This browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(28.799999999999997, 96%, 12%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ENJ%3C/text%3E%3C/svg%3E)
kubectl get pods -n teams-recording-bot
output:
NAME READY STATUS RESTARTS AGE
teams-recording-bot-0 0/1 ContainerCreating 0 42m
can't setup certifcates:
kubectl get cert -n teams-recording-bot
NAME READY SECRET AGE
ingress-tls False ingress-tls 41m
kubectl logs -n teams-recording-bot teams-recording-bot-0
Error from server (BadRequest): container "recording-bot" in pod "teams-recording-bot-0" is waiting to start: ContainerCreating
pod logs:
[Pod] [teams-recording-bot-0] FailedMount: Unable to attach or mount volumes: unmounted volumes=[certificate], unattached volumes=[kube-api-access-bnvqt certificate]: timed out waiting for the condition
11 mins ago [Pod] [teams-recording-bot-0] FailedMount: Unable to attach or mount volumes: unmounted volumes=[certificate], unattached volumes=[certificate kube-api-access-bnvqt]: timed out waiting for the condition
1 min ago [Pod] [teams-recording-bot-0] FailedMount: MountVolume.SetUp failed for volume "certificate" : secret "ingress-tls" not found
682 msecs ago [Pod] [teams-recording-bot-0] Scheduled: Successfully assigned teams-recording-bot/teams-recording-bot-0 to aksscale000000
Hello @Nikheel Jain
I hope you are doing fine.
I would like to provide my humble opinion on this case scenario.
Furthermore, there is a good point to be shared as shown below:
Use one bot channel registration per environment, since your endpoint changes when you switch between local development, staging, and production environments.
Deleting the Teams channel registration will cause a new pair of keys to be generated when it is re-enabled. This invalidates all 29:xxx and a:xxx IDs that the bot may have stored for proactive messaging.
I hope this can help you the behavior observed until now buddy : )
Looking forward to your feedback,
Best Regards,
Please "Accept the answer" if the information helped you. This will help us and others in the community as well.
Apologies for the delay in responding.
I have read through this requirements doc of the media bot to get an audio stream during the initial study of the repo. I do agree with you AKS is at PaaS service but the underlying node on which the bot will be deployed is a VM instance which is IaaS.
Moreover, the endpoint registration here means the endpoint we mentioned while creating the azure bot and giving MS teams access, Since it's the endpoint the bot will finally listen to join a call. It changes during deployment because we get cluster IP / Public IP while on local setup it works through ngrok.
Also to add these steps for deployment are mentioned in the official sample repo ref:https://github.com/microsoftgraph/microsoft-graph-comms-samples/tree/master/Samples/V1.0Samples/AksSamples/teams-recording-bot
so I believe these steps and setup are supported.
The error we are facing is of secrets not mounting if we leave the teams recording bot code apart for a second, I guess the certs should get mounted in the windows node pool as it is not the problem with code but deployment configurations. Maybe we are missing some configurations which are different from the Linux node pool.
Once Again thanks for helping.
No worries for the delay.
Let me bring back one of your statements...
I have read through this requirements doc of the media bot to get an audio stream during the initial study of the repo. I do agree with you AKS is at PaaS service but the underlying node on which the bot will be deployed is a VM instance which is IaaS.
I just wanted to remind here that the master node is a managed instance which is handled by Azure.
Cheers,
I second that
Hi,
Just wanted to shared this detail as last comment for this. See below:
Thanks for the reply.
So this means the VM machine that we are deploying doesn't support persistence volums
Thank you for your quick response.
Yeap that the whole problem and I had to read a lot as well to get to this answer : )
I hope that my assistance was useful for you buddy.
Have a good one.
Best Regards,
Please "Accept the answer" if the information helped you. This will help us and others in the community as well.
I know that it required thorough searching and I am sure it was time-consuming.
I really appreciate your efforts and you deserve the praise for going out of your way to help me.
Thanks a lot.
Btw the repo admin replied to the issue that I created on the official GitHub repo. I am going through that:
ref:
https://github.com/microsoftgraph/microsoft-graph-comms-samples/issues/570
Thanks a lot for your comment buddy and it was a pleasure.
Nice that you got response on GitHub.
Cheers : )
As the error message indicates, there is a mount failure : FailedMount: MountVolume.SetUp failed for volume "certificate"
Probably, you need to check the reason behind the mount failure and try creating pods again
Can't find any concrete reason for not mounting
If possible, you can share the code so that others can check what is wrong
the commands used to set up:
https://github.com/jinik21/TeamsRecordingBot/blob/master/03_IaC/00_AzureCLI/02_AKS_iac.ps1
deploment yaml files:
https://github.com/jinik21/TeamsRecordingBot/tree/master/00_RecordingBot/deploy/teams-recording-bot
It sounds like the path indicated in following section of deployment.yaml should be changed as a valid path like /host_mnt/c/certs
volumeMounts:
- mountPath: "C:/certs/"
name: certificate
readOnly: true
I changed the path but still it's not working:
logs:
3 mins ago [Pod] [teams-recording-bot-0] FailedMount: Unable to attach or mount volumes: unmounted volumes=[certificate], unattached volumes=[certificate kube-api-access-h9qdq]: timed out waiting for the condition
1 min ago [Pod] [teams-recording-bot-0] FailedMount: MountVolume.SetUp failed for volume "certificate" : secret "ingress-tls" not found
58 secs ago [Pod] [teams-recording-bot-0] FailedMount: Unable to attach or mount volumes: unmounted volumes=[certificate], unattached volumes=[kube-api-access-h9qdq certificate]: timed out waiting for the condition
433 msecs ago [Pod] [teams-recording-bot-0] Scheduled: Successfully assigned teams-recording-bot/teams-recording-bot-0 to aksscale000000
cert manager can't locate tls-ingress due to volume mount and also gives error
Have a try with different formats for mounting the drive. Once the mount is successful, the certificate read will; also work.
You may try like c://certs//, c:\certs\, c:\certs\, c:\certs\, //c/certs/ etc.
tried all , where can i check if volume is there or not?
any help?
Hi
Did you check if the package nfs-common is available?
Cheers,
How do i check it? I am not able to create debug node and ssh into windows nodepool.
Thanks for the response.
Have you tried it?
Yes the debug pod remains in containercreating state .
Got it.
What did you see from AKS diagnostics? Besides the logs given before...
Aks diagnostics seems fine as these are warnings the logs i was getting was from the insights live events
all the other pods nginx controller,cert manager are in ready state and working fine, just problem with teams recording bot.
https://github.com/jinik21/TeamsRecordingBot/blob/master/03_IaC/00_AzureCLI/02_AKS_iac.ps1
these are setup commands I have used
I tried changing kube version to 1.22.4 but still error was there.
setup logs:
https://github.com/jinik21/TeamsRecordingBot/runs/7086339232?check_suite_focus=true
So based on your description, you have others PODs running normally but the one stated all this time is the one affected.
If I understood correctly you might try to review IP allocation space, NSG's, node count limit and quota usage limitation...
Another step could be this one below:
az vm update -n <VM_NAME> -g <RESOURCE_GROUP_NAME>
Regards,
Thanks for response,
node count and quota usage limitations are not their I have already checked that.
was following this:
https://github.com/kubernetes/ingress-nginx/issues/5932
but couldn't get it solved for windows pool
this is the fix solution for it as per the following comment on that GitHub thread...
can you please tell where do i add it. I couldn't figure out where to add it.
I created this issue. jinik21 is my github id . didn't get any response on official microsoft repo issue till now.
I thought about it hehe but could not correlate this user against GitHub user....
On which region is this hosted? try review this >> https://releases.aks.azure.com/webpage/index.html#tabus
As well as this other link that can be used to filter a similar issue or bug fix notes.
centralus , eastus , central france, central india tried all these but got same error.
what about redeploying it?
I have to redeploy every time I make a change with changed bot name, acrname etc because azure then gives error try to restore a deleted resource. So I redeploy every time and delete previous deployment as I have Azure student access.
Runs:
https://github.com/jinik21/TeamsRecordingBot/actions
If possible can you try setting up once on your own. I can guide through step it's pretty much automated
Let me see and I can get it done during the afternoon.
When you stated that you have student access... Is this Azure Lab Services??????????
Cheers,
Appreciate that man. It has complete access the one you get with github developer pack got 100 dollar worth credit.
Steps to setup updated command:
az ad sp create-for-rbac -n "RecBotGitHubActions" --role
Owner --sdk-auth --scopes /subscriptions/yourid
this will create a json object copy it and add as github secret.
rest follow steps in git readme. basically commit will start github action1 then add ip A record and registry password and username in secret.
and manually run workflow 2 then setup will start.
follow this repo as I have fixed errors related to kub and also depreciated commands.
https://github.com/jinik21/TeamsRecordingBot
Again thanks a lot . If any doubt setting up do ping me up.
Your welcome buddy.
I shall let me know about it at the end of the day how it was as well as end results.
See ya : )
Hi,
I have gone through the link as well as README file : )
Having said that, I do have a question for ya and it is.... Is a Must to have a O365 account with Admin role on my user?
Looking forward to your feedback.
Regards,
As such setup doesn't require it but the bot will require office 365 enabled Tennant .
You can get from Microsoft developer website for free.
https://www.google.com/url?sa=t&source=web&rct=j&url=https://developer.microsoft.com/en-us/microsoft-365/dev-program&ved=2ahUKEwilwunR9tH4AhXASWwGHdknBWcQFnoECAsQAQ&usg=AOvVaw1b-HxZC7OdFAEyL1YgtJem
Great detail.
Another question is that I was reading that it takes around 2 hours to be fully provisioned.... Did I understand correctly?
Because I could be doing this and get it done after work buddy (since I am volunteer : )
Finally, can we elaborate steps taken on this TS thread?
Cheers,
It's fully automated you have to just run GitHub actions and all resources will be provisioned and run automatically.Takes around 45-50 min. Steps are as follows.:
Fell free to do it whenever possible and at your pace buddy.
Thanks a lot buddy!!
Hi Buddy
I just wanted to bother you with this replica...
I have gone through this as well as the O365 account creation.
But there is a part where you need to create an azure Bot services and found this below:
They renamed Bot Channels Registration to just Bot channels.
For creating an azure bot:
Thanks for the heads up.
Sorry for my silly question but it is just to double-check one thing... Is this an AKS cluster using Windows server or is this a WebApp instance?
Cheers,
No worries.
AKS cluster using windows node pool. basically, it creates an aks cluster(system node Linux) then adds windows nodepool and deploys on it. This is needed due to restrictions by Microsoft that the bot should be deployed on the windows server to access stream data.
detailed deployment structure ref:https://github.com/jinik21/TeamsRecordingBot/blob/master/00_RecordingBot/docs/deploy/aks.md
Thanks again for the clarification on this.
Keep doing this later and I will be getting back with any findings : )
Redefine pods may help
I tried redeploying them but no good
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(19.2, 15%, 11%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EHB%3C/text%3E%3C/svg%3E)
As the error message indicates, there is a mount failure : FailedMount: MountVolume.SetUp failed for volume "certificate"
Probably, you need to check the reason behind the mount failure and try creating pods again!!!!