Re: 2.6.11 testing
"Stan" <[email protected]>
| Newsgroups | gmane.linux.cluster.ssic.devel |
|---|---|
| Message-ID | <[email protected]> |
Thanks. > -----Original Message----- > From: John Hughes [mailto:[email protected]] > Sent: Tuesday, July 31, 2007 1:36 AM > To: Stan > Cc: 'ssic-linux-devel' > Subject: Re: [SSI-devel] 2.6.11 testing > > Stan wrote: > > Hi, John, > > What's you favorite method for brining up (debugging) a > new SSI kernel? > > UML (User-Mode-Linux/SSI?) or some other mechanism beyond a > serial port and > > hard reboots? > > > At the moment I'm having best results with qemu (with or > without kqemu) > and kvm (faster than qemu, only works on some hardware). Did you encounter any anomalies in the qemu Linux setup instructions? > > One of the problems with OpenSSI is that it's far to easy to > end up with > an unbootable system - if you can make a new image with a cp or two > things go a lot faster. > > For the networking I find vde a lot easier to set up than > tun/bridging. > > > What type of OSSI correctness tests are you using beyond DRDB? > > > I'm trying all the basic functionality: > > 1. Cross node device/file access (ok) > 2. CVIP (ok, with failover) > 3. migration (ok) > 4. boot/shutdown non-init nodes (ok) > 5. boot/shutdown potential init nodes (ok) > 6. root failover with shared disk (using qemu this is extra easy to > set up - the "shared disk" is just a file). (ok) > 7. non-root failover with shared disk (in progress) > 8. root failover with drdb (problems) > 9. non-root failover with drbd (not yet started) > > > Do you or someone else have an OSSI regression test suite? > > > I don't. Would be a nice thing to have. > > ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/