Pull Request details not available on SonarQube

SonarQube
21

Hi,

I don’t think I need to see that log. I’m happy to take your word for it.

So now can you check ce.log to see if you can see the analysis report being handled?

 
Ann

22

Hi,

No there is no mention of it in ce.log but another pr which was triggered almost a minute before 1156 does show up.

24

Hi,

Thanks for looking. I’ve pinged internally for more expert eyes.

 
Ann

25

Hello, Any updates with this? We are starting to face this more frequently, it has gone up from 1-2 PR’s per week to 1-2 PR’s everyday or more.

27

Hello,

When you run the scanner, you should see something like this in the log at the end:

INFO: ANALYSIS SUCCESSFUL, you can find the results at: https://domain.com/dashboard?id=my-project
INFO: Note that you will be able to access the updated dashboard once the server has processed the submitted analysis report
INFO: More about the report processing at https://domain.com/api/ce/task?id=AYSk11Esb6hcGbyEZurD

Can you try to access the second link, please? It should help us know if the report was actually received by SonarQube, and what is its current status in the queue.

28

Hello,
It looks something like this, the image is from a more recent pr with same issue.

image
image762×70 2.19 KB

29

I am adding few more details which I have observed in past few days:

  1. I have noticed that the pr’s which have these issues sometimes do show up on Sonarqube UI after multiple builds via jenkins(multiple scans).

  2. When they do show up after multiple tries, when I try to check the sonarqube UI for the background task it throws error at project level which looks like below image but it goes away once the background task for the PR in question is done:

    pr error
    pr error1206×137 7.27 KB

    I am also attaching an error that shows up on web.log at same time
    weblog.txt (11.5 KB)

  3. After reading some other posts on different forums some people continuously talk about waitforqualitygate() or adding a sleep, currently we dont use any of those in our Jenkins pipeline script. Is that something we should look at?

30

I don’t think this would help, as there is already an issue before the QG step:

This means that right after the analysis is submitted to SonarQube (and the scanner gets an acknowledgment and the task UUID as a response), the task does not exist on SQ anymore, which does not make sense.

Usually, when this kind of “logic” issue happens, It’s because there are several SonarQube instances running in parallel on the same database. Would it be possible that you have a stagging instance that receives by error production queries? Or two SonarQube instances connected to your production database?

31

Hi, Thanks for the reply. No we do not have multiple instances. Also the database being used is PostegreSQL and it’s hosted on same machine as well. And only thing using that database is SonarQube.

32

Hello, any update on this?