I'm creating a custom vhd image to make it available in azure market place. Its an inbuild centos 7 machine and I want to load it as an image. Is it mandatory to have WALinuxAgent agent to be installed as part of the image? Here is the documentation which says to install it https://learn.microsoft.com/en-us/azure/virtual-machines/linux/create-upload-centos#centos-70. I want to exclude that package from the image. Is that okay to do that ?
You do need it, otherwise the VM may not boot. This agent communicates with the Azure fabric to provide VM status amongst other things. You can find more information about it here: https://github.com/Azure/WALinuxAgent
Related
We are facing now a new challenge, how can we autoinstall a Linux image I created.
I am installing my custom Linux image on a Windows 10 Mini pc. I would like the auto install to run on boot for a security and automation reasons.
I used Clonezilla in order to capture and restore a custom image. It worked with S3fs as well.
I tried to enable PXE, but it didn't work. From a screen that says PXE..... I just went again to Windows loading logo.
So the process:
I created a custom image with all the dependencies and services I need in order the application to run, I captured it with Clonezilla and restored it with the same open source application. It was great. Now we would like to automate the process as much as we can in order to deploy the image on a verity of physicals servers.
My main issue right now with Clonezilla is that in order to restore from S3 for example I need to use S3fs and I have to create a credentials file with my access key id and secret access key each time I restore an image. Repetitive and insecure.
I explored the idea of autoinstall the image with PXE over the internet, interesting but I haven't fully understand the process.
We are not cancelling an payment application, if it's stable and provide support.
If anyone know how I can create a new Clonezilla image with my AWS credentials file exist it could be helpful, if anyone can share his experience with auto install custom image over the internet with PXE it would be helpful, or even an experience with a different tool that works for you, I would love to hear about it.
Please le me know if I need to add addition information and details or if this is not the right thread for this question.
Thank you for your help
Being quite new to Azure, does anyone know how to map a Blob File storage on Linux PHP7.2 box as PAAS ?
I looked at the documentation followed the instruction, but they do not work on my App Service.
https://learn.microsoft.com/en-us/azure/storage/blobs/storage-how-to-mount-container-linux
According to this documentation, Blobfuse needs to be installed in order for this to work.
I cannot install Blobfuse, lsb_release -a does not work (part of the instructions), according to terminal the installed package is linux 4.4.0-128.154.
I can find a Wordpress plugin that uses the Wordpress CLI, but that is not really what I am looking for. (I dont use Wordpress on my site, but I thought the code could give me a hint)
I am just looking for a file share to add store uploaded images.
Has anyone done this before?
Thanks in advance !
I'm looking for someone who was able to follows the documentation of Microsoft (https://learn.microsoft.com/en-us/azure/iot-edge/tutorial-deploy-function) and Jon Gallant (https://blog.jongallant.com/2017/11/azure-iot-edge-raspberrypi/) with success.
After following the whole documentation to get a working Azure Function on a Raspberry PI hosted in a docker container, the Function does not work and the edgeAgent log contains just an info that the Functions can't start. To find out what is happen, I setup the debug environment like described here: https://learn.microsoft.com/en-us/azure/iot-edge/how-to-vscode-debug-azure-function but nothing works. After hit the last F5 (as its described) I get the error, that the docker container does not exists.
Yes, Dariusz is right.
So, I went to check the status of Azure Functions Runtime and the good news is that they have now a version of their runtime for ARM. https://hub.docker.com/r/microsoft/azure-functions-runtime/tags/ for ARM.
So, we just need to update our Binding and start publishing our image for Arm.
I opened this GitHub Issue to rack the status:
https://github.com/Azure/iot-edge/issues/485
UPDATE: We have release 1.0.0-preview022 that has ARM version for Function that works for raspberry pi! Let us know if you found any problem. We should be updating our VS Code Template soon.
As of today there is no Azure Functions image for ARM released. If you look at the docker hub url https://hub.docker.com/r/microsoft/azureiotedge-functions-binding/tags/ there are just Windows Nano container and Linux x64 container available.
I am new to Azure Portal but have good knowledge with VMware ESX and vcenter. I created some vm images in Azure portal (not classic) and need to clone them. However, all online docs point to "Capture" button which for some reason is missing from my profile. I found some docs about using AZcopy tool which is fairly complex.
also, because I have some automation customized in the VM, running sysprep might break the tool/automation. so I prefer without running sysprep.
So the bottom line is that I need a way to clone single VM into multiple VMs without running sysprep. Yes I know about the conflicts and hopping I can do the modifications manually. But even if it is impossible without sysprep, fine. As long as I can clone them I'll live with sysprep somehow.
This is possible using Powershell and sysprep. This blog post has a step by step of how it's done.
http://www.codeisahighway.com/how-to-capture-your-own-custom-virtual-machine-image-under-azure-resource-manager-api/
If the Capture button is missing try to check whether your machine is a "Managed disk" one.
If it's not you can follow this tutorial to turn it on but I guess there's costs associate to it.
https://learn.microsoft.com/en-us/azure/virtual-machines/windows/convert-unmanaged-to-managed-disks
ps: last time I checked "unmanaged disks" and I spent like three hours trying to understand why the capture button has gone.
My goal is to create a Deployment Manager profile in my Websphere on Linux.
Reading tons of documentation pages gives just two methods:
1) using X GUI application WAS_root/bin/ProfileManagement/pmt.sh
In this case, according to manuals, i should choose Management option, click "Next" and choose Deployment Manager server type.
Actually when i run pmt.sh in my WAS installation there is no option to choose Deployment Manager in Management section and only one profile type that i'm able to create is AdminAgent.
2) using manageprofiles.sh script under WAS_root/bin directory.
The manuals say that i can use -serverType DEPLOYMENT_MANAGER option in order to create deployment manager profile.
Actually when i run the script:
./manageprofiles.sh -create -templatePath ../profileTemplates/management/ -ServerType DEPLOYMENT_MANAGER -isDefault -profileName dmgr -adminUserName websphere -adminPassword websphere1
I get the following message:
The following validation errors were present with the command line
arguments:
serverType: The value for this parameter must be within this set of values [ADMIN_AGENT]
That means that i don't have the ability to create Deployment Manager at all.
Please advise what steps i can perform to be able to install Deploy Manager except erase my WAS installation and install it from scratch once again.
Thanks a lot.
You have standalone aka base edition of WebSphere Application Server. To create Deployment Manager you need Network Deployment edition.
Unfortunately you will have to install it from scratch from different installation files (from ND, not base or developers edition).