JDK-6363961 : PIT b65: many regression tests failed with linux-amd64 build on 2.6 kernel linux machine
  • Type: Bug
  • Component: core-libs
  • Sub-Component: java.nio
  • Affected Version: 6
  • Priority: P2
  • Status: Closed
  • Resolution: Duplicate
  • OS: linux_2.6
  • CPU: generic
  • Submitted: 2005-12-15
  • Updated: 2010-04-02
  • Resolved: 2005-12-15
The Version table provides details related to the release that this issue/RFE will be addressed.

Unresolved : Release in which this issue/RFE will be addressed.
Resolved: Release in which this issue/RFE has been resolved.
Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.

To download the current JDK release, click here.
JDK 6
6Resolved
Related Reports
Duplicate :  
Description
System config.
Mustang PIT b65, Regression tests on linux-amd64 build on "amd-susew" and "sqe-iteam-7"
(both systems are linux 2.6 kernel)

Symptom:
There are around 32 j2se regression tests failed on linux 2.6 kernel system (amd-susew) with linux-amd64 build.
Looking at more than 10 of the test failures, they are all failing with:
"Server exception: java.io.IOException: Bad file descriptor"
This message is coming from the HttpServer in 
/test/sun/net/www/httptest

QE has tested sqe functional tests JHttp related by using the following build:
   /net/mho.sfbay/u/pit/b65/linux-amd64/346c/j2sdk1.6.0
         sqe-iteam-7 j2sdk1.6.0/bin> ./java -version
        java version "1.6.0-auto"
       Java(TM) 2 Runtime Environment, Standard
Edition (build 1.6.0-auto-346c)
     Java HotSpot(TM) 64-Bit Server VM (build
1.6.0-rc-b63, mixed mode)
And all tests are passing.

But many tests have failed when the following PIT build used:
/.automount/mho.sfbay/root/u/pit/b65/linux-amd64/jdk1.6.0/bin
sqe-iteam-7 jdk1.6.0/bin> ./java -version
java version "1.6.0-auto"
Java(TM) 2 Runtime Environment, Standard Edition
(build 1.6.0-auto-347)
Java HotSpot(TM) 64-Bit Server VM (build
1.6.0-rc-b63, mixed mode)

Comments from Dev.:
"FWIW, the issue arose on the 2.6 kernel and happened with 32-bit too. I 
assume the linux-i586 testing was on a 2.4 kernel so the issue wasn't 
noticed (as always we just have too many configurations to test :-)
-Alan"

A respin is on the way.

Comments
EVALUATION 5100121 was forward-ported to Mustang from 1.4.2_xx. Unfortunately the changes broke the epoll-based Selector - the issue was not observed in pre-putback testing because testing was done on a Linux 2.4 kernel. Anyway, once the issue was reported during PIT, we corrected the issue in conjunction with the gatekeeper. The correction was putback as an update to 5100121 and new Linux (32-bit and 64-bit) binaries were created. As the issue has been resolved I will close this bug.
15-12-2005