RE: Cloudstack Object Storage

[Alexandru Stan]

Thanks Wei!


-Original Message-
From: Wei ZHOU  
Sent: Tuesday, September 10, 2024 2:28 PM
To: users@cloudstack.apache.org
Subject: Re: Cloudstack Object Storage

It will be supported in 4.20
https://github.com/apache/cloudstack/pull/8389

-Wei

On Tue, Sep 10, 2024 at 1:25 PM Alexandru Stan 
 wrote:

> Hi everyone. Could someone please clarify what object storage 
> solutions are currently supported in ACS? Can we add something else 
> other than MinIO in the Add Object Storage section, like via Ceph RGW?
>
> Thank you!
>

Cannot copy template to 2nd zone

[Alexandru Stan]

Hello,

I have a problem with the Copy template feature in ACS, maybe someone here can 
help. I have 2 zones set up and when creating templates from a volume there is 
no option to choose zones, like it is when registering a template from URL. So 
it gets created in the zone that the VM is in. Then I want to make the template 
available for the other zone too, but I always get time out after a few 
minutes. There IS communication between the secondary storage of the first zone 
and the secondary storage of the second zone, but maybe there is something else 
that needs to be done.

Thank you!

RE: Cloud-Init not working as per Cloudstack documentation

[Alexandru Stan]

-07 22:47:05,331 - rhel.py[DEBUG]: Using DNF for package 
> management
> 2024-08-07 22:47:05,331 - subp.py[DEBUG]: Running command ['dnf', 
> '-y', 'upgrade'] with allowed return codes [0] (shell=False, 
> capture=False)
> 2024-08-07 22:47:07,103 - rhel.py[DEBUG]: Using DNF for package 
> management
> 2024-08-07 22:47:07,103 - subp.py[DEBUG]: Running command ['dnf', 
> '-y', 'install', 'epel-release', 'neofetch', 'htop', 'gcc', 'gcc-c++', 
> 'perl'] with allowed return codes [0] (shell=False, capture=False)
> 2024-08-07 22:47:08,573 - util.py[WARNING]: Failed to install packages:
> ['epel-release', 'neofetch', 'htop', 'gcc', 'gcc-c++', 'perl']
> 2024-08-07 22:47:08,575 - util.py[DEBUG]: Failed to install packages:
> ['epel-release', 'neofetch', 'htop', 'gcc', 'gcc-c++', 'perl'] 
> Traceback (most recent call last):
>   File
> "/usr/lib/python3.6/site-packages/cloudinit/config/cc_package_update_u
> pgrade_install.py",
> line 121, in handle
> cloud.distro.install_packages(pkglist)
>   File "/usr/lib/python3.6/site-packages/cloudinit/distros/rhel.py", 
> line 59, in install_packages
> self.package_command("install", pkgs=pkglist)
>   File "/usr/lib/python3.6/site-packages/cloudinit/distros/rhel.py", 
> line 201, in package_command
> subp.subp(cmd, capture=False)
>   File "/usr/lib/python3.6/site-packages/cloudinit/subp.py", line 323, 
> in subp
> stdout=out, stderr=err, exit_code=rc, cmd=args
> cloudinit.subp.ProcessExecutionError: Unexpected error while running 
> command.
> Command: ['dnf', '-y', 'install', 'epel-release', 'neofetch', 'htop', 
> 'gcc', 'gcc-c++', 'perl'] Exit code: 1
> Reason: -
> Stdout: -
> Stderr: -
> 2024-08-07 22:47:08,582 - 
> cc_package_update_upgrade_install.py[WARNING]: 1 failed with 
> exceptions, re-raising the last one
> 2024-08-07 22:47:08,582 - handlers.py[DEBUG]: finish:
> modules-final/config-package_update_upgrade_install: FAIL: running 
> config-package_update_upgrade_install with frequency once-per-instance
> 2024-08-07 22:47:08,582 - util.py[WARNING]: Running module 
> package_update_upgrade_install ( 'cloudinit.config.cc_package_update_upgrade_install' from
> '/usr/lib/python3.6/site-packages/cloudinit/config/cc_package_update_u
> pgrade_install.py'>)
> failed
> 2024-08-07 22:47:08,582 - util.py[DEBUG]: Running module 
> package_update_upgrade_install ( 'cloudinit.config.cc_package_update_upgrade_install' from
> '/usr/lib/python3.6/site-packages/cloudinit/config/cc_package_update_u
> pgrade_install.py'>)
> failed
>
>
> My main concern is the hostname that I cannot seem to set no matter 
> what I try, if I can get this at least right that would be the 
> ultimate. How can I pass the update_hostname userdata so that I can 
> manually enter custom hostnames when deploying  VM via UI?
> On 8/7/24 22:52, Alexandru Stan wrote:
>
> Hello,
>
> Can’t say about everything you mentioned, but regarding the disk resizinng I 
> can share my experience, I also had some issues before I managed to get it 
> working. So, from what I’ve searched, growpart does not seem to work with lvm 
> (could be wrong here though), so basic partitioning it is. Then, it seems 
> that it extends the partition only if it’s the last one on the disk, seems to 
> be your issue since the last partition is the swap partition. Try with the 
> root partition being the last one, it should work fine.
>
>
> From: Granwille Strauss  
>   
> 
> Sent: 7 August, 2024 20:35
> To: users@cloudstack.apache.org
> Subject: Cloud-Init not working as per Cloudstack documentation
>
>
> Evening,
>
> Please note, I do not use any API service and solely rely on the Cloudstack 
> UI.
>
> I deployed a new clean template and I followed the cloudstack documentation 
> thoroughly to use the basics of cloud-init with cloudstack. But when I 
> deployed my VM nothing seems to trigger at all, including userdata I 
> registered.
> pwd
> /etc/cloud/cloud.cfg.d
> [root@hostname cloud.cfg.d]# ll
> total 32
> -rw-r--r--. 1 root root 2070 Dec  4  2023 05_logging.cfg
> -rw-r--r--. 1 root root   22 Aug  7 17:43 49_hostkeys.cfg
> -rw-r--r--. 1 root root   88 Aug  7 17:45 50_growpartion.cfg
> -rw-r--r--. 1 root root  209 Aug  7 17:52 51_extend_volume.cfg 
> -rw-r--r--. 1 root root  143 Aug  7 17:55 52_upgrade_packages.cfg 
> -rw-r--r-

RE: Cloud-Init not working as per Cloudstack documentation

[Alexandru Stan]

Hello,

Can’t say about everything you mentioned, but regarding the disk resizinng I 
can share my experience, I also had some issues before I managed to get it 
working. So, from what I’ve searched, growpart does not seem to work with lvm 
(could be wrong here though), so basic partitioning it is. Then, it seems that 
it extends the partition only if it’s the last one on the disk, seems to be 
your issue since the last partition is the swap partition. Try with the root 
partition being the last one, it should work fine.


From: Granwille Strauss 
Sent: 7 August, 2024 20:35
To: users@cloudstack.apache.org
Subject: Cloud-Init not working as per Cloudstack documentation


Evening,

Please note, I do not use any API service and solely rely on the Cloudstack UI.

I deployed a new clean template and I followed the cloudstack documentation 
thoroughly to use the basics of cloud-init with cloudstack. But when I deployed 
my VM nothing seems to trigger at all, including userdata I registered.
pwd
/etc/cloud/cloud.cfg.d
[root@hostname cloud.cfg.d]# ll
total 32
-rw-r--r--. 1 root root 2070 Dec  4  2023 05_logging.cfg
-rw-r--r--. 1 root root   22 Aug  7 17:43 49_hostkeys.cfg
-rw-r--r--. 1 root root   88 Aug  7 17:45 50_growpartion.cfg
-rw-r--r--. 1 root root  209 Aug  7 17:52 51_extend_volume.cfg
-rw-r--r--. 1 root root  143 Aug  7 17:55 52_upgrade_packages.cfg
-rw-r--r--. 1 root root  355 Aug  7 18:00 53_install_packages.cfg
-rw-r--r--. 1 root root  357 Aug  7 17:42 80_user.cfg
-rw-r--r--. 1 root root  167 Dec  4  2023 README


For example, I deployed a new VM with its root disk overwrite from the 
template, which was 10 GB to 60 GB and the 50_growpartion.cfg did nothing:
[root@hostname cloud.cfg.d]# cat 50_growpartion.cfg
growpart:
  mode: auto
  devices:
- \"/dev/vda2\"
  ignore_growroot_disabled: false
cat 51_extend_volume.cfg
runcmd:
  - [ cloud-init-per, always, grow_FS, resize2fs, /dev/vda2 ]

grep "50_growpartion.cfg" /var/log/cloud-init.log
2024-08-07 16:50:45,881 - util.py[DEBUG]: Reading from 
/etc/cloud/cloud.cfg.d/50_growpartion.cfg (quiet=False)
2024-08-07 16:50:45,881 - util.py[DEBUG]: Read 88 bytes from 
/etc/cloud/cloud.cfg.d/50_growpartion.cfg
2024-08-07 16:50:45,928 - util.py[DEBUG]: Reading from 
/etc/cloud/cloud.cfg.d/50_growpartion.cfg (quiet=False)
2024-08-07 16:50:45,928 - util.py[DEBUG]: Read 88 bytes from 
/etc/cloud/cloud.cfg.d/50_growpartion.cfg
2024-08-07 16:50:46,046 - util.py[DEBUG]: Reading from 
/etc/cloud/cloud.cfg.d/50_growpartion.cfg (quiet=False)
2024-08-07 16:50:46,046 - util.py[DEBUG]: Read 88 bytes from 
/etc/cloud/cloud.cfg.d/50_growpartion.cfg
lsblk
NAME   MAJ:MIN RM  SIZE RO TYPE MOUNTPOINT
sr0 11:01 1024M  0 rom
vda253:00   60G  0 disk
├─vda1 253:10  512M  0 part /boot
├─vda2 253:20  7.5G  0 part /
└─vda3 253:302G  0 part [SWAP]
The same happens with the userdata specified, because I don't use an API, I 
need to set custom value, such as a hostname when I boot a VM I created the 
following userdata in the UI as per doc:


#cloud-config runcmd: - echo 'hostname {{ ds.meta_data.hostname1 }}' > 
/tmp/hostname - echo '{{ ds.meta_data.hostname1 }} > /etc/hostname' > 
/tmp/hostname - echo 'hostnamectl set-hostname {{ ds.meta_data.hostname1 }}' > 
/tmp/hostname
And I declared the hostname1 variable, which allows me to specify the hostname 
under "Advanced" settings when I create a new instance. But nothing, for some 
reason, it keeps setting the VM name as the server hostname. But periods (.) 
are not allowed in the Name field for me to continue with this method and I 
need to be able to have special characters in the hostname for FQDNS spmetimes.

The only cloud-init command that ran was the 80_user.cfg that creates a managed 
user, that was the only one that worked. My upgrade and install configs none 
ran:
cat 52_upgrade_packages.cfg
#cloud-config

# Upgrade the instance on first boot
# (ie run apt-get upgrade)
#
# Default: false
# Aliases: apt_upgrade
cat  53_install_packages.cfg
#cloud-config

# Install additional packages on first boot
#
# Default: none
#
# if packages are specified, this apt_update will be set to true
#
# packages may be supplied as a single package name or as a list
# with the format [, ] wherein the specifc
# package version will be installed.
packages:
 - epel-release
 - neofetch
 - htop
And here's my cloud.cfg:
cat cloud.cfg
datasource_list: ["CloudStack"]

datasource:
  CloudStack:
max_wait: 120
timeout: 50

network:
  config: disabled

users:
  - default

system_info:
  default_user:
name: root

disable_root: false

ssh_deletekeys: true

cloud_init_modules:
 - seed_random
 - bootcmd
 - write-files
 - growpart
 - resizefs
 - disk_setup
 - mounts
 - set_hostname
 - update_hostname
 - update_etc_hosts
 - [ssh, always]
 - users-groups

cloud_config_modules:
 - ssh-import-id
 - locale
 - [ set-passwords, always ]
 - ntp
 - timezone
 - disable-ec2-metadata
 - [runcmd, always]

cloud_final_modules:
 - scripts-v

Windows OS templates in ACS

[Alexandru Stan]

Hi everyone,

Can someone please share a definitive method to build valid kvm templates from 
Windows Server/Desktop? I am following the guide here, (except I'm not 
installing the os in virtualbox, I'm doing it in ACS from iso then I create 
template from volume): How to create Windows 10 or Windows 11 Virtual Machines 
in OpenStack (using KVM) - I Learned How To... 
(wordpress.com)
 , but while everything works fine when creating an instance from the generated 
template the password reset NEVER works. I get the change password prompt at 
first boot but if I want to change the Administrator password from ACS at any 
time later, it  just doesn't apply it, the old password is the only one that 
works. With Linux distros and cloudinit is much easier, and the documentation 
actually works for a change, but for Windows it seems I'm out of luck. The 
guide I'm using is the most recent I could find on this topic, other are way 
older and are pointing to obsolete tools (like this one How to create the 
Windows template in Apache Cloudstack? 
(assistanz.com)).

Thank you!

RE: [D] Multiple managment servers [cloudstack]

[Alexandru Stan]

Hi,

When deploying the database use the virtual ip as target, not the individual ip 
of each db server. On one of the management server you need to use the 
--deply-as-root parameter, on the other 2 you don't use it. 

Alex


-Original Message-
From: SviridoffA (via GitHub)  
Sent: Friday, July 26, 2024 2:43 PM
To: users@cloudstack.apache.org
Subject: [D] Multiple managment servers [cloudstack]


GitHub user SviridoffA created a discussion: Multiple managment servers

I would be very grateful if someone could clarify the following question for 
me. I have a MySQL cluster with a virtual IP (pacemaker / corosync), and three 
management servers. I configured the first server via 
cloudstack-setup-databases then cloudstack-setup-management, according to the 
manual. Then I repeated the procedure on the another two management servers. 
When executing cloudstack-setup-databases, I specified the -i key and specified 
my IP on each server. After completing the configuration, everything seems to 
work, the changes are displayed on all management nodes. But if you go to 
Infrastracture - Management servers, only the last configured server is 
displayed. I would be very grateful if someone could suggest the correct option 
for configuring a scheme with several management servers.

GitHub link: https://github.com/apache/cloudstack/discussions/9457


This is an automatically sent email for users@cloudstack.apache.org.
To unsubscribe, please send an email to: users-unsubscr...@cloudstack.apache.org

Network offerings bug

[Alexandru Stan]

Hi guys. Anyone here affected by this issue? Creating Guest Network: unable to 
choose Network Offering. This for all L2,Shared and Isolated networks * Issue 
#9193 * apache/cloudstack 
(github.com)
I see that the issue is closed but there is no solution, or at least not that I 
could find there. The network offerings dropdown is populated before adding the 
zone, so something happens after the zone is up & running. I'm on 4.19.1.0 if 
it makes any difference.

Thanks!

RE: No network if VR and VM are on different hosts

[Alexandru Stan]

Yes, it wasn't ACS related, it was the switch. I split the management traffic 
and the public/guest traffic into 2 different bridges, trunk mode, all vlans 
are now transported as they should.

Thanks for the help guys!


-Original Message-
From: Alex Mattioli  
Sent: Wednesday, July 24, 2024 4:20 PM
To: users@cloudstack.apache.org
Subject: RE: No network if VR and VM are on different hosts

It doesn't sound like it's ACS related. 
With access ports it would definitely not work. Have you created all VLANs in 
the switch?

Cheers,
Alex

 


-Original Message-
From: Alexandru Stan 
Sent: Wednesday, July 24, 2024 2:47 PM
To: users@cloudstack.apache.org
Subject: RE: No network if VR and VM are on different hosts

Hi Wei,

They were. I even tried with plain, access ports, and only one interface on the 
host bridge, it's the same issue, the cloud0 bridge does not come up, it's 
either down or unknown. I really don't know where to go from here, never had 
this issue while testing and I did like 1000 installations on identical 
hardware, same host os. The only difference that I can think of is the ACS 
version, in testing I used 4.19.0 and now 4.19.1, but I don't think this is 
relevant since it doesn't seem like other users experienced this issue.

Alex

-Original Message-
From: Wei ZHOU 
Sent: Wednesday, July 24, 2024 1:27 AM
To: users@cloudstack.apache.org
Subject: Re: No network if VR and VM are on different hosts

Hi,

This seems to be a very typical issue for cloud platforms. The switch ports 
need to be configured as trunk mode.

-Wei

On Tuesday, July 23, 2024, Alexandru Stan 
 wrote:

> Hello. I am dealing with a weird networking issue that I never 
> experienced before and I really can't find anything relevant in logs 
> or else. The setup is like this:
>
>   *   Kvm hosts with 2 slave interfaces (for redundancy) under a master
> bridge called br0
>   *   The 2 interfaces are connected to 2 stacked Cisco switches
>   *   Traffic for management, guest, public, etc. goes through this bridge
> only
>   *   ACS zone added with advanced network, br0 label set on each traffic
> type
>
> Everything seems fine, the system vms are up & running, hosts are 
> added without issues, etc. The cloud0 bridge is created on each host 
> as it should. Now, if I add a vm on a certain host and the virtual 
> router (simple, isolated network) is on the same host, everything 
> works fine, the vm gets its own ip, dns, internet conectivity, etc.
> But if I add a vm to a different host than where the VR is hosted, I 
> don't get any network for the vm. The virtual interfaces are created 
> on the host but the cloud0 bridge status is either down or unknown.
> Could this be related to the hardware setup or it's something ACS related?
>
> Thank you!
>

RE: No network if VR and VM are on different hosts

[Alexandru Stan]

Hi Jimmy,

>From what I've seen, libvirtd adds a virbr0 bridge, cloud0 is added when the 
>host is configured in the ACS ui. And under this bridge all the virtual 
>interfaces are created, so it is in use. 
I have only one bridge per host, with (now) only one ethernet interface under 
it. That bridge is set as traffic label in ACS on all traffic types 
(management, public, etc.). That's all to it, nothing fancy.
You guys may be right about the switch ports, in fact this was my next step, to 
get it out of the equation and see if everything works as it should. 


-Original Message-
From: Jimmy Huybrechts  
Sent: Wednesday, July 24, 2024 4:07 PM
To: users@cloudstack.apache.org
Subject: Re: No network if VR and VM are on different hosts

Hi,

Cloud0 is added here too, in a down state (it's normal as I don't use them and 
it's auto added by libvirtd)

What are the bridges defined inside Cloudstack?

It very much sounds like an issue on the switch side as Wei mentioned as well.

--
Jimmy

From: Alexandru Stan 
Date: Wednesday, 24 July 2024 at 14:47
To: users@cloudstack.apache.org 
Subject: RE: No network if VR and VM are on different hosts Hi Wei,

They were. I even tried with plain, access ports, and only one interface on the 
host bridge, it's the same issue, the cloud0 bridge does not come up, it's 
either down or unknown. I really don't know where to go from here, never had 
this issue while testing and I did like 1000 installations on identical 
hardware, same host os. The only difference that I can think of is the ACS 
version, in testing I used 4.19.0 and now 4.19.1, but I don't think this is 
relevant since it doesn't seem like other users experienced this issue.

Alex

-Original Message-
From: Wei ZHOU 
Sent: Wednesday, July 24, 2024 1:27 AM
To: users@cloudstack.apache.org
Subject: Re: No network if VR and VM are on different hosts

Hi,

This seems to be a very typical issue for cloud platforms. The switch ports 
need to be configured as trunk mode.

-Wei

On Tuesday, July 23, 2024, Alexandru Stan 
 wrote:

> Hello. I am dealing with a weird networking issue that I never 
> experienced before and I really can't find anything relevant in logs 
> or else. The setup is like this:
>
>   *   Kvm hosts with 2 slave interfaces (for redundancy) under a master
> bridge called br0
>   *   The 2 interfaces are connected to 2 stacked Cisco switches
>   *   Traffic for management, guest, public, etc. goes through this bridge
> only
>   *   ACS zone added with advanced network, br0 label set on each traffic
> type
>
> Everything seems fine, the system vms are up & running, hosts are 
> added without issues, etc. The cloud0 bridge is created on each host 
> as it should. Now, if I add a vm on a certain host and the virtual 
> router (simple, isolated network) is on the same host, everything 
> works fine, the vm gets its own ip, dns, internet conectivity, etc.
> But if I add a vm to a different host than where the VR is hosted, I 
> don't get any network for the vm. The virtual interfaces are created 
> on the host but the cloud0 bridge status is either down or unknown.
> Could this be related to the hardware setup or it's something ACS related?
>
> Thank you!
>

RE: No network if VR and VM are on different hosts

[Alexandru Stan]

Hi Wei,

They were. I even tried with plain, access ports, and only one interface on the 
host bridge, it's the same issue, the cloud0 bridge does not come up, it's 
either down or unknown. I really don't know where to go from here, never had 
this issue while testing and I did like 1000 installations on identical 
hardware, same host os. The only difference that I can think of is the ACS 
version, in testing I used 4.19.0 and now 4.19.1, but I don't think this is 
relevant since it doesn't seem like other users experienced this issue.

Alex

-Original Message-
From: Wei ZHOU  
Sent: Wednesday, July 24, 2024 1:27 AM
To: users@cloudstack.apache.org
Subject: Re: No network if VR and VM are on different hosts

Hi,

This seems to be a very typical issue for cloud platforms. The switch ports 
need to be configured as trunk mode.

-Wei

On Tuesday, July 23, 2024, Alexandru Stan 
 wrote:

> Hello. I am dealing with a weird networking issue that I never 
> experienced before and I really can't find anything relevant in logs 
> or else. The setup is like this:
>
>   *   Kvm hosts with 2 slave interfaces (for redundancy) under a master
> bridge called br0
>   *   The 2 interfaces are connected to 2 stacked Cisco switches
>   *   Traffic for management, guest, public, etc. goes through this bridge
> only
>   *   ACS zone added with advanced network, br0 label set on each traffic
> type
>
> Everything seems fine, the system vms are up & running, hosts are 
> added without issues, etc. The cloud0 bridge is created on each host 
> as it should. Now, if I add a vm on a certain host and the virtual 
> router (simple, isolated network) is on the same host, everything 
> works fine, the vm gets its own ip, dns, internet conectivity, etc. 
> But if I add a vm to a different host than where the VR is hosted, I 
> don't get any network for the vm. The virtual interfaces are created 
> on the host but the cloud0 bridge status is either down or unknown.
> Could this be related to the hardware setup or it's something ACS related?
>
> Thank you!
>

No network if VR and VM are on different hosts

[Alexandru Stan]

Hello. I am dealing with a weird networking issue that I never experienced 
before and I really can't find anything relevant in logs or else. The setup is 
like this:

  *   Kvm hosts with 2 slave interfaces (for redundancy) under a master bridge 
called br0
  *   The 2 interfaces are connected to 2 stacked Cisco switches
  *   Traffic for management, guest, public, etc. goes through this bridge only
  *   ACS zone added with advanced network, br0 label set on each traffic type

Everything seems fine, the system vms are up & running, hosts are added without 
issues, etc. The cloud0 bridge is created on each host as it should. Now, if I 
add a vm on a certain host and the virtual router (simple, isolated network) is 
on the same host, everything works fine, the vm gets its own ip, dns, internet 
conectivity, etc. But if I add a vm to a different host than where the VR is 
hosted, I don't get any network for the vm. The virtual interfaces are created 
on the host but the cloud0 bridge status is either down or unknown.
Could this be related to the hardware setup or it's something ACS related?

Thank you!

RE: Networking issue with VLAN setup

[Alexandru Stan]

Hi Wei,

So instead of the vlan interface as bridge slave, I should have the bond?



-Original Message-
From: Wei ZHOU  
Sent: Wednesday, July 17, 2024 11:25 AM
To: users@cloudstack.apache.org
Subject: Re: Networking issue with VLAN setup

It looks like there is no physical interface (ethX/bondX) added to the bridge 
for guest networks (cloudbrX).

-Wei



On Wed, Jul 17, 2024 at 10:07 AM Alexandru Stan 
 wrote:
>
> Hi everyone. I have a specific network setup that doesn't seem to be working 
> with CS for some reason, I think it's a missconfiguration on my part. So, 
> there are 2 interfaces on the host, one it's a bond interface and one it's a 
> vlan interface that's a member of the bond. Since I need a bridge for CS, I 
> added one, with the ipv4 settings and everything, and made the vlan interface 
> the slave of the bridge. When I tried to add the zone (advanced) I set the 
> bridge label on the management interface and added the ip range with untagged 
> traffic, but it couldn't add the hosts correctly, they all had alert status. 
> Removing the label from the management interface seems to work somehow, the 
> hosts are up and running, the system vms are up and connected, etc. I added 
> an isolated network and tried to create a vm, but this is where the main 
> issue starts, I cannot deploy any vm, they all fail with the same error and 
> it has something to do with networking. The error on the host is this 
> (obviously the device name varies):
>
> Process [206176] for command 
> [/usr/share/cloudstack-common/scripts/vm/network/vnet/modifyvlan.sh -v 1277 
> -p  -b br-1277 -o add ] encountered the error: [Cannot find device 
> "br-1277"/usr/share/cloudstack-common/scripts/vm/network/vnet/modifyvlan.sh: 
> line 37: /proc/sys/net/ipv6/conf/br-1277.1277/disable_ipv6: No such file or 
> directoryCannot find device "br-1277.1277"Failed to create vlan 1277 on pif: 
> br-1277.].
>
> Does anyone have a clue about what could be wrong here?
>
> Thank you!

Networking issue with VLAN setup

[Alexandru Stan]

Hi everyone. I have a specific network setup that doesn't seem to be working 
with CS for some reason, I think it's a missconfiguration on my part. So, there 
are 2 interfaces on the host, one it's a bond interface and one it's a vlan 
interface that's a member of the bond. Since I need a bridge for CS, I added 
one, with the ipv4 settings and everything, and made the vlan interface the 
slave of the bridge. When I tried to add the zone (advanced) I set the bridge 
label on the management interface and added the ip range with untagged traffic, 
but it couldn't add the hosts correctly, they all had alert status. Removing 
the label from the management interface seems to work somehow, the hosts are up 
and running, the system vms are up and connected, etc. I added an isolated 
network and tried to create a vm, but this is where the main issue starts, I 
cannot deploy any vm, they all fail with the same error and it has something to 
do with networking. The error on the host is this (obviously the device name 
varies):

Process [206176] for command 
[/usr/share/cloudstack-common/scripts/vm/network/vnet/modifyvlan.sh -v 1277 -p  
-b br-1277 -o add ] encountered the error: [Cannot find device 
"br-1277"/usr/share/cloudstack-common/scripts/vm/network/vnet/modifyvlan.sh: 
line 37: /proc/sys/net/ipv6/conf/br-1277.1277/disable_ipv6: No such file or 
directoryCannot find device "br-1277.1277"Failed to create vlan 1277 on pif: 
br-1277.].

Does anyone have a clue about what could be wrong here?

Thank you!

RE: HA for management server

[Alexandru Stan]

I did not change anything on that file until today when I modified that driver 
and all was fine after - so following that documentation for multiple 
management servers deployment results in fail if we leave it with the defaults. 
So I don't have an issue anymore, the HA is handled by galera for the DB and by 
a load balancer for the management server instance. It works fine.


-Original Message-
From: Wei ZHOU  
Sent: Thursday, May 30, 2024 1:12 PM
To: users@cloudstack.apache.org
Subject: Re: HA for management server

I thought you did not change the db.properties.

if you use mariadb driver, you can use "db.cloud.uri".
the failover mode is not supported in db.properties, which could be the root 
cause of your issue 
https://mariadb.com/kb/en/failover-and-high-availability-with-mariadb-connector-j/


-Wei


On Thu, May 30, 2024 at 11:48 AM Alexandru Stan 
 wrote:
>
> Hello,
>
> I actually managed to get it to work just by modifying the db.cloud.driver in 
> db.properties on the management servers to jdbc:mariadb instead of the 
> default jdbc:mysql. Maybe I didn't search properly but I really think this is 
> not mentioned in any documentation of multi-management servers deployment. 
> Anyway, all good now.
>
> Thank you!
>
>
>
> -Original Message-
> From: Wei ZHOU 
> Sent: Wednesday, May 29, 2024 11:29 PM
> To: users@cloudstack.apache.org
> Subject: Re: HA for management server
>
> Looks good
> Can you try db.cloud.uri in db.properties?
> See https://github.com/apache/cloudstack/pull/7895
>
> -Wei
>
>
> On Wednesday, May 29, 2024, Alexandru Stan 
>  wrote:
>
> > Yes. On the first management server with --deploy-as=root:password 
> > and on the second one without. All successful. The I ran 
> > cloudstack-setup-management and that was it. But in the browser I 
> > get the
> > 503 error...
> >
> >
> > -Original Message-
> > From: Wei ZHOU 
> > Sent: Wednesday, May 29, 2024 11:20 PM
> > To: users@cloudstack.apache.org
> > Subject: Re: HA for management server
> >
> > Have you setup the database by cloudstack-setup-database?
> >
> >
> >
> > On Wednesday, May 29, 2024, Alexandru Stan 
> > 
> > wrote:
> >
> > > Yes, I can login with the cloud user just fine via the virtual ip.
> > >
> > >
> > > -Original Message-
> > > From: Wei ZHOU 
> > > Sent: Wednesday, May 29, 2024 10:45 PM
> > > To: users@cloudstack.apache.org
> > > Subject: Re: HA for management server
> > >
> > > Yes.
> > > Are you able to access the db vip or the two db nodes using mysql 
> > > command from management server?
> > >
> > >
> > >
> > > On Wednesday, May 29, 2024, Alexandru Stan 
> > > 
> > > wrote:
> > >
> > > > You're right. This is the actual problem:
> > > >
> > > > Caused by: com.cloud.utils.exception.CloudRuntimeException: 
> > > > Unable to initialize a connection to the database for locking 
> > > > purposes Caused
> > > > by: java.sql.SQLNonTransientConnectionException: Could not 
> > > > create connection to database server. Attempted reconnect 3 times. 
> > > > Giving up.
> > > >
> > > > But everything seems right in the db.properties file. The host 
> > > > should be the virtual ip of the load balancer, right?
> > > >
> > > >
> > > > -Original Message-
> > > > From: Wei ZHOU 
> > > > Sent: Wednesday, May 29, 2024 10:28 PM
> > > > To: users@cloudstack.apache.org
> > > > Subject: Re: HA for management server
> > > >
> > > > Hi,
> > > >
> > > > The exception is misleading
> > > >
> > > > You can check the real exception by grep Exception 
> > > > management-server.log | grep -v NoSuchBeanDefinitionException
> > > >
> > > > I guess the  root cause is database configuration or db.properties.
> > > >
> > > > By the way, it is better mariadb galera cluster have 3 servers.
> > > >
> > > >
> > > > -Wei
> > > >
> > > >
> > > > On Wednesday, May 29, 2024, Alexandru Stan 
> > > > 
> > > > wrote:
> > > >
> > > > > Hello,
> > > > >
> > > > > There seems to be issues with installing the management server 
> > > > > databases on a separate server, I setup a galera cluster with 
> > >

RE: HA for management server

[Alexandru Stan]

Hello,

I actually managed to get it to work just by modifying the db.cloud.driver in 
db.properties on the management servers to jdbc:mariadb instead of the default 
jdbc:mysql. Maybe I didn't search properly but I really think this is not 
mentioned in any documentation of multi-management servers deployment. Anyway, 
all good now.

Thank you!



-Original Message-
From: Wei ZHOU  
Sent: Wednesday, May 29, 2024 11:29 PM
To: users@cloudstack.apache.org
Subject: Re: HA for management server

Looks good
Can you try db.cloud.uri in db.properties?
See https://github.com/apache/cloudstack/pull/7895

-Wei


On Wednesday, May 29, 2024, Alexandru Stan 
 wrote:

> Yes. On the first management server with --deploy-as=root:password and 
> on the second one without. All successful. The I ran 
> cloudstack-setup-management and that was it. But in the browser I get 
> the
> 503 error...
>
>
> -Original Message-
> From: Wei ZHOU 
> Sent: Wednesday, May 29, 2024 11:20 PM
> To: users@cloudstack.apache.org
> Subject: Re: HA for management server
>
> Have you setup the database by cloudstack-setup-database?
>
>
>
> On Wednesday, May 29, 2024, Alexandru Stan 
> 
> wrote:
>
> > Yes, I can login with the cloud user just fine via the virtual ip.
> >
> >
> > -Original Message-
> > From: Wei ZHOU 
> > Sent: Wednesday, May 29, 2024 10:45 PM
> > To: users@cloudstack.apache.org
> > Subject: Re: HA for management server
> >
> > Yes.
> > Are you able to access the db vip or the two db nodes using mysql 
> > command from management server?
> >
> >
> >
> > On Wednesday, May 29, 2024, Alexandru Stan 
> > 
> > wrote:
> >
> > > You're right. This is the actual problem:
> > >
> > > Caused by: com.cloud.utils.exception.CloudRuntimeException: Unable 
> > > to initialize a connection to the database for locking purposes 
> > > Caused
> > > by: java.sql.SQLNonTransientConnectionException: Could not create 
> > > connection to database server. Attempted reconnect 3 times. Giving up.
> > >
> > > But everything seems right in the db.properties file. The host 
> > > should be the virtual ip of the load balancer, right?
> > >
> > >
> > > -Original Message-
> > > From: Wei ZHOU 
> > > Sent: Wednesday, May 29, 2024 10:28 PM
> > > To: users@cloudstack.apache.org
> > > Subject: Re: HA for management server
> > >
> > > Hi,
> > >
> > > The exception is misleading
> > >
> > > You can check the real exception by grep Exception 
> > > management-server.log | grep -v NoSuchBeanDefinitionException
> > >
> > > I guess the  root cause is database configuration or db.properties.
> > >
> > > By the way, it is better mariadb galera cluster have 3 servers.
> > >
> > >
> > > -Wei
> > >
> > >
> > > On Wednesday, May 29, 2024, Alexandru Stan 
> > > 
> > > wrote:
> > >
> > > > Hello,
> > > >
> > > > There seems to be issues with installing the management server 
> > > > databases on a separate server, I setup a galera cluster with 2 
> > > > servers and tried to install the management server on a 
> > > > different server using (where it still
> > > > applies) this guide: https://www.shapeblue.com/how- 
> > > > to-deploy-high-availability-cloudstack-with-mariadb-galera-clust
> > > > er
> > > > /
> > > >
> > > > However, while the database setup seems to be fine and 
> > > > everything after, the cloudstack-management service is running 
> > > > but there's no
> UI.
> > > > Checking the status shows this:
> > > >
> > > > Failed to start module [redfish] due to: [Error creating bean 
> > > > with name 'org.springframework.beans.factory.config.
> > > PropertyPlaceholderConfigurer#0'
> > > > defined in URL [jar:file:/usr/share/cloudstack-management/lib/
> > > > cloudstack-4.19.0.1.jar!/META-INF/cloudstack/bootstrap/
> > > > spring-bootstrap-context-inheritable.xml]: Cannot resolve 
> > > > reference to bean 'DefaultConfigResources' while setting bean 
> > > > property 'locations'; nested exception is org.springframework.beans.
> factory.
> > > NoSuchBeanDefinitionException:
> > > > No bean named 'DefaultConfigResources' available].
> > > >
> > > > Any 

RE: HA for management server

[Alexandru Stan]

I would but I don't really understand what I need to add to the file. This 
parameters maybe, adjusted with my ips?

db.cloud.driver=jdbc:mariadb
db.cloud.uri=jdbc:mariadb:sequential://192.168.201.160:3306,192.168.201.161:3306,192.168.201.162:3306/cloud?autoReconnect=true&prepStmtCacheSize=517&cachePrepStmts=true&sessionVariables=sql_mode='STRICT_TRANS_TABLES,NO_ZERO_IN_DATE,NO_ZERO_DATE,ERROR_FOR_DIVISION_BY_ZERO,NO_ENGINE_SUBSTITUTION'&serverTimezone=UTC



-Original Message-
From: Wei ZHOU  
Sent: Wednesday, May 29, 2024 11:29 PM
To: users@cloudstack.apache.org
Subject: Re: HA for management server

Looks good
Can you try db.cloud.uri in db.properties?
See https://github.com/apache/cloudstack/pull/7895

-Wei


On Wednesday, May 29, 2024, Alexandru Stan 
 wrote:

> Yes. On the first management server with --deploy-as=root:password and 
> on the second one without. All successful. The I ran 
> cloudstack-setup-management and that was it. But in the browser I get 
> the
> 503 error...
>
>
> -Original Message-
> From: Wei ZHOU 
> Sent: Wednesday, May 29, 2024 11:20 PM
> To: users@cloudstack.apache.org
> Subject: Re: HA for management server
>
> Have you setup the database by cloudstack-setup-database?
>
>
>
> On Wednesday, May 29, 2024, Alexandru Stan 
> 
> wrote:
>
> > Yes, I can login with the cloud user just fine via the virtual ip.
> >
> >
> > -Original Message-
> > From: Wei ZHOU 
> > Sent: Wednesday, May 29, 2024 10:45 PM
> > To: users@cloudstack.apache.org
> > Subject: Re: HA for management server
> >
> > Yes.
> > Are you able to access the db vip or the two db nodes using mysql 
> > command from management server?
> >
> >
> >
> > On Wednesday, May 29, 2024, Alexandru Stan 
> > 
> > wrote:
> >
> > > You're right. This is the actual problem:
> > >
> > > Caused by: com.cloud.utils.exception.CloudRuntimeException: Unable 
> > > to initialize a connection to the database for locking purposes 
> > > Caused
> > > by: java.sql.SQLNonTransientConnectionException: Could not create 
> > > connection to database server. Attempted reconnect 3 times. Giving up.
> > >
> > > But everything seems right in the db.properties file. The host 
> > > should be the virtual ip of the load balancer, right?
> > >
> > >
> > > -Original Message-
> > > From: Wei ZHOU 
> > > Sent: Wednesday, May 29, 2024 10:28 PM
> > > To: users@cloudstack.apache.org
> > > Subject: Re: HA for management server
> > >
> > > Hi,
> > >
> > > The exception is misleading
> > >
> > > You can check the real exception by grep Exception 
> > > management-server.log | grep -v NoSuchBeanDefinitionException
> > >
> > > I guess the  root cause is database configuration or db.properties.
> > >
> > > By the way, it is better mariadb galera cluster have 3 servers.
> > >
> > >
> > > -Wei
> > >
> > >
> > > On Wednesday, May 29, 2024, Alexandru Stan 
> > > 
> > > wrote:
> > >
> > > > Hello,
> > > >
> > > > There seems to be issues with installing the management server 
> > > > databases on a separate server, I setup a galera cluster with 2 
> > > > servers and tried to install the management server on a 
> > > > different server using (where it still
> > > > applies) this guide: https://www.shapeblue.com/how- 
> > > > to-deploy-high-availability-cloudstack-with-mariadb-galera-clust
> > > > er
> > > > /
> > > >
> > > > However, while the database setup seems to be fine and 
> > > > everything after, the cloudstack-management service is running 
> > > > but there's no
> UI.
> > > > Checking the status shows this:
> > > >
> > > > Failed to start module [redfish] due to: [Error creating bean 
> > > > with name 'org.springframework.beans.factory.config.
> > > PropertyPlaceholderConfigurer#0'
> > > > defined in URL [jar:file:/usr/share/cloudstack-management/lib/
> > > > cloudstack-4.19.0.1.jar!/META-INF/cloudstack/bootstrap/
> > > > spring-bootstrap-context-inheritable.xml]: Cannot resolve 
> > > > reference to bean 'DefaultConfigResources' while setting bean 
> > > > property 'locations'; nested exception is org.springframework.beans.
> factory.
> > > NoSuchBeanDefinitionEx

RE: HA for management server

[Alexandru Stan]

Yes. On the first management server with --deploy-as=root:password and on the 
second one without. All successful. The I ran cloudstack-setup-management and 
that was it. But in the browser I get the 503 error...


-Original Message-
From: Wei ZHOU  
Sent: Wednesday, May 29, 2024 11:20 PM
To: users@cloudstack.apache.org
Subject: Re: HA for management server

Have you setup the database by cloudstack-setup-database?



On Wednesday, May 29, 2024, Alexandru Stan 
 wrote:

> Yes, I can login with the cloud user just fine via the virtual ip.
>
>
> -Original Message-
> From: Wei ZHOU 
> Sent: Wednesday, May 29, 2024 10:45 PM
> To: users@cloudstack.apache.org
> Subject: Re: HA for management server
>
> Yes.
> Are you able to access the db vip or the two db nodes using mysql 
> command from management server?
>
>
>
> On Wednesday, May 29, 2024, Alexandru Stan 
> 
> wrote:
>
> > You're right. This is the actual problem:
> >
> > Caused by: com.cloud.utils.exception.CloudRuntimeException: Unable 
> > to initialize a connection to the database for locking purposes 
> > Caused
> > by: java.sql.SQLNonTransientConnectionException: Could not create 
> > connection to database server. Attempted reconnect 3 times. Giving up.
> >
> > But everything seems right in the db.properties file. The host 
> > should be the virtual ip of the load balancer, right?
> >
> >
> > -Original Message-
> > From: Wei ZHOU 
> > Sent: Wednesday, May 29, 2024 10:28 PM
> > To: users@cloudstack.apache.org
> > Subject: Re: HA for management server
> >
> > Hi,
> >
> > The exception is misleading
> >
> > You can check the real exception by
> > grep Exception management-server.log | grep -v 
> > NoSuchBeanDefinitionException
> >
> > I guess the  root cause is database configuration or db.properties.
> >
> > By the way, it is better mariadb galera cluster have 3 servers.
> >
> >
> > -Wei
> >
> >
> > On Wednesday, May 29, 2024, Alexandru Stan 
> > 
> > wrote:
> >
> > > Hello,
> > >
> > > There seems to be issues with installing the management server 
> > > databases on a separate server, I setup a galera cluster with 2 
> > > servers and tried to install the management server on a different 
> > > server using (where it still
> > > applies) this guide: https://www.shapeblue.com/how- 
> > > to-deploy-high-availability-cloudstack-with-mariadb-galera-cluster
> > > /
> > >
> > > However, while the database setup seems to be fine and everything 
> > > after, the cloudstack-management service is running but there's no UI.
> > > Checking the status shows this:
> > >
> > > Failed to start module [redfish] due to: [Error creating bean with 
> > > name 'org.springframework.beans.factory.config.
> > PropertyPlaceholderConfigurer#0'
> > > defined in URL [jar:file:/usr/share/cloudstack-management/lib/
> > > cloudstack-4.19.0.1.jar!/META-INF/cloudstack/bootstrap/
> > > spring-bootstrap-context-inheritable.xml]: Cannot resolve 
> > > reference to bean 'DefaultConfigResources' while setting bean 
> > > property 'locations'; nested exception is 
> > > org.springframework.beans.factory.
> > NoSuchBeanDefinitionException:
> > > No bean named 'DefaultConfigResources' available].
> > >
> > > Any idea why this happens?
> > >
> > >
> > > -Original Message-
> > > From: Jayanth Babu A 
> > > Sent: Monday, May 27, 2024 12:28 PM
> > > To: users@cloudstack.apache.org
> > > Subject: Re: HA for management server
> > >
> > > Hello,
> > > Yes, DB clustering with Galera or some other sort should be looked
> into.
> > > I'm not sure if [1] has worked for someone so far.
> > >
> > > [1] https://docs.cloudstack.apache.org/en/latest/
> > > adminguide/reliability.html#configuring-database-high-availability
> > >
> > > Thanks,
> > > Jayanth
> > > 
> > > From: Alexandru Stan 
> > > Sent: Monday, May 27, 2024 2:48 PM
> > > To: users@cloudstack.apache.org 
> > > Subject: RE: HA for management server
> > >
> > > Hi,
> > >
> > > Actually no, I'm just documenting the process so far, so nothing 
> > > was done for the DB. For this I assume the info in the official 
> > > documentation would be ok? Or somethi

RE: HA for management server

[Alexandru Stan]

Yes, I can login with the cloud user just fine via the virtual ip.


-Original Message-
From: Wei ZHOU  
Sent: Wednesday, May 29, 2024 10:45 PM
To: users@cloudstack.apache.org
Subject: Re: HA for management server

Yes.
Are you able to access the db vip or the two db nodes using mysql command from 
management server?



On Wednesday, May 29, 2024, Alexandru Stan 
 wrote:

> You're right. This is the actual problem:
>
> Caused by: com.cloud.utils.exception.CloudRuntimeException: Unable to 
> initialize a connection to the database for locking purposes Caused 
> by: java.sql.SQLNonTransientConnectionException: Could not create 
> connection to database server. Attempted reconnect 3 times. Giving up.
>
> But everything seems right in the db.properties file. The host should 
> be the virtual ip of the load balancer, right?
>
>
> -Original Message-
> From: Wei ZHOU 
> Sent: Wednesday, May 29, 2024 10:28 PM
> To: users@cloudstack.apache.org
> Subject: Re: HA for management server
>
> Hi,
>
> The exception is misleading
>
> You can check the real exception by
> grep Exception management-server.log | grep -v 
> NoSuchBeanDefinitionException
>
> I guess the  root cause is database configuration or db.properties.
>
> By the way, it is better mariadb galera cluster have 3 servers.
>
>
> -Wei
>
>
> On Wednesday, May 29, 2024, Alexandru Stan 
> 
> wrote:
>
> > Hello,
> >
> > There seems to be issues with installing the management server 
> > databases on a separate server, I setup a galera cluster with 2 
> > servers and tried to install the management server on a different 
> > server using (where it still
> > applies) this guide: https://www.shapeblue.com/how- 
> > to-deploy-high-availability-cloudstack-with-mariadb-galera-cluster/
> >
> > However, while the database setup seems to be fine and everything 
> > after, the cloudstack-management service is running but there's no UI.
> > Checking the status shows this:
> >
> > Failed to start module [redfish] due to: [Error creating bean with 
> > name 'org.springframework.beans.factory.config.
> PropertyPlaceholderConfigurer#0'
> > defined in URL [jar:file:/usr/share/cloudstack-management/lib/
> > cloudstack-4.19.0.1.jar!/META-INF/cloudstack/bootstrap/
> > spring-bootstrap-context-inheritable.xml]: Cannot resolve reference 
> > to bean 'DefaultConfigResources' while setting bean property 
> > 'locations'; nested exception is org.springframework.beans.factory.
> NoSuchBeanDefinitionException:
> > No bean named 'DefaultConfigResources' available].
> >
> > Any idea why this happens?
> >
> >
> > -Original Message-
> > From: Jayanth Babu A 
> > Sent: Monday, May 27, 2024 12:28 PM
> > To: users@cloudstack.apache.org
> > Subject: Re: HA for management server
> >
> > Hello,
> > Yes, DB clustering with Galera or some other sort should be looked into.
> > I'm not sure if [1] has worked for someone so far.
> >
> > [1] https://docs.cloudstack.apache.org/en/latest/
> > adminguide/reliability.html#configuring-database-high-availability
> >
> > Thanks,
> > Jayanth
> > 
> > From: Alexandru Stan 
> > Sent: Monday, May 27, 2024 2:48 PM
> > To: users@cloudstack.apache.org 
> > Subject: RE: HA for management server
> >
> > Hi,
> >
> > Actually no, I'm just documenting the process so far, so nothing was 
> > done for the DB. For this I assume the info in the official 
> > documentation would be ok? Or something more advanced would be 
> > needed,
> like a galera cluster?
> >
> >
> >
> > -Original Message-
> > From: Jayanth Babu A 
> > Sent: Monday, May 27, 2024 12:12 PM
> > To: users@cloudstack.apache.org
> > Subject: Re: HA for management server
> >
> > Hello Alexandru,
> > Please see [1] and [2]. I assume your DB HA is taken care of.
> >
> > [1] https://checkpoint.url-protection.com/v1/url?o=https%
> > 3A//docs.cloudstack.apache.org/en/latest/adminguide/reliability.html
> > &g = ZmEwYjAwNmYwMjIxNTU5ZQ==&h=NmQyMzY5Nzc5YTU2ZDkyYjc4N2ExND
> > U4Y2Q0OWFkMjQ4OGRhMGZiZjdiNDA5N2NmNmRhM2EwYjUzZGEwMmI2Mg==&p=
> > Y3AxZTpueHRnZW5pbmZpbml0ZWRhdGFjZW50ZXI6YzpvOmY0ZWFkYmU5NDkz
> > MjMyMjdjY2EyM2IxN2YzOGZiMWE4OnYxOnA6VA==
> > [2] https://checkpoint.url-protection.com/v1/url?o=https%
> > 3A//github.com/apache/cloudstack/issues/8221&g=YzVlNDk3YjI5NTEwMTM5M
> > g=
> > =&h=
> > OTBmN2Q4ZjA3YzRlOTA

RE: HA for management server

[Alexandru Stan]

You're right. This is the actual problem:

Caused by: com.cloud.utils.exception.CloudRuntimeException: Unable to 
initialize a connection to the database for locking purposes
Caused by: java.sql.SQLNonTransientConnectionException: Could not create 
connection to database server. Attempted reconnect 3 times. Giving up.

But everything seems right in the db.properties file. The host should be the 
virtual ip of the load balancer, right?


-Original Message-
From: Wei ZHOU  
Sent: Wednesday, May 29, 2024 10:28 PM
To: users@cloudstack.apache.org
Subject: Re: HA for management server

Hi,

The exception is misleading

You can check the real exception by
grep Exception management-server.log | grep -v NoSuchBeanDefinitionException

I guess the  root cause is database configuration or db.properties.

By the way, it is better mariadb galera cluster have 3 servers.


-Wei


On Wednesday, May 29, 2024, Alexandru Stan 
 wrote:

> Hello,
>
> There seems to be issues with installing the management server 
> databases on a separate server, I setup a galera cluster with 2 
> servers and tried to install the management server on a different 
> server using (where it still
> applies) this guide: https://www.shapeblue.com/how- 
> to-deploy-high-availability-cloudstack-with-mariadb-galera-cluster/
>
> However, while the database setup seems to be fine and everything 
> after, the cloudstack-management service is running but there's no UI. 
> Checking the status shows this:
>
> Failed to start module [redfish] due to: [Error creating bean with 
> name 
> 'org.springframework.beans.factory.config.PropertyPlaceholderConfigurer#0'
> defined in URL [jar:file:/usr/share/cloudstack-management/lib/
> cloudstack-4.19.0.1.jar!/META-INF/cloudstack/bootstrap/
> spring-bootstrap-context-inheritable.xml]: Cannot resolve reference to 
> bean 'DefaultConfigResources' while setting bean property 'locations'; 
> nested exception is 
> org.springframework.beans.factory.NoSuchBeanDefinitionException:
> No bean named 'DefaultConfigResources' available].
>
> Any idea why this happens?
>
>
> -Original Message-
> From: Jayanth Babu A 
> Sent: Monday, May 27, 2024 12:28 PM
> To: users@cloudstack.apache.org
> Subject: Re: HA for management server
>
> Hello,
> Yes, DB clustering with Galera or some other sort should be looked into.
> I'm not sure if [1] has worked for someone so far.
>
> [1] https://docs.cloudstack.apache.org/en/latest/
> adminguide/reliability.html#configuring-database-high-availability
>
> Thanks,
> Jayanth
> 
> From: Alexandru Stan 
> Sent: Monday, May 27, 2024 2:48 PM
> To: users@cloudstack.apache.org 
> Subject: RE: HA for management server
>
> Hi,
>
> Actually no, I'm just documenting the process so far, so nothing was 
> done for the DB. For this I assume the info in the official 
> documentation would be ok? Or something more advanced would be needed, like a 
> galera cluster?
>
>
>
> -Original Message-
> From: Jayanth Babu A 
> Sent: Monday, May 27, 2024 12:12 PM
> To: users@cloudstack.apache.org
> Subject: Re: HA for management server
>
> Hello Alexandru,
> Please see [1] and [2]. I assume your DB HA is taken care of.
>
> [1] https://checkpoint.url-protection.com/v1/url?o=https%
> 3A//docs.cloudstack.apache.org/en/latest/adminguide/reliability.html&g
> = ZmEwYjAwNmYwMjIxNTU5ZQ==&h=NmQyMzY5Nzc5YTU2ZDkyYjc4N2ExND
> U4Y2Q0OWFkMjQ4OGRhMGZiZjdiNDA5N2NmNmRhM2EwYjUzZGEwMmI2Mg==&p=
> Y3AxZTpueHRnZW5pbmZpbml0ZWRhdGFjZW50ZXI6YzpvOmY0ZWFkYmU5NDkz
> MjMyMjdjY2EyM2IxN2YzOGZiMWE4OnYxOnA6VA==
> [2] https://checkpoint.url-protection.com/v1/url?o=https%
> 3A//github.com/apache/cloudstack/issues/8221&g=YzVlNDk3YjI5NTEwMTM5Mg=
> =&h=
> OTBmN2Q4ZjA3YzRlOTAxZDk1YzcwMmExMTAzYTY4N2VmYjg5MGZkZGQyMDE2
> NTM3MTU4MGUxNDRjNzMzN2E0ZQ==&p=Y3AxZTpueHRnZW5pbmZpbml0ZWRhdG
> FjZW50ZXI6YzpvOmY0ZWFkYmU5NDkzMjMyMjdjY2EyM2IxN2YzOGZiMWE4OnYxOnA6VA==
>
> Regards,
> Jayanth Reddy
> Engineering Manager
>
> From: Alexandru Stan 
> Date: Monday, 27 May 2024 at 2:38 PM
> To: users@cloudstack.apache.org 
> Subject: HA for management server
> Hello,
>
> Is there any documentation on deployment of multiple management 
> servers in high availability? Or does anyone have good info on this? 
> The official documentation does not contain anything relevant unfortunately.
>
> Thank you!
> Disclaimer *** This e-mail contains PRIVILEGED AND CONFIDENTIAL 
> INFORMATION intended solely for the use of the addressee(s). If you 
> are not the intended recipient, please notify the sender by e-mail and 
> delete the original message. Further, you are not author

RE: HA for management server

[Alexandru Stan]

Hello,

There seems to be issues with installing the management server databases on a 
separate server, I setup a galera cluster with 2 servers and tried to install 
the management server on a different server using (where it still applies) this 
guide: 
https://www.shapeblue.com/how-to-deploy-high-availability-cloudstack-with-mariadb-galera-cluster/

However, while the database setup seems to be fine and everything after, the 
cloudstack-management service is running but there's no UI. Checking the status 
shows this:

Failed to start module [redfish] due to: [Error creating bean with name 
'org.springframework.beans.factory.config.PropertyPlaceholderConfigurer#0' 
defined in URL 
[jar:file:/usr/share/cloudstack-management/lib/cloudstack-4.19.0.1.jar!/META-INF/cloudstack/bootstrap/spring-bootstrap-context-inheritable.xml]:
 Cannot resolve reference to bean 'DefaultConfigResources' while setting bean 
property 'locations'; nested exception is 
org.springframework.beans.factory.NoSuchBeanDefinitionException: No bean named 
'DefaultConfigResources' available].

Any idea why this happens?


-Original Message-
From: Jayanth Babu A  
Sent: Monday, May 27, 2024 12:28 PM
To: users@cloudstack.apache.org
Subject: Re: HA for management server

Hello,
Yes, DB clustering with Galera or some other sort should be looked into. I'm 
not sure if [1] has worked for someone so far.

[1] 
https://docs.cloudstack.apache.org/en/latest/adminguide/reliability.html#configuring-database-high-availability

Thanks,
Jayanth
____
From: Alexandru Stan 
Sent: Monday, May 27, 2024 2:48 PM
To: users@cloudstack.apache.org 
Subject: RE: HA for management server

Hi,

Actually no, I'm just documenting the process so far, so nothing was done for 
the DB. For this I assume the info in the official documentation would be ok? 
Or something more advanced would be needed, like a galera cluster?



-Original Message-
From: Jayanth Babu A 
Sent: Monday, May 27, 2024 12:12 PM
To: users@cloudstack.apache.org
Subject: Re: HA for management server

Hello Alexandru,
Please see [1] and [2]. I assume your DB HA is taken care of.

[1] 
https://checkpoint.url-protection.com/v1/url?o=https%3A//docs.cloudstack.apache.org/en/latest/adminguide/reliability.html&g=ZmEwYjAwNmYwMjIxNTU5ZQ==&h=NmQyMzY5Nzc5YTU2ZDkyYjc4N2ExNDU4Y2Q0OWFkMjQ4OGRhMGZiZjdiNDA5N2NmNmRhM2EwYjUzZGEwMmI2Mg==&p=Y3AxZTpueHRnZW5pbmZpbml0ZWRhdGFjZW50ZXI6YzpvOmY0ZWFkYmU5NDkzMjMyMjdjY2EyM2IxN2YzOGZiMWE4OnYxOnA6VA==
[2] 
https://checkpoint.url-protection.com/v1/url?o=https%3A//github.com/apache/cloudstack/issues/8221&g=YzVlNDk3YjI5NTEwMTM5Mg==&h=OTBmN2Q4ZjA3YzRlOTAxZDk1YzcwMmExMTAzYTY4N2VmYjg5MGZkZGQyMDE2NTM3MTU4MGUxNDRjNzMzN2E0ZQ==&p=Y3AxZTpueHRnZW5pbmZpbml0ZWRhdGFjZW50ZXI6YzpvOmY0ZWFkYmU5NDkzMjMyMjdjY2EyM2IxN2YzOGZiMWE4OnYxOnA6VA==

Regards,
Jayanth Reddy
Engineering Manager

From: Alexandru Stan 
Date: Monday, 27 May 2024 at 2:38 PM
To: users@cloudstack.apache.org 
Subject: HA for management server
Hello,

Is there any documentation on deployment of multiple management servers in high 
availability? Or does anyone have good info on this? The official documentation 
does not contain anything relevant unfortunately.

Thank you!
Disclaimer *** This e-mail contains PRIVILEGED AND CONFIDENTIAL INFORMATION 
intended solely for the use of the addressee(s). If you are not the intended 
recipient, please notify the sender by e-mail and delete the original message. 
Further, you are not authorised to copy, disclose, or distribute this e-mail or 
its contents to any other person and any such actions are unlawful and strictly 
prohibited. This e-mail may contain viruses. NxtGen Datacenter & Cloud 
Technologies Private Ltd (“NxtGen”) has taken every reasonable precaution to 
minimize this risk but is not liable for any damage you may sustain as a result 
of any virus in this e-mail. You should carry out your own virus checks before 
opening the e-mail or attachment. NxtGen reserves the right to monitor and 
review the content of all messages sent to or from this e-mail address. 
Messages sent to or from this e-mail address may be stored on the NxtGen e-mail 
system. *** End of Disclaimer ***NXTGEN*** Disclaimer *** This e-mail contains 
PRIVILEGED AND CONFIDENTIAL INFORMATION intended solely for the use of the 
addressee(s). If you are not the intended recipient, please notify the sender 
by e-mail and delete the original message. Further, you are not authorised to 
copy, disclose, or distribute this e-mail or its contents to any other person 
and any such actions are unlawful and strictly prohibited. This e-mail may 
contain viruses. NxtGen Datacenter & Cloud Technologies Private Ltd (“NxtGen”) 
has taken every reasonable precaution to minimize this risk but is not liable 
for any damage you may sustain as a result of any virus in this e-mail. You 
should carry out your own viru

RE: Windows VM performance in ACS

[Alexandru Stan]

Hello,

Unfortunately no, it's the same. What is weird is that on a server instance 
(like Windows Server 2019) it's way better, maybe the issue applies only to 
desktop os's. I'll keep testing and get back if I have new information.


-Original Message-
From: Jayanth Babu A  
Sent: Tuesday, May 28, 2024 7:41 AM
To: users@cloudstack.apache.org
Subject: Re: Windows VM performance in ACS

Hi,
Just wanted to follow-up to see if you were able to resolve this.

Regards,
Jayanth Reddy

From: Jayanth Babu A 
Date: Friday, 24 May 2024 at 7:23 AM
To: users@cloudstack.apache.org 
Subject: Re: Windows VM performance in ACS Hello Alexandru, Alright. Please 
past the output of # virsh dumpxml 

Thanks,
Jayanth

________
From: Alexandru Stan 
Sent: Friday, May 24, 2024 2:01:35 am
To: users@cloudstack.apache.org 
Subject: RE: Windows VM performance in ACS

Did that and I'm not really sure that the situation is improved. Right after 
booting I started installing the latest CU for May and it was done in 1 hour 
and 30 minutes. I know that these things sometimes take a while to install but 
this seems way too much...


-Original Message-
From: Jayanth Babu A 
Sent: 23 May, 2024 21:55
To: users@cloudstack.apache.org
Subject: Re: Windows VM performance in ACS

Hello Alexandru,
Thanks for sharing. The reason could be that the VM is using IDE instead of 
virtio. Please stop the VM and add the key "rootDiskController" with value 
"virtio" and see the difference. You'll have to ensure that virtio drivers are 
installed on the VM so that it would be able to detect the disks.

Thanks,
Jayanth

____
From: Alexandru Stan 
Sent: Friday, May 24, 2024 12:21:03 am
To: users@cloudstack.apache.org 
Subject: RE: Windows VM performance in ACS

Hi Jayanth,

I don't have that key. In the settings section of the vm there are 2 keys, 
Message.ReservedCapacityFreed.Flag and rootdisksize.
I don't have a template yet, actually this vm was installed from a Windows ISO 
(that is registered as Windows PV in ACS) and I was hoping to transform it into 
a template with cloudbase-init, but since the performance is so bad I didn't 
get past the updates phase...

Alex


-Original Message-
From: Jayanth Babu A 
Sent: Thursday, May 23, 2024 4:54 PM
To: users@cloudstack.apache.org
Subject: Re: Windows VM performance in ACS

Hello Alexandru,
I’m not able to see the image. Would you please check the value for the key 
“rootDiskController” in both template and the Windows VM if the template OS 
type is “Windows PV” already? You’ll find it in the settings section.

Regards,
Jayanth Reddy

From: Alexandru Stan 
Date: Thursday, 23 May 2024 at 5:53 PM
To: users@cloudstack.apache.org 
Subject: RE: Windows VM performance in ACS Yes, I installed the guest tools 
right at first boot.

[cid:image001.png@01DAAD25.1CAA02B0]

From: Gary Dixon 
Sent: Thursday, May 23, 2024 2:32 PM
To: users@cloudstack.apache.org
Subject: RE: Windows VM performance in ACS

Have you checked in the VM's device manager that all devices are using virtio 
drivers (disk, disk controller, nic adapter etc) 
[cid:image535355.png@39364AC9.99EC2427]
Gary Dixon
Quadris Cloud Manager
[cid:image073403.png@5F5163CD.BCC16405]
+44 (0) 161 537 4980
 +44 (0) 7989 717661
[cid:image544450.png@462E4508.F6776E5F]
gary.di...@quadris.co.uk<mailto:gary.di...@quadris.co.uk>
[cid:image958545.png@2524980D.56B1E4D4]
https://checkpoint.url-protection.com/v1/url?o=www.quadris.com&g=OWQ3Y2VjYjI5MzFmMDhlYQ==&h=ODIxMzViOWYyNWQ3YzFhMTNhYmJiY2FjNWFjM2NlMTQxYzA0OTkxZmIxNzlkMTI0ZDQ4NWQ3ZjZhY2E2ZTVmNg==&p=Y3AxZTpueHRnZW5pbmZpbml0ZWRhdGFjZW50ZXI6YzpvOmEzMTc5ZmM5MjI1ZTc3NTIxOGEyMTUwOTlhZWE2NGNhOnYxOnA6VA==<https://checkpoint.url-protection.com/v1/url?o=http%3A//www.quadris.com&g=NjRjMmNhNWQxNWIyNTc5YQ==&h=Yzc1OTFlZWZiYjA5ZThkYzZkYzVkNzAzZGVkOWE5NGViZTliZGRhODBiYmM2ZGZhNDg1OTNmN2VjNThhYmJhNA==&p=Y3AxZTpueHRnZW5pbmZpbml0ZWRhdGFjZW50ZXI6YzpvOjdlOWE2YmVlMWZlMDUxZjU5YTM1MWZkYzc2YTVlOThmOnYxOmg6VA==><https://checkpoint.url-protection.com/v1/url?o=www.quadris.com&g=OWQ3Y2VjYjI5MzFmMDhlYQ==&h=ODIxMzViOWYyNWQ3YzFhMTNhYmJiY2FjNWFjM2NlMTQxYzA0OTkxZmIxNzlkMTI0ZDQ4NWQ3ZjZhY2E2ZTVmNg==&p=Y3AxZTpueHRnZW5pbmZpbml0ZWRhdGFjZW50ZXI6YzpvOmEzMTc5ZmM5MjI1ZTc3NTIxOGEyMTUwOTlhZWE2NGNhOnYxOnA6VA==%3chttps://checkpoint.url-protection.com/v1/url?o=http%3A//www.quadris.com&g=NjRjMmNhNWQxNWIyNTc5YQ==&h=Yzc1OTFlZWZiYjA5ZThkYzZkYzVkNzAzZGVkOWE5NGViZTliZGRhODBiYmM2ZGZhNDg1OTNmN2VjNThhYmJhNA==&p=Y3AxZTpueHRnZW5pbmZpbml0ZWRhdGFjZW50ZXI6YzpvOjdlOWE2YmVlMWZlMDUxZjU5YTM1MWZkYzc2YTVlOThmOnYxOmg6VA==%3e>
[cid:image378974.png@A6154DF9.C6D27CA5]
Innovation House, 12‑13 Bredbury Business Park Bredbury Park Way, Bredbury, 
Stockport, SK6 2SN -Original Message-
From: Alexandru Stan 
mailto:alexandru.s...@lifeincloud.com.INVALID>>
Sent: Thursday,

RE: HA for management server

[Alexandru Stan]

Hi,

Actually no, I'm just documenting the process so far, so nothing was done for 
the DB. For this I assume the info in the official documentation would be ok? 
Or something more advanced would be needed, like a galera cluster?



-Original Message-
From: Jayanth Babu A  
Sent: Monday, May 27, 2024 12:12 PM
To: users@cloudstack.apache.org
Subject: Re: HA for management server

Hello Alexandru,
Please see [1] and [2]. I assume your DB HA is taken care of.

[1] https://docs.cloudstack.apache.org/en/latest/adminguide/reliability.html
[2] https://github.com/apache/cloudstack/issues/8221

Regards,
Jayanth Reddy
Engineering Manager

From: Alexandru Stan 
Date: Monday, 27 May 2024 at 2:38 PM
To: users@cloudstack.apache.org 
Subject: HA for management server
Hello,

Is there any documentation on deployment of multiple management servers in high 
availability? Or does anyone have good info on this? The official documentation 
does not contain anything relevant unfortunately.

Thank you!
Disclaimer *** This e-mail contains PRIVILEGED AND CONFIDENTIAL INFORMATION 
intended solely for the use of the addressee(s). If you are not the intended 
recipient, please notify the sender by e-mail and delete the original message. 
Further, you are not authorised to copy, disclose, or distribute this e-mail or 
its contents to any other person and any such actions are unlawful and strictly 
prohibited. This e-mail may contain viruses. NxtGen Datacenter & Cloud 
Technologies Private Ltd (“NxtGen”) has taken every reasonable precaution to 
minimize this risk but is not liable for any damage you may sustain as a result 
of any virus in this e-mail. You should carry out your own virus checks before 
opening the e-mail or attachment. NxtGen reserves the right to monitor and 
review the content of all messages sent to or from this e-mail address. 
Messages sent to or from this e-mail address may be stored on the NxtGen e-mail 
system. *** End of Disclaimer ***NXTGEN***

HA for management server

[Alexandru Stan]

Hello,

Is there any documentation on deployment of multiple management servers in high 
availability? Or does anyone have good info on this? The official documentation 
does not contain anything relevant unfortunately.

Thank you!

RE: Windows VM performance in ACS

[Alexandru Stan]

Did that and I'm not really sure that the situation is improved. Right after 
booting I started installing the latest CU for May and it was done in 1 hour 
and 30 minutes. I know that these things sometimes take a while to install but 
this seems way too much...


-Original Message-
From: Jayanth Babu A  
Sent: 23 May, 2024 21:55
To: users@cloudstack.apache.org
Subject: Re: Windows VM performance in ACS

Hello Alexandru,
Thanks for sharing. The reason could be that the VM is using IDE instead of 
virtio. Please stop the VM and add the key "rootDiskController" with value 
"virtio" and see the difference. You'll have to ensure that virtio drivers are 
installed on the VM so that it would be able to detect the disks.

Thanks,
Jayanth

____
From: Alexandru Stan 
Sent: Friday, May 24, 2024 12:21:03 am
To: users@cloudstack.apache.org 
Subject: RE: Windows VM performance in ACS

Hi Jayanth,

I don't have that key. In the settings section of the vm there are 2 keys, 
Message.ReservedCapacityFreed.Flag and rootdisksize.
I don't have a template yet, actually this vm was installed from a Windows ISO 
(that is registered as Windows PV in ACS) and I was hoping to transform it into 
a template with cloudbase-init, but since the performance is so bad I didn't 
get past the updates phase...

Alex


-Original Message-
From: Jayanth Babu A 
Sent: Thursday, May 23, 2024 4:54 PM
To: users@cloudstack.apache.org
Subject: Re: Windows VM performance in ACS

Hello Alexandru,
I’m not able to see the image. Would you please check the value for the key 
“rootDiskController” in both template and the Windows VM if the template OS 
type is “Windows PV” already? You’ll find it in the settings section.

Regards,
Jayanth Reddy

From: Alexandru Stan 
Date: Thursday, 23 May 2024 at 5:53 PM
To: users@cloudstack.apache.org 
Subject: RE: Windows VM performance in ACS Yes, I installed the guest tools 
right at first boot.

[cid:image001.png@01DAAD25.1CAA02B0]

From: Gary Dixon 
Sent: Thursday, May 23, 2024 2:32 PM
To: users@cloudstack.apache.org
Subject: RE: Windows VM performance in ACS

Have you checked in the VM's device manager that all devices are using virtio 
drivers (disk, disk controller, nic adapter etc) 
[cid:image535355.png@39364AC9.99EC2427]
Gary Dixon
Quadris Cloud Manager
[cid:image073403.png@5F5163CD.BCC16405]
+44 (0) 161 537 4980
 +44 (0) 7989 717661
[cid:image544450.png@462E4508.F6776E5F]
gary.di...@quadris.co.uk<mailto:gary.di...@quadris.co.uk>
[cid:image958545.png@2524980D.56B1E4D4]
https://checkpoint.url-protection.com/v1/url?o=www.quadris.com&g=OWQ3Y2VjYjI5MzFmMDhlYQ==&h=ODIxMzViOWYyNWQ3YzFhMTNhYmJiY2FjNWFjM2NlMTQxYzA0OTkxZmIxNzlkMTI0ZDQ4NWQ3ZjZhY2E2ZTVmNg==&p=Y3AxZTpueHRnZW5pbmZpbml0ZWRhdGFjZW50ZXI6YzpvOmEzMTc5ZmM5MjI1ZTc3NTIxOGEyMTUwOTlhZWE2NGNhOnYxOnA6VA==<https://checkpoint.url-protection.com/v1/url?o=http%3A//www.quadris.com&g=NjRjMmNhNWQxNWIyNTc5YQ==&h=Yzc1OTFlZWZiYjA5ZThkYzZkYzVkNzAzZGVkOWE5NGViZTliZGRhODBiYmM2ZGZhNDg1OTNmN2VjNThhYmJhNA==&p=Y3AxZTpueHRnZW5pbmZpbml0ZWRhdGFjZW50ZXI6YzpvOjdlOWE2YmVlMWZlMDUxZjU5YTM1MWZkYzc2YTVlOThmOnYxOmg6VA==>
[cid:image378974.png@A6154DF9.C6D27CA5]
Innovation House, 12‑13 Bredbury Business Park Bredbury Park Way, Bredbury, 
Stockport, SK6 2SN -Original Message-
From: Alexandru Stan 
mailto:alexandru.s...@lifeincloud.com.INVALID>>
Sent: Thursday, May 23, 2024 11:17 AM
To: users@cloudstack.apache.org<mailto:users@cloudstack.apache.org>
Subject: RE: Windows VM performance in ACS

Hi Gary,

That’s what I used when registering the ISO. Doesn’t seem to help at all 
unfortunately.


From: Gary Dixon 
mailto:gary.di...@quadris.co.uk.INVALID>>
Sent: Thursday, May 23, 2024 1:14 PM
To: users@cloudstack.apache.org<mailto:users@cloudstack.apache.org>
Subject: RE: Windows VM performance in ACS

HI Alexandru

We use "Windows PV" OS type on our Windows VM's - this provides the Hyper-V 
enlightenments.

BR

Gary

[cid:image795113.png@DAEF2351.E3855BD4]
Gary Dixon
Quadris Cloud Manager
[cid:image752066.png@3A68EBBE.280BF225]
+44 (0) 161 537 4980
+44 (0) 7989 717661
[cid:image143416.png@CD1E068B.3BF4E714]
gary.di...@quadris.co.uk<mailto:gary.di...@quadris.co.uk<mailto:gary.di...@quadris.co.uk%3cmailto:gary.di...@quadris.co.uk>>
[cid:image596607.png@BDE27FFF.72E872DF]
https://checkpoint.url-protection.com/v1/url?o=http%3A//www.quadris.com/&g=N2M5MzIzZDYzMzk1YzU1ZA==&h=NTIwNzczY2E1ZjI5MGJkNzY4YmYyYzhlMmJmMmRhZjI2ZTU2N2E5MmQxMThmZDM1NTE5YWQwNjU2NTBkODZiYQ==&p=Y3AxZTpueHRnZW5pbmZpbml0ZWRhdGFjZW50ZXI6YzpvOmEzMTc5ZmM5MjI1ZTc3NTIxOGEyMTUwOTlhZWE2NGNhOnYxOnA6VA==<https://checkpoint.url-protection.com/v1/url?o=http%3A//www.quadris.com/&g=MjI3YTZlZjZhNTJhYzNlZg==&h=NzE1OTIyZGVhOWE3MmRiYjI4YTE0YWQ4MDI3NWZhMjg3NmQ2MGU3MzBhNzZiOGI1NmQyMDllZmU2MWEwMjJiNg==&p=Y3AxZTpueHRnZW5pbmZpbml0ZWRhdGFjZW50Z

RE: Windows VM performance in ACS

[Alexandru Stan]

Hi Jayanth,

I don't have that key. In the settings section of the vm there are 2 keys, 
Message.ReservedCapacityFreed.Flag and rootdisksize. 
I don't have a template yet, actually this vm was installed from a Windows ISO 
(that is registered as Windows PV in ACS) and I was hoping to transform it into 
a template with cloudbase-init, but since the performance is so bad I didn't 
get past the updates phase...

Alex


-Original Message-
From: Jayanth Babu A  
Sent: Thursday, May 23, 2024 4:54 PM
To: users@cloudstack.apache.org
Subject: Re: Windows VM performance in ACS

Hello Alexandru,
I’m not able to see the image. Would you please check the value for the key 
“rootDiskController” in both template and the Windows VM if the template OS 
type is “Windows PV” already? You’ll find it in the settings section.

Regards,
Jayanth Reddy

From: Alexandru Stan 
Date: Thursday, 23 May 2024 at 5:53 PM
To: users@cloudstack.apache.org 
Subject: RE: Windows VM performance in ACS Yes, I installed the guest tools 
right at first boot.

[cid:image001.png@01DAAD25.1CAA02B0]

From: Gary Dixon 
Sent: Thursday, May 23, 2024 2:32 PM
To: users@cloudstack.apache.org
Subject: RE: Windows VM performance in ACS

Have you checked in the VM's device manager that all devices are using virtio 
drivers (disk, disk controller, nic adapter etc) 
[cid:image535355.png@39364AC9.99EC2427]
Gary Dixon
Quadris Cloud Manager
[cid:image073403.png@5F5163CD.BCC16405]
+44 (0) 161 537 4980
 +44 (0) 7989 717661
[cid:image544450.png@462E4508.F6776E5F]
gary.di...@quadris.co.uk<mailto:gary.di...@quadris.co.uk>
[cid:image958545.png@2524980D.56B1E4D4]
www.quadris.com<https://checkpoint.url-protection.com/v1/url?o=http%3A//www.quadris.com&g=NjRjMmNhNWQxNWIyNTc5YQ==&h=Yzc1OTFlZWZiYjA5ZThkYzZkYzVkNzAzZGVkOWE5NGViZTliZGRhODBiYmM2ZGZhNDg1OTNmN2VjNThhYmJhNA==&p=Y3AxZTpueHRnZW5pbmZpbml0ZWRhdGFjZW50ZXI6YzpvOjdlOWE2YmVlMWZlMDUxZjU5YTM1MWZkYzc2YTVlOThmOnYxOmg6VA==>
[cid:image378974.png@A6154DF9.C6D27CA5]
Innovation House, 12‑13 Bredbury Business Park Bredbury Park Way, Bredbury, 
Stockport, SK6 2SN -----Original Message-
From: Alexandru Stan 
mailto:alexandru.s...@lifeincloud.com.INVALID>>
Sent: Thursday, May 23, 2024 11:17 AM
To: users@cloudstack.apache.org<mailto:users@cloudstack.apache.org>
Subject: RE: Windows VM performance in ACS

Hi Gary,

That’s what I used when registering the ISO. Doesn’t seem to help at all 
unfortunately.


From: Gary Dixon 
mailto:gary.di...@quadris.co.uk.INVALID>>
Sent: Thursday, May 23, 2024 1:14 PM
To: users@cloudstack.apache.org<mailto:users@cloudstack.apache.org>
Subject: RE: Windows VM performance in ACS

HI Alexandru

We use "Windows PV" OS type on our Windows VM's - this provides the Hyper-V 
enlightenments.

BR

Gary

[cid:image795113.png@DAEF2351.E3855BD4]
Gary Dixon
Quadris Cloud Manager
[cid:image752066.png@3A68EBBE.280BF225]
+44 (0) 161 537 4980
+44 (0) 7989 717661
[cid:image143416.png@CD1E068B.3BF4E714]
gary.di...@quadris.co.uk<mailto:gary.di...@quadris.co.uk<mailto:gary.di...@quadris.co.uk%3cmailto:gary.di...@quadris.co.uk>>
[cid:image596607.png@BDE27FFF.72E872DF]
http://www.quadris.com/<http://www.quadris.com/<https://checkpoint.url-protection.com/v1/url?o=http%3A//www.quadris.com/%253chttp%3A/www.quadris.com/&g=ZjQ2Y2VhYzY0YzMzM2NmZA==&h=MTY1YzQ3YTMwMDBmNmNhMjQzZGZhNDk0MGU0ZTdkM2FmMmFhMmEyNzkwY2Y1MWM3YjU0OGFlODY2MjgzZTAwNw==&p=Y3AxZTpueHRnZW5pbmZpbml0ZWRhdGFjZW50ZXI6YzpvOjdlOWE2YmVlMWZlMDUxZjU5YTM1MWZkYzc2YTVlOThmOnYxOmg6VA==>>
[cid:image382983.png@B32DDBFC.505CE7AD]
Innovation House, 12‑13 Bredbury Business Park Bredbury Park Way, Bredbury, 
Stockport, SK6 2SN -Original Message-
From: Alexandru Stan 
mailto:alexandru.s...@lifeincloud.com.INVALID<mailto:alexandru.s...@lifeincloud.com.INVALID%3cmailto:alexandru.s...@lifeincloud.com.INVALID>>>
Sent: Thursday, May 23, 2024 10:40 AM
To: 
users@cloudstack.apache.org<mailto:users@cloudstack.apache.org<mailto:users@cloudstack.apache.org%3cmailto:users@cloudstack.apache.org>>
Subject: Windows VM performance in ACS

Hi everyone,

Is there anything that must be done to improve the performance of Windows VMs 
in ACS? I tried to deploy a W10 instance, with the virtio drivers, and the 
performance is very, very poor (for example it took almost 3 hours to install a 
CU, the actual setup took about 50 minutes and so on). The hypervisor kvm on a 
HP server, with ssd's for storage. In other apps there are those hyper-v 
enlightments that can be enabled for Windows guests and it really makes a 
difference, but I have no idea if this is a thing in ACS too. If someone has 
some info on this, please share.

Thank you!
Disclaimer *** This e-mail contains PRIVILEGED AND CONFIDENTIAL INFORMATION 
intended solely for the use of the addressee(s). If you are not the intended 
recipient, please notify the sender by e-mail and

RE: Windows VM performance in ACS

[Alexandru Stan]

Yes, I installed the guest tools right at first boot.

[cid:image001.png@01DAAD25.1CAA02B0]

From: Gary Dixon 
Sent: Thursday, May 23, 2024 2:32 PM
To: users@cloudstack.apache.org
Subject: RE: Windows VM performance in ACS

Have you checked in the VM's device manager that all devices are using virtio 
drivers (disk, disk controller, nic adapter etc)

[cid:image535355.png@39364AC9.99EC2427]
Gary Dixon
Quadris Cloud Manager
[cid:image073403.png@5F5163CD.BCC16405]
+44 (0) 161 537 4980
 +44 (0) 7989 717661
[cid:image544450.png@462E4508.F6776E5F]
gary.di...@quadris.co.uk<mailto:gary.di...@quadris.co.uk>
[cid:image958545.png@2524980D.56B1E4D4]
www.quadris.com<http://www.quadris.com>
[cid:image378974.png@A6154DF9.C6D27CA5]
Innovation House, 12‑13 Bredbury Business Park
Bredbury Park Way, Bredbury, Stockport, SK6 2SN
-Original Message-
From: Alexandru Stan 
mailto:alexandru.s...@lifeincloud.com.INVALID>>
Sent: Thursday, May 23, 2024 11:17 AM
To: users@cloudstack.apache.org<mailto:users@cloudstack.apache.org>
Subject: RE: Windows VM performance in ACS

Hi Gary,

That’s what I used when registering the ISO. Doesn’t seem to help at all 
unfortunately.


From: Gary Dixon 
mailto:gary.di...@quadris.co.uk.INVALID>>
Sent: Thursday, May 23, 2024 1:14 PM
To: users@cloudstack.apache.org<mailto:users@cloudstack.apache.org>
Subject: RE: Windows VM performance in ACS

HI Alexandru

We use "Windows PV" OS type on our Windows VM's - this provides the Hyper-V 
enlightenments.

BR

Gary

[cid:image795113.png@DAEF2351.E3855BD4]
Gary Dixon
Quadris Cloud Manager
[cid:image752066.png@3A68EBBE.280BF225]
+44 (0) 161 537 4980
+44 (0) 7989 717661
[cid:image143416.png@CD1E068B.3BF4E714]
gary.di...@quadris.co.uk<mailto:gary.di...@quadris.co.uk<mailto:gary.di...@quadris.co.uk%3cmailto:gary.di...@quadris.co.uk>>
[cid:image596607.png@BDE27FFF.72E872DF]
http://www.quadris.com/<http://www.quadris.com/<http://www.quadris.com/%3chttp:/www.quadris.com/>>
[cid:image382983.png@B32DDBFC.505CE7AD]
Innovation House, 12‑13 Bredbury Business Park Bredbury Park Way, Bredbury, 
Stockport, SK6 2SN -Original Message-
From: Alexandru Stan 
mailto:alexandru.s...@lifeincloud.com.INVALID<mailto:alexandru.s...@lifeincloud.com.INVALID%3cmailto:alexandru.s...@lifeincloud.com.INVALID>>>
Sent: Thursday, May 23, 2024 10:40 AM
To: 
users@cloudstack.apache.org<mailto:users@cloudstack.apache.org<mailto:users@cloudstack.apache.org%3cmailto:users@cloudstack.apache.org>>
Subject: Windows VM performance in ACS

Hi everyone,

Is there anything that must be done to improve the performance of Windows VMs 
in ACS? I tried to deploy a W10 instance, with the virtio drivers, and the 
performance is very, very poor (for example it took almost 3 hours to install a 
CU, the actual setup took about 50 minutes and so on). The hypervisor kvm on a 
HP server, with ssd's for storage. In other apps there are those hyper-v 
enlightments that can be enabled for Windows guests and it really makes a 
difference, but I have no idea if this is a thing in ACS too. If someone has 
some info on this, please share.

Thank you!

RE: Windows VM performance in ACS

[Alexandru Stan]

Hi Gary,

That’s what I used when registering the ISO. Doesn’t seem to help at all 
unfortunately.


From: Gary Dixon 
Sent: Thursday, May 23, 2024 1:14 PM
To: users@cloudstack.apache.org
Subject: RE: Windows VM performance in ACS

HI Alexandru

We use "Windows PV" OS type on our Windows VM's - this provides the Hyper-V 
enlightenments.

BR

Gary

[cid:image795113.png@DAEF2351.E3855BD4]
Gary Dixon
Quadris Cloud Manager
[cid:image752066.png@3A68EBBE.280BF225]
+44 (0) 161 537 4980
 +44 (0) 7989 717661
[cid:image143416.png@CD1E068B.3BF4E714]
gary.di...@quadris.co.uk<mailto:gary.di...@quadris.co.uk>
[cid:image596607.png@BDE27FFF.72E872DF]
www.quadris.com<http://www.quadris.com>
[cid:image382983.png@B32DDBFC.505CE7AD]
Innovation House, 12‑13 Bredbury Business Park
Bredbury Park Way, Bredbury, Stockport, SK6 2SN
-Original Message-
From: Alexandru Stan 
mailto:alexandru.s...@lifeincloud.com.INVALID>>
Sent: Thursday, May 23, 2024 10:40 AM
To: users@cloudstack.apache.org<mailto:users@cloudstack.apache.org>
Subject: Windows VM performance in ACS

Hi everyone,

Is there anything that must be done to improve the performance of Windows VMs 
in ACS? I tried to deploy a W10 instance, with the virtio drivers, and the 
performance is very, very poor (for example it took almost 3 hours to install a 
CU, the actual setup took about 50 minutes and so on). The hypervisor kvm on a 
HP server, with ssd's for storage. In other apps there are those hyper-v 
enlightments that can be enabled for Windows guests and it really makes a 
difference, but I have no idea if this is a thing in ACS too. If someone has 
some info on this, please share.

Thank you!

Windows VM performance in ACS

[Alexandru Stan]

Hi everyone,

Is there anything that must be done to improve the performance of Windows VMs 
in ACS? I tried to deploy a W10 instance, with the virtio drivers, and the 
performance is very, very poor (for example it took almost 3 hours to install a 
CU, the actual setup took about 50 minutes and so on). The hypervisor kvm on a 
HP server, with ssd's for storage. In other apps there are those hyper-v 
enlightments that can be enabled for Windows guests and it really makes a 
difference, but I have no idea if this is a thing in ACS too. If someone has 
some info on this, please share.

Thank you!

ACS VM HA and CPU type

[Alexandru Stan]

Hi everyone. I have a couple of questions regarding these features in ACS, 
maybe someone here can share some info. So, first of all, is there anything 
else that I have to do, other than setting the cpu type in the agent properties 
file, to make the host cpu model show up correctly in a vm? I'm trying to set 
it to host-passtrough but doesn't seem to work, in a Windows vm the cpu still 
shows as qemu-something.

The other matter is about the vm high availability feature. It works for me, as 
in the vms are restarted on another host if the original host goes down, but 
what I don't know is what happens "behind the doors". I know there are some 
health checks being done and maybe other stuff, but what, exactly? Is there any 
way to influence this, like shorten the time between the moment when the host 
goes down and the vm power-on on a different host? Right now it takes about 
15-20 minutes for this to happen, seems quite long. I tried to modify some 
parameters in the global configuration but it makes no difference.

Thank you!

RE: Storage solution for Cloudstack

[Alexandru Stan]

Hi Bryan,

Not much luck with this I'm afraid. I followed the documentation to the letter 
and I managed to add a 3-node primary storage in CS, I can see the random named 
linstor resources created by CS, yet I cannot create any instances, there's 
always error at startup, like the one bellow.

Unable to orchestrate start VM instance 
{"id":18,"instanceName":"i-2-18-VM","type":"User","uuid":"867503d9-3912-47e4-85be-a30b3e088aa4"}
 due to [Create volume from template (ID = 4) failed: Resource definition 
'cs-null' not found.].

This is the method I used, except I didn't create a zfs pool, I created a 
thin-lvm one: 
https://linbit.com/drbd-user-guide/linstor-guide-1_0-en/#s-linstor-cloudstack-installing-linstor

Any idea why this is happening?


-Original Message-
From: Bryan Tiang  
Sent: Friday, March 29, 2024 4:38 PM
To: users@cloudstack.apache.org; users@cloudstack.apache.org
Subject: RE: Storage solution for Cloudstack

Hey Alexandru,

The architecture is not quite the same because Linstors secret sauce is that 
they use DRBD, which is what allows them to get such high performance and low 
latency.

You can check up more about it here: 
https://linbit.com/blog/how-does-linstor-compare-to-ceph/#:~:text=Ceph's%20use%20of%20the%20CRUSH,Ceph%20isn't%20operating%20normally.

And some performance benchmarks with ceps, here: 
https://blog.palark.com/kubernetes-storage-performance-linstor-ceph-mayastor-vitastor/

Regards,
Bryan
On 29 Mar 2024 at 4:11 PM +0800, Bryan Tiang , wrote:
>
>
> Bryan

RE: Storage solution for Cloudstack

[Alexandru Stan]

Hi Bryan,

Interesting, thank you for the suggestion. Is this similar to ceph, 
installation wise? I mean a controller and storage nodes added to it?



-Original Message-
From: Bryan Tiang  
Sent: Friday, March 29, 2024 4:11 PM
To: users@cloudstack.apache.org; users@cloudstack.apache.org
Subject: Re: Storage solution for Cloudstack

Hey Alexandru,

We are using Cloudstack + Linstor right now and are live with a few clients.

Its open source, performance is pretty great, and they support things like VM 
Snapshots, Volume Snapshots, Storage Replication, Encryption in Rest/Transit, 
VM HA and more.

I think it suits what you’re looking for.

Regards,
Bryan
On 29 Mar 2024 at 10:04 PM +0800, Alexandru Stan 
, wrote:
> Hi everyone,
>
> We have a specific scenario in witch we want to have a functional Cloudstack 
> cluster but we can't seem to find/decide on a working storage solution, so if 
> anyone can recommend a setup that meets all our requirements, please do. So, 
> we are using KVM for hypervisors and we want to be able to offer vm snapshots 
> functionality to the clients and also vm HA - that excludes Ceph. If we go 
> for NFS, we don't have redundancy, since we're using nvme disks only and I 
> believe we all know what ar the choices of raid controllers for this right 
> now.
> We're considering Storpool too, but we've yet to receive a price list, and we 
> assume it's not cheap but who knows.
>
> Thank you!

Storage solution for Cloudstack

[Alexandru Stan]

Hi everyone,

We have a specific scenario in witch we want to have a functional Cloudstack 
cluster but we can't seem to find/decide on a working storage solution, so if 
anyone can recommend a setup that meets all our requirements, please do. So, we 
are using KVM for hypervisors and we want to be able to offer vm snapshots 
functionality to the clients and also vm HA - that excludes Ceph. If we go for 
NFS, we don't have redundancy, since we're using nvme disks only and I believe 
we all know what ar the choices of raid controllers for this right now.
We're considering Storpool too, but we've yet to receive a price list, and we 
assume it's not cheap but who knows.

Thank you!