Doh!
It looks like what happened was I tried to start a build without using
`nohup`. So, at some point something is trying to do output that is
not being redirected by the nightly build script (which is supposed to
redirect all output to the log), and that was causing the process to
get a TTOU.
I'm going to run it in the foreground and see if I can figure out what
the output is.
Bleah.
On 2009-05-04, at 10:07EDT, Henry Minsky wrote:
I don't know what would have changed, but I thought there was a
'headless'
flag to run Java with that we use
-Djava.awt.headless=true
On Mon, May 4, 2009 at 9:42 AM, P T Withington <[email protected]> wrote:
On 2009-05-04, at 09:06EDT, P T Withington wrote:
Here:
update-version:
[copy] Copying 4 files to
/home/svnbuild/builds/13798-openlaszlo-branches-4.2
laszlo-explorer:
[copy] Copying 1 file to
/home/svnbuild/builds/13798-openlaszlo-branches-4.2/laszlo-
explorer/coverpages/welcome
[style] Processing
/home/svnbuild/builds/13798-openlaszlo-branches-4.2/laszlo-
explorer/nav.xml
to /home/svnbuild/builds/13798-o\
penlaszlo-branches-4.2/index.html
[style] Loading stylesheet
/home/svnbuild/builds/13798-openlaszlo-branches-4.2/quick-
index.xslt
init:
[mkdir] Created dir:
/home/svnbuild/builds/13798-openlaszlo-branches-4.2/WEB-INF/lps/
schema/build
rng:
ps shows:
svnbuild 16346 0.2 9.5 480608 99960 pts/0 Tl 03:45 0:20
/usr/lib/jvm/sun-jdk-1.5/bin/java -classpath /home/svnbuild/lib/ap\
ache-ant-1.6.5/lib/ant-launcher.jar
-Dant.home=/home/svnbuild/lib/apache-ant-1.6.5
-Dant.library.dir=/home/svnbuild/lib/apache-ant-\
1.6.5/lib org.apache.tools.ant.launch.Launcher -cp .
-Dbuild.revision=13798 -Dbuild.branch=branches/4.2 -
Dbuild.version=4.2.0.3
-Db\
uild.name=13798-openlaszlo-branches-4.2 -Dbuild.id=branches/
4...@13778(13798) -Dtomcat.contextpath=/lps-4.2-nightly -
Dbuild.target=\
dist-one init do-linux-build
svnbuild 16371 0.0 0.0 272 16 pts/0 T 03:49 0:00
/opt/sun-jdk-1.5.0.13/jre/bin/java -jar /home/svnbuild/lib/apache-\
ant-1.6.5/lib/trang.jar lzx.rnc build/lzx.direct.rng
'T' status indicates the process is 'stopped' by job control?
Ideas requested.
I have an idea: Something is trying to do some @#!-ing Java
operation that
interacts with the display, so it is getting sent SIGTTOU. Bleah.
What has
changed that is causing this all of a sudden?
--
Henry Minsky
Software Architect
[email protected]
