- ABP Framework version: v8.2.3
- UI Type: Blazor Server
- Database System: EF Core (SQL Server)
- Tiered (for MVC) or Auth Server Separated (for Angular): yes
- Exception message and full stack trace:
- Steps to reproduce the issue:
Hi,
This just started to happen, when I log in to our production (test) environment everything is OK for a few seconds, then I seem to be losing authentication, and getting an 'Unauthorized' exception when calling a service method on the homepage. I have a video of this here:
https://drive.google.com/file/d/1xJpaKd_RJFRgb90Ol0IxINk0oW9c6n6Y/view?usp=drive_link
I have detailed errors on so I can see the exception in the browser console:
Do you know why / how this is happening and how to fix?
Thanks,
Karim Ainsworth
9 Answer(s)
-
0
Here are the full debug logs for all 3 appservices:
https://drive.google.com/file/d/1tAimv5aRdwYhnA0LvwKTcaGj9_WoxdUS/view?usp=drive_link
-
0
Hi,
I'm requesting the access
-
0
Hi, I have given you access.
-
0
Hi,
i guess this is related to Redis configuration.
are they use the same redis server and same cache key prefix?
-
0
Hi,
Ah OK, that might make sense. Let me make sure I have Redis explicitly set up in the EnvVars. Just in appsettings now, as so:
"Redis": { "Configuration": "hathor.redis.cache.windows.net:6380,password=XXXXXXXXXXXXXXXXXXXXXXXXXXXX=,ssl=True,abortConnect=False" },I have 2 completely separate environments now, not using Deployment Slots. One for Test, one for Production, running the exact same codebase. Using the same Redis connection string for both environments, all AppServices.
Is that OK, would you suggest a different setup / configuration? I plan now to move the Redis config into the EnvVars in Azure to have a bit better control. I will let you know how that works.
Any other suggestions?
Thanks,
Karim
-
0
Hi,
All projects should use the same Redis config.(connection string, key prefix)
Configure<AbpDistributedCacheOptions>(options => { options.KeyPrefix = ... }); -
0
Hi,
OK I have confirmed that all KeyPrefixes are the same. Just to make sure, the KeyPrefix can be anything, correct? I have it as "ESv2:" for all projects.
I also copied the Redis config in to an Environment Variable and used the same value for all AppServices:
However I am still getting the exact same error:
any other ideas?
Thanks,
Karim
-
0
Just to make sure, the KeyPrefix can be anything,
yes.
However, I am still getting the exact same error:
i think this is a problem, is the problem can reproduce in local?
-
0
Hi,
In local it works fine, in our identical Production environment ( https:hathor.events ) it also works fine, with the exact same codebase, both were deployed at the same time. The configuration also looks the same to me, I have checked and double checked.
Here is a video of the production deploy:
https://1drv.ms/v/s!AkJmGuHQuob7kqUmF4Iietrj4J1G6A?e=vbVyIK
And now, when I go to https://test-dashboard.hathor.events I get a 400 error before the login UI. This has also been a huge issue for me when deploying to production, seems there may be intermittent issues connecting to the database. I have another ticket open for that.
I just need some help getting to the bottom of these issues when deploying. We have spent so much time on this project, and everything works fine locally but it is so unstable after deploying. I need to demo to our client and I can't do so, this is becoming critical for us.
Seems like everything is revolving around the Auth project in one way or another.
Here are the new logs, same exact deployment, now with 400:
https://1drv.ms/u/s!AkJmGuHQuob7kqUl46X837Ty84Ga_Q?e=zyRTum
[ERR] An error occurred using the connection to database 'ESv2-testing' on server 'tcp:hathor-hk.database.windows.net,1433'
I really need help getting things stable in production, any ideas or insight would be very much appreciated.
Thanks,
Karim
