🎨 Speeds up sending messages to user with socketio

[pcrespov]

What do these changes do?

This PR follows up from the incident reported by @matusdrobuliak66 and @GitHK on delays experienced consuming messages from socketio queue in rabbitMQ. @GitHK pointed out that the problem was in the delay caused by accessing redis everytime a socketio message was sent. @sanderegg proposed a solution that avoids this access by using instead the new rooms associated to user_id, i.e. rooms named f"user:{user_id}".

• ♻️ minor renaming and cleanup
• ♻️ Reactivated test on socketio

Related issue/s

• Follows up from incident 48

How to test

• Manually tested by checking that messages to the user were still delivered in different tabs and browsers (red arrows)
  
  

Dev Checklist

• No ENV changes or I properly updated ENV (read the instruction)

DevOps Checklist

[codecov]

Codecov Report

Attention: 2 lines in your changes are missing coverage. Please review.

Comparison is base (3c2ef61) 87.3% compared to head (522c9e4) 76.8%.

Additional details and impacted files

@@            Coverage Diff            @@
##           master   #5331      +/-   ##
=========================================
- Coverage    87.3%   76.8%   -10.5%     
=========================================
  Files        1321     542     -779     
  Lines       54151   27180   -26971     
  Branches     1172     202     -970     
=========================================
- Hits        47282   20883   -26399     
+ Misses       6619    6247     -372     
+ Partials      250      50     -200     

Flag	Coverage Δ
integrationtests	60.6% <72.2%> (-4.4%)	⬇️
unittests	86.5% <72.2%> (+1.4%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files	Coverage Δ
...rc/simcore_service_webserver/payments/_socketio.py	100.0% <100.0%> (ø)
...simcore_service_webserver/projects/projects_api.py	84.3% <100.0%> (ø)
...rc/simcore_service_webserver/socketio/_handlers.py	86.0% <100.0%> (ø)
...src/simcore_service_webserver/socketio/messages.py	100.0% <100.0%> (ø)
...tifications/_rabbitmq_exclusive_queue_consumers.py	88.7% <66.6%> (ø)

... and 917 files with indirect coverage changes

[GitHK]

Very nice.
Pleas make sure that if you start locally a service who's image is not present you are able to see the progress bar. After this is merged I will have a look at it's performance.

[matusdrobuliak66]

Thank you! nice solution.

[pcrespov]

Thank you! nice solution.

it's @sanderegg credit ;-)

[sanderegg]

thanks a lot for this!

[pcrespov]

Very nice. Pleas make sure that if you start locally a service who's image is not present you are able to see the progress bar. After this is merged I will have a look at it's performance.

@GitHK there is not test for that feature? I am not sure I can reproduce it. I propose to test in master and also create an automatic for it. Socketio plugin is really undertested! I can help with that.

[GitHK]

Very nice. Pleas make sure that if you start locally a service who's image is not present you are able to see the progress bar. After this is merged I will have a look at it's performance.

@GitHK there is not test for that feature? I am not sure I can reproduce it. I propose to test in master and also create an automatic for it. Socketio plugin is really undertested! I can help with that.

I will test it on was-master when you have it merged since I need to see the images being pulled. That's what causes a lot of messages to be delivered and the socktio delivery of messages to be slow.

[bisgaard-itis]

Looks good to me. Thanks

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues

Measures
0 Security Hotspots
No data about Coverage
0.0% Duplication on New Code

See analysis details on SonarCloud