I can explain the oddity about the TeamCity output. Basically when you run tests in parallel, Gallio has to do a bit of magic to serialize the results for TeamCity in a manner that preserves the overall structure of the tests. If we just submitted test results to TeamCity in some arbitrarily interleaved order then TeamCity would be unable to reconstruct the right test structure and you might see some very strange arrangements of test suites, etc. in the TeamCity reports. As a result of this "deinterleaving" process, Gallio may buffer up test results for quite a long time until it reaches a suitable place in the lifecycle where it can stream them out. When it does so you might see 1000 test results appear all in one go. How does the total execution time on TeamCity without NCover compare with TDNet test runs on your development workstation? Are the development and build machines all of similar class? Are they under similar load? Does the build machine have enough RAM? Jeff.
_____ From: [email protected] [mailto:[email protected]] On Behalf Of Bruno Wouters Sent: Tuesday, September 08, 2009 12:24 AM To: [email protected] Subject: MbUnit RE: Known issues with ncover runner and Parallelizable attribute? Hi all, It does not seem to be a problem with NCover. If I run it without ncover it runs just as slow. After 10minutes of testing, only 69 tests are shown as passed in teamcity but if I look in the log I see a lot of lines like this: [08:50:27]: [gallio] [starting] MbUnit v3.0.6.787/BULTest/CachingTest/DoesPersonHaveItemPriv_NotLoaded_ResultIsNull [08:50:27]: [gallio] [passed] Test MbUnit v3.0.6.787/BULTest/CachingTest/DoesPersonHaveItemPriv_NotLoaded_ResultIsNull Certainly more than 69 [passed] lines, probably more than 600. If I run it from visual studio all tests (#3000) pass in max 11minutes and on teamcity it takes more than 30minutes. Thanks! Greets, Bruno From: Bruno Wouters [mailto:[email protected]] Sent: maandag 7 september 2009 22:54 To: '[email protected]' Subject: Known issues with ncover runner and Parallelizable attribute? Hi all, Are there any known issues with ncover runner and the Parallelizable attribute? Running the tests locally without ncover (just in visual studio with testdriven.net) works great. In teamcity with ncover it seems to be extremely slow. Any ideas? Thanks! Greets, Bruno --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "MbUnit.User" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/mbunituser?hl=en -~----------~----~----~----~------~----~------~--~---
