On Tue, May 10, 2011 at 12:58 AM, Gaojinchao <[email protected]> wrote: > Thanks for your reply. > > 1/ I have other question: It needs clear something when one test case had > failed ?
Are you suggesting that there is a ripple effect if a test fails in that lots of the subsequent tests fail? That should not be the case but I believe I've seen it happen. If a test does not cleanup properly, it may for example leave a zk ensemble running, one that has some state in it and it then messes up the subsequent test. > In my case. The test cases auto run at night. > I find some test case failed randomly and try to dig it. > > > Tests run: 1, Failures: 1, Errors: 0, Skipped: 0, Time elapsed: 7.599 sec <<< > FAILURE! > testMergeTool(org.apache.hadoop.hbase.util.TestMergeTool) Time elapsed: > 7.586 sec <<< FAILURE! So, TestMergeTool always fails or sometimes? Can you look at the log to see why the assertion failed? Is there an exception earlier up in the log? > 2/ DNS Error test case: > TestClockSkewDetection > testBadOriginalRootLocation > testScanner > TestCatalogTracker > > Error logs: > Tests run: 1, Failures: 0, Errors: 1, Skipped: 0, Time elapsed: 0.031 sec <<< > FAILURE! > testClockSkewDetection(org.apache.hadoop.hbase.master.TestClockSkewDetection) > Time elapsed: 0.02 sec <<< ERROR! > java.lang.IllegalArgumentException: Could not resolve the DNS name of > example.org:1234 Lets figure a way to fix these. Its not good that our test depends on the network and a working DNS. Please file an issue with this data. I don't think it'll be fixed for 0.90.x but we can address it in 0.92.0 (in 0.90.x, the deserialization of HServerAddress creates an InetSocketAddress which does a resolve on construction). St.Ack
