As the new autoscaling functionality became available in windows azure a few weeks ago i enabled this on my service and everything was great.
Then I deleted the deployment and deployed it again. And now I get an error saying that "Autoscale failed for [role name]" and I can only manually scale it.
I also tried deleting the service alltogether and recreate it from cratch with no improvments. This is the only service that I have this problem with and if I deploy the same solution to Another service it works.
Does anyone know how to get around this?
Related
I have a SpringBoot application deployed in PCF scaled to 12 instances. One or two instances are going down. I want to restart those 2 instances instead of restarting the application automatically. How do you restart a single instance automatically in PCF when it is going down or when it crashed?
I am using the following command to restart a single instance manually:
cf restart-app-instance APP_NAME INDEX
There is nothing you need to do. The Cloud Foundry platform will monitor your application via the configured health check and automatically restart it.
If that's not happening, you would want to look into why it's not being restarted automatically and not try to hack together some other way of restarting it.
Can you try something.
Delete the instance at the index:
cf curl DELETE /v2/apps/$(cf app APP_NAME --guid)/instances/INDEX_OF_THE_INSTANCE
Reference: http://apidocs.cloudfoundry.org/272/apps/terminate_the_running_app_instance_at_the_given_index.html
Wait for sometime, so that CF can recreate the instance.
I have not tried this in my lab, and if I get any update I will post it here in comments.
Thank,
Chandan Patra
I am not able to update AMI on my launch configuration. Here is the error I see
and it rolls back. I checked the reason on google and it should work fine even if there are running instances, just that it won't update the already running instances. But, it does not work like that in our case. Please suggest!
Have an issue when we try to deploy webjob to a web app via Visual Studio.
If we try to set 10 minutes interval it returns 409 response
http://grab.by/Rirs
If we try with one hour interval it is successful.
We have Standard app service plan, so it should be supported.
http://grab.by/Rirw
Always on is activated for web app.
We have also tried another approach that described here
There is Publish Succeeded. but if we look on webjob in VS Cloud Explorer it has Fail status
Any ideas how can we solve it?
After some researching, found the issue, but actually in old Azure portal.
It was necessary to set standard tier for the job collections that was created for jobs under web-apps.
New portal doesn't have simple way to get that (or I didn't find it as simple like I done that with old portal) and again the super-fine thing it's possibility to configure job schedule on the old portal, while it's not possible on new one... or again I didn't find that
I am currenlty using Web Deploy 3 with Automatic Backup. Yesterday, backup failed with error message 'ERROR_PACKAGE_TOO_LARGE'.
as per MSDN
Resolution - Use the archiveDir provider when creating a package instead. Currently there is no solution for this limit with respect to automatic backups.
Now archiveDir works fine on command line
msdeploy -verb:sync -source:apphostconfig="Site" -dest:archivedir=c:\archive
I am pretty new to Web Deploy, can someone please help me to automate this with Web Deploy using VS2013? I don't want to login to server to manually back up my site and then come back to VS to publish. Is there any setting in .pubxml(Publish Profile) to achieve it?
Thanks
I have been trying hard to deploy a java application through cloudfoundry into an Amazon EC2 instance using the console as described in this screencast video on cloud foundry.
http://classic.cloudfoundry.com/screencasts.html
but only failing to get the console where I can deploy by choosing a war from the local file system, as explained in the video..
What is the correct URL where I can find a console similar to the one shown in the video ??
And I have already tried signing up for the url http://classis.cloudfoundry.com, but not getting an approval or acknowledgement mail from the team even after 4 days..
Did anyone face similar issue in the recent times?
If yes please help!
The Cloud Foundry Classic service (at http://classic.cloudfoundry.com) is no longer accepting new accounts.