-
Notifications
You must be signed in to change notification settings - Fork 311
Deploy to Azure - site dies after 24-48 hours #542
Comments
@bobwiller Yeah ... and what a night to have this happen, right??? I'm sure everyone is out partying their rear-ends off. There is no reported Azure issue with that exception AFAIK. It's probably a config problem with your app. The critical bit is ...
The runtime is specified in How are you publishing your project prior to deployment to Azure? Do you have a |
That error looks like Helios is left over (AspNet.Loader.dll) in the bin folder of your deployed application |
Oh, that's a new one! |
Just had the same issue about an hour ago - @guardrex - Here is my global.json although it does not appear to get deployed to Azure. { When I publish, I am simply right-clicking the project, selecting "Publish..." and doing a deployment to my Azure Web app. On the Settings tab, my Configuration is set to "Release - Any CPU" and Target DNX Version is set to "Default (dnx-clr-win-x86.1.0.0.0-rc1-update1)" if I use the Kudu cmd window, I get 'dnvm' is not recognized as an internal or external command. @davidfowl: I was hoping it was going to be that easy, but AspNet.Loader.dll does not appear to be present (although this is a project I upgraded from beta 5, and had the same issues, although not as frequently) Here is my StartUp class if this helps:
} Thanks for the replies guys, I appreciate the help so far. |
@bobwiller Is your delete all files checked in the publish profile? If it isn't, that could lead to the |
Yep! |
@bobwiller I am going to assume that this is resolved once Helios is removed from the server. Closing. Re-open it or create a new issue if you still need help. |
Sorry @glennc I was traveling last week and wasn't able to respond. This is still causing me headaches: I went through and deleted all the files under / via FTP, and then published per @guardrex (checking remove additional files and compile files into NuGet packages), and I still go 48-72 hours until the site stops working. Only now, I also no longer get any errors in the event viewer. The other thing that is different now is that when I try to publish I get the following error: "An error was encountered when processing operation 'Delete Directory' on 'D:\home\site\wwwroot\App_Data\Jobs'." So I need to ftp and delete that Jobs folder before republishing. I just turned on some additional logging in the Azure portal hoping to catch what is going on, but this continues to be a problem (at least for me). Sorry for the delay in responding. Let me know if there is anything else I should look for, or can do from a troubleshooting perspective. |
@bobwiller There is a report of the Azure logging causing issues (but only one report of that, so it might be nothing): #364 (comment) Is this a free tier app? There is a report of running out of space on the free tier: aspnet/dnx#3192 |
Sorry if this isn't the right place to post this, but I am pulling my hair out: I have a simple Angular app built using ASP.NET 5 rc1 (not using MVC) and then using Visual Studio 2015 to deploy to an Azure Web App.
The site works fine for 24-48 hours and at some point it fails to load/render properly. I sense it is missing something to start or restart itself properly, but the deploy to azure docs are not complete.
My web.config looks like this:
and my project.json:
And here is the error:
What else is missing?? I am pulling my hair out with this and any help would be appreciated.
Bob
The text was updated successfully, but these errors were encountered: