On Wed, 2005-08-24 at 11:36 -0700, Bryce Harrington wrote: > On Wed, Aug 17, 2005 at 03:46:38PM -0400, Silbermann, Martine wrote: > > Currently test case 2, 3, 4 for CPU hotplug have scripts associated to > > them. Test case 6 has been done for TOP and Mary generated a list of all > > the other tools that we'll keep track of in the context of this test > > case. There's an existing script for test case 1 that will be integrated > > in the suite. Test case 5 is a stress test (which can also be used to > > keep track of performance changes). First Mary will use the database > > test to do stress testing; later on we might combine NFS and hotplug > > testing. The problem will be spotting problems in an automated way. The > > stress tests are expected to run over long periods of time. > > Over the past couple days, MarkD and I hammered out test case 1. The > thing that made this one difficult compared with the other cases was > that it required several verifications but did not explain how to do > those, so it took time for Mark and I to develop approaches for doing > this. There are still a few remaining questions, but we were able to > get the script essentially operational. I plan to do a bit more work > before I publish it, hopefully later this week. > > MarkD also added sar to the hotplug 6 test case. There are a couple > other tools to add, however there are no known failure situations for > them; we could simply check to make sure the tools don't terminate > during a hotplug event, though. > In the con call on 08/02 Mary mentioned that SAR was not behaving properly when CPUs are added (it didn't crash but didn't show the new CPUs). David Lively took the AR to look at the problem but I haven't seen a patch from him. Did you get one? or do you no longer see this problem with SAR? Thx - Martine > > Last week I got the framework set up to run through complete test sets - > it automatically downloads the patch, applies it, builds the kernel, and > invokes the lhcs_regression test suite, capturing all logs and results. > The major missing piece is to install the kernel and reboot the ppc64 > machine after building the kernel and before running the regression > test. > > For rebooting the ppc64 machine, we finally got the hmc hooked up on our > lab network properly a couple weeks ago, so I think we've reached the > end of that whole ppc64 hardware saga. ;-) I have the hmc manual > sitting on my desk, and plan to learn how to do remote power control, > etc. in the coming week, so I can finish automating that bit. After > that, I think we'll be fully in business. > > Bryce
