Re: RFR: 8274453: (sctp) com/sun/nio/sctp/SctpChannel/CloseDescriptors.java test should be resilient to lsof warnings
On Tue, 28 Sep 2021 15:54:09 GMT, Aleksey Shipilev wrote: > Recently added test fails on my desktop in the course of newly added `tier4` > runs. The reason is simple: the test verifies that lsof outputs no more than > LIMIT_LINES (2) lines. And on my machine, lsof prints some warnings that blow > that limit. In the patch, lsof -w is used to suppress the warnings. > > Additional testing: > - [x] Affected test now passes All right, I am integrating soon then. - PR: https://git.openjdk.java.net/jdk/pull/5739
Re: RFR: 8274453: (sctp) com/sun/nio/sctp/SctpChannel/CloseDescriptors.java test should be resilient to lsof warnings
On Wed, 29 Sep 2021 15:05:21 GMT, Aleksey Shipilev wrote: > Ok, thanks. @AlanBateman, want to ack? I don't have anything to add except to say that close and async close implementations in the SCTP classes needs to be re-written and brought into line with the implementation in NIO. Also I'm not a fan of using losf in tests because the file descriptor count can increase/decrease for many reasons. - PR: https://git.openjdk.java.net/jdk/pull/5739
Re: RFR: 8274453: (sctp) com/sun/nio/sctp/SctpChannel/CloseDescriptors.java test should be resilient to lsof warnings
On Tue, 28 Sep 2021 15:54:09 GMT, Aleksey Shipilev wrote: > Recently added test fails on my desktop in the course of newly added `tier4` > runs. The reason is simple: the test verifies that lsof outputs no more than > LIMIT_LINES (2) lines. And on my machine, lsof prints some warnings that blow > that limit. In the patch, lsof -w is used to suppress the warnings. > > Additional testing: > - [x] Affected test now passes Ok, thanks. @AlanBateman, want to ack? - PR: https://git.openjdk.java.net/jdk/pull/5739
Re: RFR: 8274453: (sctp) com/sun/nio/sctp/SctpChannel/CloseDescriptors.java test should be resilient to lsof warnings
On Tue, 28 Sep 2021 15:54:09 GMT, Aleksey Shipilev wrote: > Recently added test fails on my desktop in the course of newly added `tier4` > runs. The reason is simple: the test verifies that lsof outputs no more than > LIMIT_LINES (2) lines. And on my machine, lsof prints some warnings that blow > that limit. In the patch, lsof -w is used to suppress the warnings. > > Additional testing: > - [x] Affected test now passes Marked as reviewed by dfuchs (Reviewer). I saw the test failed once on a linux ARM 64 machine - but it didn't fail again after 50 more runs. I also believe that the failure has nothing to do with your changes, so from my perspective your good to go. In case you're interested here is the tail of the log file that shows the failure. I suspect this a one off where the OS hadn't reclaimed the file descriptor within the expected delay (maybe due to other activity occurring on the machine). ... 99 0 1 2 3 4 5 6 7 8 9 end COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME java2958962 mach5-one5u unix 0x 0t0 217247452 type=STREAM java2958962 mach5-one8u unix 0x 0t0 217252225 type=STREAM --System.err:(13/835)-- java.lang.RuntimeException: Failed: detected unclosed FD. at CloseDescriptors.main(CloseDescriptors.java:92) at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:77) at java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.base/java.lang.reflect.Method.invoke(Method.java:568) at com.sun.javatest.regtest.agent.MainWrapper$MainThread.run(MainWrapper.java:127) at java.base/java.lang.Thread.run(Thread.java:833) JavaTest Message: Test threw exception: java.lang.RuntimeException: Failed: detected unclosed FD. JavaTest Message: shutting down test - PR: https://git.openjdk.java.net/jdk/pull/5739
Re: RFR: 8274453: (sctp) com/sun/nio/sctp/SctpChannel/CloseDescriptors.java test should be resilient to lsof warnings
On Tue, 28 Sep 2021 15:54:09 GMT, Aleksey Shipilev wrote: > Recently added test fails on my desktop in the course of newly added `tier4` > runs. The reason is simple: the test verifies that lsof outputs no more than > LIMIT_LINES (2) lines. And on my machine, lsof prints some warnings that blow > that limit. In the patch, lsof -w is used to suppress the warnings. > > Additional testing: > - [x] Affected test now passes @shipilev I will send your patch to our CI and come back with results. - PR: https://git.openjdk.java.net/jdk/pull/5739