Replies: 2 comments 17 replies
-
|
I am not sure if this a ARM issue, but it's looking for a bridge with the name : br-1460 and that fails. Are you sure this Agent is configured properly in the agent.properties? |
Beta Was this translation helpful? Give feedback.
1 reply
-
|
yes we have configured the agent with cloudbr0 bridge which we have been doing with the x86 hosts, the normal instance making on the x86 is working peoperly but only for this arm based system we are facing this issue even we have added this host in a different cluster which is set to arm |
Beta Was this translation helpful? Give feedback.
-
|
Currently we are trying to access the nvidia jetson agx thor to onboard it on our cloudstack platform but we are still facing same kind of error Can you guys give us any solution on how can we use these kind of devices like nvidia jetson orin, nvidia agx thor, or nvidia dgx spark to onboard them on cloudstack and use them for ml related works? |
Beta Was this translation helpful? Give feedback.
16 replies
-
|
Hey @prashanthr2, These are you desired outputs listed below. This is the output of This is the |
Beta Was this translation helpful? Give feedback.
-
|
@arpanbht Cloudstack tries to set network bandwidth for the VM nic to 200MBps (by default), this can be seen as part of libvirtxml as below Quick search on this tells me that HTB module is missing in the linux kernel. try below commands to see if the module exists lsmod | grep sch_htb if the above command returns nothing, try to load it manually as below. If modprobe fails as well. Install the package(based on your OS on KVM host) as mentioned here https://stackoverflow.com/questions/64755248/linux-fedora-tc-qdisc-gets-error-specified-qdisc-not-found and see if that resolves your issue sudo modprobe sch_htb If the issue still happens after installing the required package, send new set of debug logs after reproducing the issue. |
Beta Was this translation helpful? Give feedback.
-
|
Our machine is NVIDIA Jetson AGX Thor, an arm based machine and this command Our machine has nvidia's custom jetpack ubuntu os installed into it and for driver issues related to other tasks which we do on that machine we cannot reinstall it with custom ubuntu server image. |
Beta Was this translation helpful? Give feedback.
-
|
@arpanbht AFAIK, ubuntu provided by NVIDIA misses some kernel modules. As a quick workaround, you can try removing the network throttling by updating Another option is to rebuild the kernel provided by NVIDIA with the missing modules. Let me know how it goes. It'll help others in the community trying to run CloudStack on Jetson devices. |
Beta Was this translation helpful? Give feedback.
-
|
@vishesh92 
After 5 to 10 minutes the login screen is showing but if I type ubuntu as a username and enter the password it is saying incorrect login. If I create a vm using an x86 template on a x86 machine then no issues are coming I can easily get into the vm. This is for a single template this output is coming but for another kind of template this is the output screen I am getting below, 
And this is the log which is coming during creating the instance on arm based machine jetson agx thor. Now when I am creating a vm using a gpu offering as jetson agx thor has gpu the vms are getting stuck on starting phase and this is the debug log for your reference |
Beta Was this translation helpful? Give feedback.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
I have made a cluster which is arm based and trying to add nvidia jetson orin as a kvm host so its getting added to the management server and up and running.
But while creating a vm inside this host I am facing this kind of error
what is this error coming while creating an instance in this host give me solution
What is the problem? Cannot understand the error log.
Beta Was this translation helpful? Give feedback.
All reactions