-
Notifications
You must be signed in to change notification settings - Fork 291
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
AI (Internal): Operation to stop does not match the current operation #693
Comments
Same problem! |
Could you please help us find the root cause of the issue
@Ruud-cb Thanks a lot for the info! |
@lmolkova Here are my responses:
timestamp [UTC]: 2018-02-27T19:02:08.058Z Please let me know if there is any info I can gather. I will admit my mind is not fresh on this as I've not revisited it until today. |
Thanks for the info. Now we can narrow down the problem to the HTTP dependency collection.
Make sure these are HTTP dependencies. We can dig more into this issue as well, but please open another issue for this.
In analytics, you may specify any timestamp in the past. Are you able to find any Regarding the new trace you observe |
I am no longer able to see any taces with "AI (Internal): Operation to stop does not match the current operation.". I queried a subset of that string and find nothing, which likely is a result of data retention policies of AppInsights as you suggest. So I'm not sure where to go with that. Seems kindof odd since I'm using the same version of the appinsights sdk. I'm glad to hear the "Current Activity is null" issue will be fixed. I'm not sure why they suddenly started showing up. It feels like something changed out from under me. Any ideas why the change in behavior? |
This error does not happen frequently. It could be some very specific case in the HttpClient diagnostics that causes it. E.g. some network issue that is not handled in diagnostics pipeline. So you may start observing it again at some point. If you do, please come back :) Let's hope that @Ruud-cb or someone else experiencing this issue would be able to find telemetry following this event.
For this issue, we know just one reason to cause it so far: some temp network issues transmitting telemetry to AppInsights endpoint. Unfortunately, this event does not give us enough info (until 2.6), so we cannot tell if it happens in any other cases. I do not think it could be caused by your code. If your service and the rest of Azure services you use are in different data centers, network glitches would be expected at a reasonable rate. If you see a huge spike in these events and/or unexpected decrease in number of http dependency telemetry items tracked - this would indicate an issue. |
Hi, Just for the record, it seems to occur on local computers only, not when the app is in production. 1 It seems to happen before some front-end requests. NPM related files that get loaded. Another example: Not sure if it helps but there is only 1 I noticed in production that occured before a http timeout. Another error though:
Also running NET Core. |
Thanks for the info, it really helps. It's interesting that localhost requests have this problem. I also see that your dependencies are reported twice. I'm pretty sure you are hitting microsoft/ApplicationInsights-dotnet-server#723. It was fixed in ApplicationInsight AspNetCore SDK 2.2-beta2. could you please confirm that you are indeed, have Microsoft.ApplicationInsights.AspNetCore 2.2.0-beta1 and Microsoft.ApplicationInsights.DependencyCollector 2.5.0-beta1? Regarding production issue you observe, it's microsoft/ApplicationInsights-dotnet-server#780. We target to fix it in the upcoming stable release.
you should have both: request and dependency telemetry when you call your backed instrumentation with AI. This is expected behavior. More on this
can you elaborate? where you are making these requests from? Are these Http apis? |
Should be fixed with microsoft/ApplicationInsights-dotnet-server#848 in 2.6.0-beta3 Microsoft.ApplicationInsights (base SDK) and 2.3.0-beta1 Microsoft.ApplicationInsights.AspNetCore. Please reopen if reproduces |
I have the same problem. I use Microsoft.ApplicationInsights.AspNetCore Version="2.5.1" and have two TelemetryProcessor and one TelemetryInitializer. Could the TelemetryProcessor or the TelemetryInitializer be a problem? We don't use a TelemteryClient. |
I've got a simple DotNet Core 2.0 app using AppInsights. I've started using the AppInsights SDK as well and the TrackTrace() method in particular. It's working fine, but queries on the traces table in Analyitics are showing a number of traces with a message of "AI (Internal): Operation to stop does not match the current operation". I'm not generating these and they are polluting my data. Any idea why they are generated and how I can make them stop? I'm using SDK v2.4.0-32153
Here are the trace details (with some values obscured):
timestamp [UTC]: 2018-01-10T19:39:20.651Z
message: AI (Internal): Operation to stop does not match the current operation.
operation_SyntheticSource: SDKTelemetry
client_Type: PC
client_IP: my-ip-address
client_CountryOrRegion: United States
appId: 43dda5df-4e1c-46a9-82ee-977811e262b5
appName: debug-fossil
iKey: my-ikey-value
sdkVersion: dotnet:2.4.0-32153
itemId: f754898b-f63d-11e7-8081-3b3156d2a6a2
itemType: trace
itemCount: 1
The text was updated successfully, but these errors were encountered: