<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<br>
<div class="moz-cite-prefix">On 12/13/13 03:33 PM, Alan Bateman
wrote:<br>
</div>
<blockquote cite="mid:52AB28B7.6090505@oracle.com" type="cite">
<meta http-equiv="Context-Type" content="text/html;
charset=ISO-8859-1">
On 13/12/2013 14:35, Balchandra Vaidya wrote:
<blockquote cite="mid:52AB1B36.7030008@oracle.com" type="cite"> <br>
All,<br>
<br>
I request to review a change for "8030035 - Create a stable<br>
test group in TEST.groups"<br>
<br>
<a moz-do-not-send="true" class="moz-txt-link-freetext"
href="http://cr.openjdk.java.net/%7Ebvaidya/8/8030035/webrev.01/">http://cr.openjdk.java.net/~bvaidya/8/8030035/webrev.01/</a><br>
<a moz-do-not-send="true" class="moz-txt-link-freetext"
href="https://bugs.openjdk.java.net/browse/JDK-8030035">https://bugs.openjdk.java.net/browse/JDK-8030035</a><br>
</blockquote>
<br>
I'm not sure about including javax/print in this list. They might
be stable but they only pass if there is a printer attached. I see
you have them listed as "known issues" [1] but I think it really
reduces the chances that all the tests in the proposed group will
pass.<br>
</blockquote>
<br>
In the java/print group, 11 tests pass and 5 tests fail when printer
is<br>
not attached. I can remove it from the jdk_stable group.<br>
<br>
<blockquote cite="mid:52AB28B7.6090505@oracle.com" type="cite"> <br>
I'm also not sure about the SCTP test. We put them into their own
group because they are highly sensitive to the kernel
configuration and/or the libsctp that is installed. They do pass
if SCTP is not in the kernel but they have been troublesome on
older kernels that had it configured. I guess I can't really
object to including this and we can always remove it if there are
complaints.<br>
</blockquote>
<br>
I didn't see any issue with these tests on OEL 6. I am happy to
remove <br>
it from the jdk_stable group if anyone complain on any other OSes.<br>
<br>
<blockquote cite="mid:52AB28B7.6090505@oracle.com" type="cite"> <br>
As regards the rest then there are few tests in these areas that
fail when running headless. I'm interested in hearing other views
on that topic. Volker's suggestion to have a stable and
stable-headful is one approach. I guess I just object to tests
doing strange polkas on my desktop. Also it is very appealing in
server environments so not have to setup a virtual frame buffer.<br>
</blockquote>
<br>
I found 13 tests from jdk_stable group that fail in headless mode.
In the short term,<br>
I can add a note in knownissues.html [1] page. In the long term,
hoping to have a<br>
solution for those failing tests. <br>
<br>
<br>
Thanks<br>
Balchandra<br>
<br>
<br>
<blockquote cite="mid:52AB28B7.6090505@oracle.com" type="cite"> <br>
-Alan.<br>
<br>
[1] <a moz-do-not-send="true" class="moz-txt-link-freetext"
href="http://download.java.net/jdk8/testresults/docs/knownissues.html">http://download.java.net/jdk8/testresults/docs/knownissues.html</a><br>
</blockquote>
<br>
</body>
</html>