--- Begin Message ---
Hi Andi!
In that case you will be interested to know that almost all functional
tests pass as well. However, TestSwitchTimezone fails like this:
#
# An unexpected error has been detected by Java Runtime Environment:
#
# SIGSEGV (0xb) at pc=0x000000000041a224, pid=8833, tid=47284893239008
#
# Java VM: OpenJDK 64-Bit Server VM (12.0-b01 mixed mode linux-amd64)
# Problematic frame:
# C [python+0x1a224] PySequence_Contains+0x4
#
# An error report file with more information is saved as:
# /home/jonas/chandler/hs_err_pid8833.log
#
# If you would like to submit a bug report, please visit:
# http://java.sun.com/webapps/bugreport/crash.jsp
# The crash happened outside the Java Virtual Machine in native code.
# See problematic frame for where to report the bug.
#
***Error exit code=-6
Oh... and you probably already know that the Gnome screensaver should be
turned off. But I missed that today and wondered why random tests
started failing.
/Jonas
Andi Vajda wrote:
On Wed, 28 Nov 2007, Jonas Beckman wrote:
I've been busy making a brand new Swedish translation. It's now complete
and I mailed it to Brian earlier, since I can't commit localizations yet.
Anyway... your Ubuntu 64-bit version now passes all unit tests on a
virtual machine with two processors. Your suggestion idea that the
kernel had to be SMP-enabled to use multiple processors under VMWare was
correct.
There is no need for a complete re-install. First, check "dmesg | grep
processor". That should show that only one processor is used and three
are unused. Shut down the VM and change the VM settings to use two
processors. Start up again and use the Synaptics packagemanager to
search for "Linux-image". Pick the generic image for x86-64, install (or
re-install) it and re-boot. That's it! "dmesg | grep processor" will now
show that two processors are used and two unused.
You may have to turn off indexing to avoid freezing in the middle of
testing. The trackerd service is problematic (see this thread:
https://bugs.launchpad.net/ubuntu/+source/tracker/+bug/131983).
Note that Fusion's virtual SMP lacks the hardware supported scheduling
that some of VMWare's server products has. It basically just
swaps memory from processor to processor now and then. But I still think
it's very nice to be able to test SMP this way at all.
Hi Jonas,
Thank you very much for the info.
I intend to try this shortly.
Andi..
--- End Message ---
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
Open Source Applications Foundation "chandler-dev" mailing list
http://lists.osafoundation.org/mailman/listinfo/chandler-dev
