Daneil, thank you so much for the video. You saved my job.
@DevOpsDirective
4 жыл бұрын
@Daniel Persson, great overview of horizontal autoscaling -- Subscribed! What motivated the choice to use the custom autoscale.sh script and poll RabbitMQ directly vs configuring a custom metric using custom.metrics.k8s.io which would then enable the normal HPA resource to handle the autoscaling?
@DanielPersson
4 жыл бұрын
Hi DD. Well, I did not know about that option. I think the solution we have now is really direct and solves the problem but will scale up and down really quickly. And perhaps a custom metric would be even greater and more stable. When I have the time I will look into how to do this and see if I could get a custom metric to work. Thank you for watching my videos. Best regards Daniel
@DevOpsDirective
4 жыл бұрын
@@DanielPersson Got it -- thanks for the response! I have not implemented a custom metric myself so I am not sure how much work that would be.
@OctoCoreHU
4 жыл бұрын
@@DevOpsDirective It isn't much work and totally worth to do with the built in tools, instead of this script.
@mouryanaidu4753
2 жыл бұрын
hai sir im getting failed to get current pod for deployment error
@DanielPersson
2 жыл бұрын
Hi Mourya. Thank you for watching my videos. It's hard for me to help you without the actual error. Either you can take the error message you get when deploying and put in into a search engine and you usually get information from others that might have encountered the same thing or even some that have solved it. If that doesn't work then send the error message to me and I'll try go do some Googling in order to find some answer for you but there is still bugs out there that only one or a few have encountered and aren't yet solved so I can't garantee a solution for your problem. I hope this helps. Best regards Daniel
@mouryanaidu4753
2 жыл бұрын
actually the thing is we are running already rammit mq in the server for that we need to scalingg everything i just followed u but at the end in the logs im getting failer to get current podss
@mouryanaidu4753
2 жыл бұрын
my error queue messages are null required pods are 0 Failed to get Current pods numbers for my deployment
@DanielPersson
2 жыл бұрын
@@mouryanaidu4753 Hi Maurya That is strange indeed. If the server can download the kubectl during the build and the AUTOSCALING environment variable are configured correctly. Mainly you should look at the namespace and the application name. If those are correctly configured, then it should just work :) You could, of course, try to run this locally and replace the parameters to see if you have misconfigured it. kubectl -n $namespace describe deploy $deployment | grep desired | awk '{print $2}' | head -n1 I hope this helps. Best regards Daniel
@mouryanaidu4753
2 жыл бұрын
@@DanielPersson hi sir thanq after running through that command i got 1
@mzw8374
3 жыл бұрын
Helo, how can I do for adding the plugin like
@DanielPersson
3 жыл бұрын
Hi Z Actually this is not used as a plugin for the kubernetes framework. You just need to run the component as another docker container. I hope this helps. Thank you for watching my videos. Best regards Daniel
Пікірлер: 17