Actually, We have 3 (sic!) versions of spring data (1x, 2.0 and 2.2) and all of them are correct.
But any versions of spring-data is not needed for platforms test. чт, 25 июн. 2020 г. в 16:50, Pavel Tupitsyn <ptupit...@apache.org>: > Ivan, thank you so much for getting to the bottom of this issue! > > Why do we have two versions of spring-data, and which one is correct? > > On Thu, Jun 25, 2020 at 4:34 PM Ivan Daschinsky <ivanda...@gmail.com> > wrote: > > > The main reason of failure is jar-hell. When .NET or C++ tests are > started, > > if IGNITE_NATIVE_TEST_CLASSPATH is set to true, source directory is > > iterated and files libs, target/classes etc. > > are added to classpath. But neither readdir(), FindNextFileA() > > or Directory.EnumerateDirectories() do guarantee any ordering. > > But in spring-data-2.0 and spring-data-2.2 there are different version of > > spring. So jar hell occurs and tests fails. > > > > I created small fix (ignore module dir that contains spring-data), and > > tests start working on aitc-lin11_02. > > > > Ticket for this problem is [1] > > > > [1] -- https://issues.apache.org/jira/browse/IGNITE-13187 > > > > > > > > чт, 25 июн. 2020 г. в 11:46, Ivan Daschinsky <ivanda...@gmail.com>: > > > > > Hi igniters! > > > > > > I found that .NET Core suite and C++ suites fails on two agents -- > > > aitc-lin01_04 and aitc-lin11_02. > > > > > > So I think the best solution is to exclude these agents as requirement > > for > > > these suites and debug problems on these > > > agents separately > > > > > > вт, 23 июн. 2020 г. в 09:33, Ivan Daschinsky <ivanda...@gmail.com>: > > > > > >> Hi folks! > > >> > > >> It seems, that this test mysteriously fails only on one > > >> agent aitc-lin01_04 with this error. [1] > > >> Locally this .NET Core tests doesn't fail for about 20 minutes of > > >> continuously run (Mac OS X 10.15, .NET Core 3.1.200) > > >> > > >> According to stacktrace, it seems that there is problem with build > > >> configuration and invalid jars are in classpath. > > >> > > >> Petr Ivanov, may be it is good idea to clean up aitc-lin01_04? > > >> > > >> -- > > >> Sincerely yours, Ivan Daschinskiy > > >> > > > > > > > > > -- > > > Sincerely yours, Ivan Daschinskiy > > > > > > > > > -- > > Sincerely yours, Ivan Daschinskiy > > > -- Sincerely yours, Ivan Daschinskiy