V9 Linux Changing Culture and Domain breaks Umbraco Frontent
Hi Guys,
I am testing Umbraco 9 as I would like to run it on Linux once Published to reduce some of my hosting cots. I have been testing out the V9 Beta and found that if you change the 'Culture' or a node ad add a second domain to it for example or the Frontend of the Web Site will return a 500 error. Unfortunately no of details are shown to help determine the cause of the issue.
To help give others an idea on my steps for what I have done so far;
Built and then Published the Application (no changed made to code, no custom code added)
Copied the published app to the '/var/www/umbraco9' folder on my Linux test server
Started the DOTNET Core application, I have then used Apache as a reverse proxy as described by Microsoft (https://docs.microsoft.com/en-us/aspnet/core/host-and-deploy/linux-apache?view=aspnetcore-5.0) please note this guide is for CentOS but I am using Ubuntu the steps are the same and work well
Create a basic doc type and master page, with <h2>hello world</h2> in it
Create a 'root folder' for the site one and then create a sub page called 'Home Page'. Change the Culture of this page to be the loopback address (required for Apache Reserve Proxy) and publish.
At this point if you try to go to the Web Page (Frontend) you will get a 500 error from the web server. If you then remove this page and then try and create it at the route of the 'Content' tree you will still get the same 500 error.
What is strange if you re-deploy the Umbraco9 app and use the same steps but create the page at the root of the 'Content' tree you get the Hello World, but then if you move the page to a sub section under a folder or another node item and change the culture again to make the page the start page of the web site you get the same issue again.
I am not sure if anyone else has run into this issue with Umbraco 9 or if I am just doing something wrong. I have had a look to see if I can find any logs that might help me determine what is worn but at t the moment I haven't found anything that has been of use.
Any information or advise on this would be appreciated but for the moment I have re-deployed the Test Web site and kept everything starting at the root of the content tree.
Maybe test with a culture of /au or /uk or whatever, leave out the http etc.
Depending on your hosts setup I find local urls like localhost to be very touchy.
Test by using the secondary site as a relative change first, that way you can narrow down where the issue is being caused.
When the 500 error occurred did it generate anything in your logs?
I ran into a somewhat similar issue. Had an Umbraco 8 website and upgraded it now to Umbraco 9. Locally everything works fine.
Now I deployed it to my Windows Server on IIS, bound it to a different domain (for testing) and pointed to DB Connection String to a clone of the original Umbraco 8 Db. When I run it, it prompts me to upgrade - which works fine. The backoffice shows all my content where it should be, all media, all users and members ... Everything looks good. The backoffice works wonderfully.
BUT: Opening the frontend only returns a 500. I don't get it.
No I never figured it out. I ended up reverting to default Cultures. It’s something I won’t to test again just haven’t had time.
In your web.config file are the Umbraco 9 files referenced correctly? I know on one of my tests it didn’t update a file path (this was in early v9 beta).
If you change you culture to local host and browser locally does it work?
Guys, JUST use the /au /eu /uk, try not to use different url hosts (yahoo.com, google.com), this way you don't 'change it to localhost', you just make sure the path is correct.
Does it still 500 on the FE? if so, what is the error in the log say?
I fixed it. My problem was a misconfiguration - probably it has nothing to do with the cultures and TPark's problem.
Ich had the models builder set to Source Code locally and did not set that on the server. So there it was set to InMemory. The namespaces didn't match, so it couldn't bind the models.
V9 Linux Changing Culture and Domain breaks Umbraco Frontent
Hi Guys, I am testing Umbraco 9 as I would like to run it on Linux once Published to reduce some of my hosting cots. I have been testing out the V9 Beta and found that if you change the 'Culture' or a node ad add a second domain to it for example or the Frontend of the Web Site will return a 500 error. Unfortunately no of details are shown to help determine the cause of the issue.
To help give others an idea on my steps for what I have done so far;
<h2>hello world</h2>in itAt this point if you try to go to the Web Page (Frontend) you will get a 500 error from the web server. If you then remove this page and then try and create it at the route of the 'Content' tree you will still get the same 500 error.
What is strange if you re-deploy the Umbraco9 app and use the same steps but create the page at the root of the 'Content' tree you get the Hello World, but then if you move the page to a sub section under a folder or another node item and change the culture again to make the page the start page of the web site you get the same issue again.
I am not sure if anyone else has run into this issue with Umbraco 9 or if I am just doing something wrong. I have had a look to see if I can find any logs that might help me determine what is worn but at t the moment I haven't found anything that has been of use.
Sample Content Tree - WebSite1 | - Home (Culture changed to http://localhost:5000) -WebSite2 | - Home (Culture changed to http://localhost2:5000)
Please note localhot2 is used as a example
Any information or advise on this would be appreciated but for the moment I have re-deployed the Test Web site and kept everything starting at the root of the content tree.
Maybe test with a culture of /au or /uk or whatever, leave out the http etc. Depending on your hosts setup I find local urls like localhost to be very touchy.
Test by using the secondary site as a relative change first, that way you can narrow down where the issue is being caused.
When the 500 error occurred did it generate anything in your logs?
Did you get it up and running eventually?
I ran into a somewhat similar issue. Had an Umbraco 8 website and upgraded it now to Umbraco 9. Locally everything works fine.
Now I deployed it to my Windows Server on IIS, bound it to a different domain (for testing) and pointed to DB Connection String to a clone of the original Umbraco 8 Db. When I run it, it prompts me to upgrade - which works fine. The backoffice shows all my content where it should be, all media, all users and members ... Everything looks good. The backoffice works wonderfully.
BUT: Opening the frontend only returns a 500. I don't get it.
Hi David,
No I never figured it out. I ended up reverting to default Cultures. It’s something I won’t to test again just haven’t had time.
In your web.config file are the Umbraco 9 files referenced correctly? I know on one of my tests it didn’t update a file path (this was in early v9 beta).
If you change you culture to local host and browser locally does it work?
Guys, JUST use the /au /eu /uk, try not to use different url hosts (yahoo.com, google.com), this way you don't 'change it to localhost', you just make sure the path is correct.
Does it still 500 on the FE? if so, what is the error in the log say?
I fixed it. My problem was a misconfiguration - probably it has nothing to do with the cultures and TPark's problem.
Ich had the models builder set to Source Code locally and did not set that on the server. So there it was set to InMemory. The namespaces didn't match, so it couldn't bind the models.
is working on a reply...