New upstream version available (2.0.4)

Bug #288957 reported by agent 8131
26
This bug affects 1 person
Affects Status Importance Assigned to Milestone
virtualbox-ose (Ubuntu)
Fix Released
Wishlist
Unassigned

Bug Description

Binary package hint: virtualbox-ose

I know it might be a problem to incorporate this release in Intrepid but a couple of items from the changelog for 2.0.4 caught my interest:

# Linux additions: fixed seek() for files bigger than 2GB (bug #2379)
# Linux additions: support Ubuntu 8.10
# Linux additions: clipboard fixes (bug #2015)

Certainly it would be nice if the version of VirtualBox shipping with Ubuntu 8.10 has support for Ubuntu 8.10. I'm running Ubuntu 8.10 guests currently but that's using vt-x. I assume this is some sort of software support but the changelog is not clear. But given the number of bugs fixed this would probably be worth updating in Intrepid if at all possible.

Changes since the current version in Ubuntu Intrepid (from http://www.virtualbox.org/wiki/Changelog)

    * VMM: better error reporting for VT-x failures
    * VMM: don’t overflow the release log with PATM messages (bug #1775)
    * VMM: fixed save state restore in real mode (software virtualization only)
    * GUI: work around a Qt bug on Mac OS X (bug #2321)
    * GUI: properly install the Qt4 accessible plugin (bug #629)
    * SATA: error message when starting a VM with a VMDK connected to a SATA port (bug #2182)
    * SATA: fixed Guru mediation when booting OpenSolaris/64; most likely applies to other guests as well (bug #2292)
    * Network: don’t crash when changing the adapter link state if no host driver is attached (bug #2333)
    * VHD: fixed bug which prevents booting from VHD images bigger than 4GB (bug #2085)
    * VRDP: fixed a repaint problem when the guest resolution was not equal to the client resolution
    * Clipboard: don’t crash when host service initialization takes longer than expected (Linux hosts only; bug #2001)
    * Windows hosts: VBoxSVC.exe crash (bug #2212)
    * Windows hosts: VBoxSVC.exe memory leak due to a Windows WMI memory leak (Vista only) (bug #2242)
    * Windows hosts: VBoxSVC.exe delays GUI startup
    * Linux hosts: handle jiffies counter overflow (VM stuck after 300 seconds of host uptime; bug #2247)
    * Solaris hosts: fixed host or guest side networking going stale while using host interface networking (bug #2474)
    * Solaris hosts: added support for using unplumbed network interfaces and Crossbow Virtual Network Interfaces (VNICs) with host interface networking
    * Solaris hosts: reworked threading model improves performance for host interface networking
    * Windows additions: fixed crash when accessing deep directory structures in a shared folder
    * Windows additions: improved shared folder name resolving (bug #1728)
    * Windows additions: fixed Windows 2000 shutdown crash (bug #2254)
    * Windows additions: fixed error code for MoveFile() if the target exists (bug #2350)
    * Linux additions: fixed seek() for files bigger than 2GB (bug #2379)
    * Linux additions: support Ubuntu 8.10
    * Linux additions: clipboard fixes (bug #2015)
    * Web services: improved documentation and fixed example (bug #1642)

Tags: upgrade
Daniel Hahler (blueyed)
description: updated
Daniel Hahler (blueyed)
Changed in virtualbox-ose:
assignee: nobody → blueyed
status: New → In Progress
Revision history for this message
Daniel Hahler (blueyed) wrote :
Revision history for this message
Daniel Hahler (blueyed) wrote :

Subscribing motu-release for ACKing.

My upload would include the following changes:
  [ Jonathan Ernst ]
  * Virtualbox should go in Accessories, not in System tools (LP: #288590)

  [ Daniel Hahler ]
  * New upstream version (LP: #288957)
    - Updated debian/patches/09-initscript.dpatch (one hunk rejected/obsolete)
  * Drop debconf question about losing snapshots because of upgrading: it is
    no longer relevant and the default (cancel upgrade) caused a lot of
    trouble (LP: #288178, #221736)
    - debian/virtualbox-ose.preinst.in, debian/virtualbox-ose.templates,
      debian/rules
  * Improve patch for disabling the "check for updates" dialog (LP: #272212)
    - debian/patches/13-disable-update-action.dpatch
  * Disable registration request at Sun (LP: #285684, #271603)
    - debian/patches/14-disable-registration-request.dpatch

The package builds fine, installs and upgrades.
I've checked that it runs as expected (tested with an Intrepid guest).

description: updated
Changed in virtualbox-ose:
assignee: blueyed → nobody
importance: Undecided → Wishlist
status: In Progress → New
Revision history for this message
Daniel Hahler (blueyed) wrote :
Revision history for this message
Daniel Hahler (blueyed) wrote :

I've uploaded this package to my PPA for testing: https://launchpad.net/~blueyed/+archive

Revision history for this message
Scott Kitterman (kitterman) wrote : Re: [Bug 288957] [NEW] New upstream version available (2.0.4)

How much testing has this had?

Revision history for this message
Hew (hew) wrote :

I just did some brief testing with the PPA version and everything seems to work for me. Looks like bug 272212 has been correctly resolved :-)

Revision history for this message
Luca Falavigna (dktrkranz) wrote :

It seems it has lots of bugfixes and support for 8.10 is quite important IMO, ACK #1.
Is there any chance to fix build failures on unofficial ports?

Revision history for this message
Daniel Hahler (blueyed) wrote :

> How much testing has this had?

Not that much (from myself); as I've said I've tested that it works, but cannot test everything of course.

> Is there any chance to fix build failures on unofficial ports?

What are you referring to? Do you have some build log where it fails? Anyway, as long as you don't have a (trivial) patch, I'd say it's difficult.

Revision history for this message
Scott Kitterman (kitterman) wrote :

Daniel: I'll approve this if you'll agree to be responsible for working with upstream and getting patches/preparing SRU for any regressions.

Revision history for this message
Daniel Hahler (blueyed) wrote : Re: [Bug 288957] Re: New upstream version available (2.0.4)

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ok.
I'll upload it in a minute then.

- --
http://daniel.hahler.de/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFJBLxXfAK/hT/mPgARAlnoAKDyd206fvFPNAbOrqp3TeaXbm+ZbQCfeAgo
6dNjPnA4WTgiZM6SPME0CLs=
=I0Y/
-----END PGP SIGNATURE-----

Revision history for this message
Scott Kitterman (kitterman) wrote :

Ack #2. Approved

Changed in virtualbox-ose:
status: New → Confirmed
Revision history for this message
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package virtualbox-ose - 2.0.4-dfsg-0ubuntu1

---------------
virtualbox-ose (2.0.4-dfsg-0ubuntu1) intrepid; urgency=low

  [ Jonathan Ernst ]
  * Virtualbox should go in Accessories, not in System tools (LP: #288590)

  [ Daniel Hahler ]
  * New upstream version (LP: #288957)
    - Updated debian/patches/09-initscript.dpatch (one hunk rejected/obsolete)
  * Drop debconf question about losing snapshots because of upgrading: it is
    no longer relevant and the default (cancel upgrade) caused a lot of
    trouble (LP: #288178, #221736)
    - debian/virtualbox-ose.preinst.in, debian/virtualbox-ose.templates,
      debian/rules
  * Improve patch for disabling the "check for updates" dialog (LP: #272212)
    - debian/patches/13-disable-update-action.dpatch
  * Disable registration request at Sun (LP: #285684, #271603)
    - debian/patches/14-disable-registration-request.dpatch

 -- Daniel Hahler <email address hidden> Sun, 26 Oct 2008 02:23:42 +0100

Changed in virtualbox-ose:
status: Confirmed → Fix Released
Revision history for this message
Matti Lindell (mlind) wrote :

Thanks for the new upstream version. It seems that the package was accidentally uploaded as a native package though.

Revision history for this message
Daniel Hahler (blueyed) wrote :

Matti, what do you mean? It should be 2.0.4-dfsg-0ubuntu1, as mentioned in the changelog entry cited above.

Revision history for this message
Matti Lindell (mlind) wrote :

Daniel, there's no .diff.gz anymore, but the package naming suggests otherwise as you said.

Revision history for this message
Daniel Hahler (blueyed) wrote :

Ouch.. somehow I've seem to have screwed this up.

I've tried to fix it now, by creating a orig.tar.gz ("debian/rules dfsg-free"), bumping the version in debian/changelog and uploading it again, but it's likely to not work out as expected, since dput does not upload the new orig.tar.gz..

Suggestions about how to fix this?

Revision history for this message
Siegfried Gevatter (rainct) wrote :

I've only read the last comment, but: shouldn't the .orig.tar.gz be uploaded if you use "debuild -S -sa"? If the archive rejects this (it probably will) using a version number like <version number>really<the same version number>-0ubuntuX or anything else should do it. I hope this helps.

Revision history for this message
William Grant (wgrant) wrote :

You should be able to just upload -0ubuntu2 - the .orig.tar.gz was never published, so there won't be a conflict.

Revision history for this message
Scott Kitterman (kitterman) wrote :

At this point just leave it the way it is. It'll need a bump the first time
the kernel gets incompatible. Fix it then.

Revision history for this message
Daniel Hahler (blueyed) wrote :

Scott, a new kernel did never require an update of virtualbox-ose, but only virtualbox-ose-modules.
But since Intrepid, v-o-m is gone and DKMS is used instead.

Yes, "-sa" probably does the trick. I'm uploading 2.0.4-dfsg-0ubuntu2 with a reference to this bug report/discussion.
I think having the orig.tar.gz and diff.gz separated is really good to have and since there is no update foreseeable in the near future, it should get fixed now already.

Revision history for this message
Scott Kitterman (kitterman) wrote :

The diff was more than just native/non-native. Let's discuss on IRC.

Revision history for this message
Daniel Hahler (blueyed) wrote :

The upload got rejected, I've talked about it with Scott on IRC. Seems like we won't fix it.

IRC log:
<blueyed> ScottK: the diff should have been only native/non-native, plus DP description for the latest patch.
<ScottK> blueyed: When I debdiffed it there was more than that.
<blueyed> ScottK: the .cvsignore files probably?! it seems like they go ignored in the native package, but not in the orig.tar.gz..
<blueyed> ScottK: anyway.. I'll have to go for some hours now..
<ScottK> blueyed: As a practical matter I think it's fine as it is.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Duplicates of this bug

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.