priviliged users.
>
> ---
> Nux
> www.nux.ro [1]
>
> On 2022-12-06 15:26, Jeremy Hansen wrote:
>
> > Unfortunately I think it gets stuck before ever reaching grub. I tried
> > to catch it on the console after a reboot and this is all I see. I
> > never get to the grub menu.
> >
> > Is there a way I can alter the qcow2 image before I create a template?
> >
> > I thought perhaps I could see the IP it obtains via dhcp, but I see
> > nothing coming from that VM which tells me it's not getting to the
> > point of bringing up its interface.
> >
> > Thanks
> > -jeremy
> >
> > On Tuesday, Dec 06, 2022 at 4:59 AM, Nux  wrote:
> >
> > You can't alter the kernel parameters from libvirt as the image loads
> > its own kernel, but you can alter the parameters at their grub screen.
> > Try to open the console proxy as soon as possible and hit up or down
> > arrow so grub doesn't proceed, then hit "e" key (I think) to edit the
> > appropriate vmlinuz entry.
> >
> > My hunch is that this being a generic image (made mostly with EC2 and
> > possibly Openstack in mind) they send all output to a serial console
> > instead of vga, so you would need to remove any parameters such as
> > console=ttyS0 and then boot it.
> >
> > ---
> > Nux
> > www.nux.ro [1]
> >
> > On 2022-12-06 04:26, Jeremy Hansen wrote:
> >
> > I'm trying to boot a Rocky cloud image for Rocky 9.1 using their qcow2
> > image.
> >
> > As soon as I start the VM, I get:
> >
> > "Probing EDD (edd=off to disable)... ok"
> >
> > and then nothing. I don't see a DHCP lease being pulled, so I assume
> > the VM isn't actually proceeding with boot.
> >
> > Anyone else see this? This is on Cloudstack 4.17.1.0. I posted a
> > similar message on the Rocky mail list. Is there an easy way via virsh
> > or cloudstack to alter the kernel command line parameters in an
> > existing image before it boots?
> >
> > Thanks
> > -jeremy
> >
> > Links:
> > --
> > [1] http://www.nux.ro
>
>
> Links:
> --
> [1] http://www.nux.ro


signature.asc
Description: PGP signature

Re: Issues with Rocky 9.1 qcow2 image

[Nux]

God knows, I've built you one with the Cloudstack bits enabled, you can 
grab it from below. It's got the root login enabled, so that's your 
default user. Let me know if it works.


http://dl.openvm.eu/cloudstack/rockylinux/9/rockylinux-9-kvm-rootuser.qcow2.bz2

I'll build more later on with non-priviliged users.

---
Nux
www.nux.ro [1]

On 2022-12-06 15:26, Jeremy Hansen wrote:

Unfortunately I think it gets stuck before ever reaching grub. I tried 
to catch it on the console after a reboot and this is all I see. I 
never get to the grub menu.


Is there a way I can alter the qcow2 image before I create a template?

I thought perhaps I could see the IP it obtains via dhcp, but I see 
nothing coming from that VM which tells me it's not getting to the 
point of bringing up its interface.


Thanks
-jeremy

On Tuesday, Dec 06, 2022 at 4:59 AM, Nux  wrote:

You can't alter the kernel parameters from libvirt as the image loads
its own kernel, but you can alter the parameters at their grub screen.
Try to open the console proxy as soon as possible and hit up or down
arrow so grub doesn't proceed, then hit "e" key (I think) to edit the
appropriate vmlinuz entry.

My hunch is that this being a generic image (made mostly with EC2 and
possibly Openstack in mind) they send all output to a serial console
instead of vga, so you would need to remove any parameters such as
console=ttyS0 and then boot it.

---
Nux
www.nux.ro [1]

On 2022-12-06 04:26, Jeremy Hansen wrote:

I'm trying to boot a Rocky cloud image for Rocky 9.1 using their qcow2
image.

As soon as I start the VM, I get:

"Probing EDD (edd=off to disable)... ok"

and then nothing. I don't see a DHCP lease being pulled, so I assume
the VM isn't actually proceeding with boot.

Anyone else see this? This is on Cloudstack 4.17.1.0. I posted a
similar message on the Rocky mail list. Is there an easy way via virsh
or cloudstack to alter the kernel command line parameters in an
existing image before it boots?

Thanks
-jeremy

Links:
--
[1] http://www.nux.ro



Links:
--
[1] http://www.nux.ro

RE: VM instances and RVR cannot communicate with each other when on different hosts

[Alex Mattioli]

From my experience I had a few issues with RVRs myself, some of those were:

  1.  They’d fail to elect a master and continuous flip over the connection
  2.  When going anywhere above 1gbps the Master would fail to send broadcasts 
messages and the Backup VR would become master, ending up with 2 masters and a 
messy situation.
  3.  In ESX, depending on version, you might need to tweak your security 
parameters for the port group.

Am sure that Andrija has more examples.

Cheers
Alex

From: Gary Dixon 
Sent: 06 December 2022 15:25
To: users@cloudstack.apache.org
Subject: RE: VM instances and RVR cannot communicate with each other when on 
different hosts

Thanks for the response Andrija

Do you have any real world examples of the RVR feature 'breaking' or any 
further info as to why they are not recommended in a production environment /

BR

Gary
Gary Dixon​
Senior Technical Consultant
T:  +44 161 537 4990
E:  vms@quadris‑support.com
W: www.quadris.co.uk
[cid:image784358.png@1C346EC9.9B031216]
The information contained in this e-mail from Quadris may be confidential and 
privileged for the private use of the named recipient.  The contents of this 
e-mail may not necessarily represent the official views of Quadris.  If you 
have received this information in error you must not copy, distribute or take 
any action or reliance on its contents.  Please destroy any hard copies and 
delete this message.
 


-Original Message-
From: Andrija Panic mailto:andrija.pa...@gmail.com>>
Sent: 05 December 2022 12:59
To: users@cloudstack.apache.org
Subject: Re: VM instances and RVR cannot communicate with each other when on 
different hosts

Hi,

RVR is a feature that is known to break from time to time, and is NOT 
recommended for the Production use, at least that's what we advise your 
customers to do/avoid doing.

However, if 2 VMs can not communicate while on different hosts, but CAN 
communicate while on the same host - this might indicate VLAN (trunking) 
misconfiguration on the switches - the VLANs via which those VMs are 
communicating, has to be trunked to all the hypervisors in your Zone - so that 
host1 and hostN both can talk over the specific VLAN (here, I'm guessing you 
are hitting issues with Guest traffic - so check that all your VLANs for Guest 
traffic are properly trunked on all switch ports to which all of your servers 
are connected)

Best,

On Fri, 18 Nov 2022 at 15:08, Gary Dixon 
mailto:gary.di...@quadris.co.uk.invalid>>
wrote:

> Hi
>
>
>
> I am hoping someone could help with a new Dev Cloudstack system we are
> trying to setup based on Ubuntu 20.04 KVM hosts and mgmt. servers with
> CS
> 4.15.2 and an Adv Zone with VPC’s
>
>
>
> We spotted that the RVR’s in the VPC’s are both in the ‘MASTER’ state
> indicating that they cannot communicate with each other. Also testing
> within a guest VM – it is able to ping another guest VM in the same
> network – but only when on the same KVM host.
>
> If we live migrate one of the VM’s to a different KVM host then the
> ping breaks.
>
> Our guest network is using VXLAN isolation method and all network
> labesl in CS are correct. We are trying to setup this Dev system to
> match our production system as closely as possible and all networking
> works perfectly in the Production system
>
>
>
> One thing we have noticed is if we put a KVM host into maintenance
> mode – thus destroying all the brvx-xxx interfaces on there and then
> bring it back out of maintenance mode and migrate VM’s to it and
> Restart the VPC with cleanup enabled – communication between VM’s
> cross host works again for a few minutes. The RVR’s go into Master and
> Backup status briefly – but then after a few minutes both VR’s go into
> ‘MASTER’ state and VM’s cannot ping each other when on different hosts.
>
> Any pointers/help would be greatly appreciated
>
>
>
> BR
>
>
>
> Gary
> Gary Dixon​
> Senior Technical Consultant
> T: +44 161 537 4990
> E: *v* 
> <+44%207989717661>ms@quadris‑support.com
> W:
> https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.q
> uadris.co.uk%2Fdata=05%7C01%7CGary.Dixon%40quadris.co.uk%7Cefdf1f
> 0e844d4aee130308dad6c081c2%7Cf1d6abf3d3b44894ae16db0fb93a96a2%7C0%7C0%
> 7C638058419553464258%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQI
> joiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7Csdata=Fq
> bSS%2FHgLkhzM0UTyBY5eH3bepYiwjcvmvIa4%2BWTDds%3Dreserved=0
> The information contained in this e-mail from Quadris may be
> confidential and privileged for the private use of the named
> recipient. The contents of this e-mail may not necessarily represent the 
> official views of Quadris.
> If you have received this information in error you must not copy,
> distribute or take any action or reliance on its contents. Please
> destroy any hard copies and delete this message.
>


--

Andrija Panić

Re: Issues with Rocky 9.1 qcow2 image

[Jeremy Hansen]

Unfortunately I think it gets stuck before ever reaching grub. I tried to catch 
it on the console after a reboot and this is all I see. I never get to the grub 
menu.

Is there a way I can alter the qcow2 image before I create a template?

I thought perhaps I could see the IP it obtains via dhcp, but I see nothing 
coming from that VM which tells me it’s not getting to the point of bringing up 
its interface.

Thanks
-jeremy

> On Tuesday, Dec 06, 2022 at 4:59 AM, Nux  (mailto:n...@li.nux.ro)> wrote:
>
>
> You can't alter the kernel parameters from libvirt as the image loads
> its own kernel, but you can alter the parameters at their grub screen.
> Try to open the console proxy as soon as possible and hit up or down
> arrow so grub doesn't proceed, then hit "e" key (I think) to edit the
> appropriate vmlinuz entry.
>
> My hunch is that this being a generic image (made mostly with EC2 and
> possibly Openstack in mind) they send all output to a serial console
> instead of vga, so you would need to remove any parameters such as
> console=ttyS0 and then boot it.
>
> ---
> Nux
> www.nux.ro [1]
>
> On 2022-12-06 04:26, Jeremy Hansen wrote:
>
> > I'm trying to boot a Rocky cloud image for Rocky 9.1 using their qcow2
> > image.
> >
> > As soon as I start the VM, I get:
> >
> > "Probing EDD (edd=off to disable)... ok"
> >
> > and then nothing. I don't see a DHCP lease being pulled, so I assume
> > the VM isn't actually proceeding with boot.
> >
> > Anyone else see this? This is on Cloudstack 4.17.1.0. I posted a
> > similar message on the Rocky mail list. Is there an easy way via virsh
> > or cloudstack to alter the kernel command line parameters in an
> > existing image before it boots?
> >
> > Thanks
> > -jeremy
>
>
> Links:
> --
> [1] http://www.nux.ro


signature.asc
Description: PGP signature

RE: VM instances and RVR cannot communicate with each other when on different hosts

[Gary Dixon]

Thanks for the response Andrija

Do you have any real world examples of the RVR feature 'breaking' or any 
further info as to why they are not recommended in a production environment /

BR

Gary


Gary Dixon
Senior Technical Consultant
T:  +44 161 537 4990
E:  v...@quadris-support.com
W: www.quadris.co.uk
The information contained in this e-mail from Quadris may be confidential and 
privileged for the private use of the named recipient.  The contents of this 
e-mail may not necessarily represent the official views of Quadris.  If you 
have received this information in error you must not copy, distribute or take 
any action or reliance on its contents.  Please destroy any hard copies and 
delete this message.
-Original Message-
From: Andrija Panic 
Sent: 05 December 2022 12:59
To: users@cloudstack.apache.org
Subject: Re: VM instances and RVR cannot communicate with each other when on 
different hosts

Hi,

RVR is a feature that is known to break from time to time, and is NOT 
recommended for the Production use, at least that's what we advise your 
customers to do/avoid doing.

However, if 2 VMs can not communicate while on different hosts, but CAN 
communicate while on the same host - this might indicate VLAN (trunking) 
misconfiguration on the switches - the VLANs via which those VMs are 
communicating, has to be trunked to all the hypervisors in your Zone - so that 
host1 and hostN both can talk over the specific VLAN (here, I'm guessing you 
are hitting issues with Guest traffic - so check that all your VLANs for Guest 
traffic are properly trunked on all switch ports to which all of your servers 
are connected)

Best,

On Fri, 18 Nov 2022 at 15:08, Gary Dixon 
wrote:

> Hi
>
>
>
> I am hoping someone could help with a new Dev Cloudstack system we are
> trying to setup based on Ubuntu 20.04 KVM hosts and mgmt. servers with
> CS
> 4.15.2 and an Adv Zone with VPC’s
>
>
>
> We spotted that the RVR’s in the VPC’s are both in the ‘MASTER’ state
> indicating that they cannot communicate with each other. Also testing
> within a guest VM – it is able to ping another guest VM in the same
> network – but only when on the same KVM host.
>
> If we live migrate one of the VM’s to a different KVM host then the
> ping breaks.
>
> Our guest network is using VXLAN isolation method and all network
> labesl in CS are correct. We  are trying to setup this Dev system to
> match our production system as closely as possible and all networking
> works perfectly in the Production system
>
>
>
> One thing we have noticed is if we put a KVM host into maintenance
> mode – thus destroying all the brvx-xxx interfaces on there and then
> bring it back out of maintenance mode and migrate VM’s to it and
> Restart the VPC with cleanup enabled – communication between VM’s
> cross host works again for a few minutes. The RVR’s go into Master and
> Backup status briefly – but then after a few minutes both VR’s go into
> ‘MASTER’ state and VM’s cannot ping each other when on different hosts.
>
> Any pointers/help would be greatly appreciated
>
>
>
> BR
>
>
>
> Gary
> Gary Dixon​
> Senior Technical Consultant
> T:  +44 161 537 4990
> E:  *v* <+44%207989717661>ms@quadris‑support.com
> W:
> https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.q
> uadris.co.uk%2Fdata=05%7C01%7CGary.Dixon%40quadris.co.uk%7Cefdf1f
> 0e844d4aee130308dad6c081c2%7Cf1d6abf3d3b44894ae16db0fb93a96a2%7C0%7C0%
> 7C638058419553464258%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQI
> joiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7Csdata=Fq
> bSS%2FHgLkhzM0UTyBY5eH3bepYiwjcvmvIa4%2BWTDds%3Dreserved=0
> The information contained in this e-mail from Quadris may be
> confidential and privileged for the private use of the named
> recipient.  The contents of this e-mail may not necessarily represent the 
> official views of Quadris.
> If you have received this information in error you must not copy,
> distribute or take any action or reliance on its contents.  Please
> destroy any hard copies and delete this message.
>


--

Andrija Panić

RE: Multiple secondary storage VMs are created out of nowhere

[Alex Mattioli]

Hi,
CloudStack automatically auto scales your SSVMs as necessary, there are two 
global settings that control that, namely:

secstorage.capacity.standby The minimal number of command execution 
sessions that system is able to serve immediately(standby capacity)
secstorage.session.max  The max number of command execution sessions that a 
SSVM can handle

If your number of sessions is > ( secstorage.session.max- 
secstorage.capacity.standby) then ACS will deploy new SSVMs until you have 
enough standby capacity.

This could happen if ACS suddenly has to deal with many more session than it 
typically does. You can safely delete the extra SSVMs.

Regards
Alex

 


-Original Message-
From: Mevludin Blazevic  
Sent: 06 December 2022 14:14
To: users@cloudstack.apache.org
Subject: Multiple secondary storage VMs are created out of nowhere

Hi all,

after I added another secondary storage NFS server to my zone to deleted the 
old secondary storage, it seems like Cloudstack started another 5 secondary 
storage VMs along the "old" one. On my test environment I did not see this 
behavior. Moreover, in the docs, I cannot see what's triggering the management 
server to start multiple system vms. Any ideas?

Thanks,

Mevludin

Multiple secondary storage VMs are created out of nowhere

[Mevludin Blazevic]

Hi all,

after I added another secondary storage NFS server to my zone to deleted 
the old secondary storage, it seems like Cloudstack started another 5 
secondary storage VMs along the "old" one. On my test environment I did 
not see this behavior. Moreover, in the docs, I cannot see what's 
triggering the management server to start multiple system vms. Any ideas?


Thanks,

Mevludin

Re: Consoleproxy VM is no longer created

[Nux]

There could be a few reasons.
Make sure your management and hypervisor logs are set to DEBUG when you 
are investigating.
If nothing shows up on the management side, keep an eye on the 
hypervisor, see if a VM creation is even attempted, check libvirt (or 
whatever) logs etc.


---
Nux
www.nux.ro

On 2022-12-06 04:37, Antoine Boucher wrote:

Hello,

We had a failure with a secondary storage that created several issues.

Nevertheless, after fixing the issues, we discovered that the system
consoleproxy and secondarystorage vm were hung.

We deleted both system VMs, the secondarystorage vm came back without
issues, but the consoleproxy vm is not being created.

We will investigate further, but we have found nothing out of the
ordinary in the management log file so far.

Has anyone had a similar issue before?

Regards,
Antoine




Confidentiality Warning: This message and any attachments are intended
only for the use of the intended recipient(s), are confidential, and
may be privileged. If you are not the intended recipient, you are
hereby notified that any review, retransmission, conversion to hard
copy, copying, circulation or other use of this message and any
attachments is strictly prohibited. If you are not the intended
recipient, please notify the sender immediately by return e-mail, and
delete this message and any attachments from your system.

Re: VM Volume Limit 1024 GB

[Slavka Peleva]

Hi Rafael,

you can increase the value of the
global setting `custom.diskoffering.size.max`

Best regards,
Slavka

On Tue, Dec 6, 2022 at 12:18 PM Nux  wrote:

> Check the user or domain limits.
>
> ---
> Nux
> www.nux.ro
>
> On 2022-12-06 10:15, Rafael del Valle wrote:
> > Hi!
> >
> > I ​cannot create an instance with 2000 GB data volume despite
> > available resources in ACS 4.16.1.
> >
> > I get a message:
> >
> > VM Creation failed. Volume size: 2000GB is out of allowed range. Max:
> > 1024 Min:1
> >
> > The configuration parameter storage.max.volume.size is 2000
> >
> >
> >
> > I cannot see any related account limit either.
> >
> > The volume type is Local Storage.
> >
> >
> > Where is the limit coming from?
> >
> >
> > R.
>

Re: VM Volume Limit 1024 GB

[Nux]

Check the user or domain limits.

---
Nux
www.nux.ro

On 2022-12-06 10:15, Rafael del Valle wrote:

Hi!

I ​cannot create an instance with 2000 GB data volume despite
available resources in ACS 4.16.1.

I get a message:

VM Creation failed. Volume size: 2000GB is out of allowed range. Max: 
1024 Min:1


The configuration parameter storage.max.volume.size is 2000



I cannot see any related account limit either.

The volume type is Local Storage.


Where is the limit coming from?


R.

VM Volume Limit 1024 GB

[Rafael del Valle]

Hi!

I ​cannot create an instance with 2000 GB data volume despite available 
resources in ACS 4.16.1.

I get a message:

VM Creation failed. Volume size: 2000GB is out of allowed range. Max: 1024 Min:1

The configuration parameter storage.max.volume.size is 2000



I cannot see any related account limit either.

The volume type is Local Storage.


Where is the limit coming from?


R.

Re: Issues with Rocky 9.1 qcow2 image

[Nux]

You can't alter the kernel parameters from libvirt as the image loads 
its own kernel, but you can alter the parameters at their grub screen. 
Try to open the console proxy as soon as possible and hit up or down 
arrow so grub doesn't proceed, then hit "e" key (I think) to edit the 
appropriate vmlinuz entry.


My hunch is that this being a generic image (made mostly with EC2 and 
possibly Openstack in mind) they send all output to a serial console 
instead of vga, so you would need to remove any parameters such as 
console=ttyS0 and then boot it.


---
Nux
www.nux.ro [1]

On 2022-12-06 04:26, Jeremy Hansen wrote:

I'm trying to boot a Rocky cloud image for Rocky 9.1 using their qcow2 
image.


As soon as I start the VM, I get:

"Probing EDD (edd=off to disable)... ok"

and then nothing.  I don't see a DHCP lease being pulled, so I assume 
the VM isn't actually proceeding with boot.


Anyone else see this?  This is on Cloudstack 4.17.1.0.  I posted a 
similar message on the Rocky mail list.  Is there an easy way via virsh 
or cloudstack to alter the kernel command line parameters in an 
existing image before it boots?


Thanks
-jeremy



Links:
--
[1] http://www.nux.ro