facter does not recognize KVM VMs as 'virtual'

Hmpf, it's been fixed already in 1.5.7-3ubuntu1, by way of Debian.

Oliver, looking at your patch, a couple of minor changes are necessary:

* the distribution target is set to 'lucid', should be 'lucid-proposed' in debian/changelog
* The version in the changelog should be 1.5.6-2ubuntu2.1 (this is SRU policy)

Changelog entry of the fixed version:

 facter (1.5.7-2) unstable; urgency=low
 .
   [ Stig Sandbeck Mathisen ]
   * Fix "upgrade from 1.3.5 to 1.5.1 requires puppet to be restarted",
     added a note in debian/NEWS about restarting puppet (Closes: #515726)
   * Add versioned dependency for ruby1.8 (Closes: #585867)
   * debian/control: bump standards version: no changes
   * debian/control: Remove unused build dependency on "cdbs"
   * Add versioned dependency for ruby1.8 (Closes: #585867)
   * Cherry-pick domain/fqdn fact fix from upstream (Closes: #515591)
 .
   [ Micah Anderson ]
   * debian/control: add Vcs-Git, Vcs-Browser, and Homepage fields
   * debian/NEWS: fix lintian syntax-error-in-debian-news-file
   * debian/facter.8: fix manpage-section-mismatch
   * cherry pick upstream virtualization support:
     . openvz fix [#4156]
     . add kvm support (Closes: #603001)
     . fix Linux-Vserver of virtual fact [#3397]
     . add xendomains fact

Corrected debdiff for Lucid, thanks to Clint for the advice!

Nominating for SRU, fulfills: (1) have an obviously safe patch and (2) affect an application rather than critical infrastructure packages

Justification: it fixes obviously wrong behaviour in regard to a package that is central to the Ubuntu Server product (KVM).

The fix is from upstream and has been verified to exist in Natty.

Regression potential: Existing puppet manifests which evaluate either the "virtual" or "is_virtual" fact will be impacted when run on KVM guests. However, the existence of such manifests is to be doubted, since they would be broken in such an environment anyway. Administrators that have worked around this bug may possibly notice a regression, though.

Hi Oliver,

Thanks for your debdiffs. They look good to me. I have sponsored your changes into lucid-proposed and maverick-proposed. As soon as they build, we'll need you (or anyone else) to test the resulting packages and ensure sure that they fix this bug and don't cause other regressions.

Thanks again!

your patch pilot,
Dustin

Accepted facter into lucid-proposed, the package will build now and be available in a few hours. Please test and give feedback here. See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Thank you in advance!

Accepted facter into maverick-proposed, the package will build now and be available in a few hours. Please test and give feedback here. See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Thank you in advance!

I have just verified that the new packages fix this bug for both Lucid and Maverick. I did not notice any regressions on either VM, nor on a physical system running Lucid.

This bug was fixed in the package facter - 1.5.6-2ubuntu2.1

---------------
facter (1.5.6-2ubuntu2.1) lucid-proposed; urgency=low

  * Cherry-picked KVM detection for 'virtual' fact from upstream commit
    62b6773a63bb96273fbc. (LP: #708080).
 -- Oliver Brakmann <email address hidden> Wed, 26 Jan 2011 21:36:23 +0100

This bug was fixed in the package facter - 1.5.7-1ubuntu1.1

---------------
facter (1.5.7-1ubuntu1.1) maverick-proposed; urgency=low

  * Cherry-picked KVM detection for 'virtual' fact from upstream commit
    62b6773a63bb96273fbc. (LP: #708080).
 -- Oliver Brakmann <email address hidden> Thu, 27 Jan 2011 19:03:10 +0100