lp://qastaging/~ubuntu-branches/ubuntu/lucid/openafs/lucid-201103250310

Branch merges

Propose for merging

No branches dependent on this one.

On hold for merging into lp://qastaging/ubuntu/lucid/openafs

Ubuntu Development Team: Pending requested 2011-03-28

Diff: 0 lines api

Related source package recipes

No recipes using this branch. (?)

Create packaging recipe

Related snap packages

Related bugs

Related blueprints

20. By Russ Allbery on 2009-11-03

* Apply upstream deltas:
  - [7833e472] Make afsd.pod reflect reality
  - [c9974c7a] Avoid prematurely destroying callback_rxcon
  - [9b37972e] Linux: 2.6.32 - Adapt to writeback changes
  - [abdf72bc] Linux: Avoid deadlock in readdir - release GLOCK for
    filldir
  - [bdb4f98a] Protect rx_call iovq from simultaneous attempts to empty
    it
  - [c08609ae] Avoid using released hosts
  - [a410b7fd] Linux - Fix disk cache access for selinux/AppArmor
    constrained processes (LP: #415766)
  - [49094ccf] Add automatic sysname detection for ARM Linux
  - [525b594a] Make ktc_curpag generally available (LP: #446521)
* In the postinst of openafs-modules-dkms, if the openafs module is
  already added in DKMS, try to remove it first before adding it. This
  should more correctly handle the case of a user installing this
  package without the correct kernel headers, having it fail in
  postinst, and then installing the correct headers and having dpkg
  attempt to configure the package again. Thanks, Philipp Kaluza.
  (Closes: #553542)

19. By Russ Allbery on 2009-09-30

When changing the name of the source package built by
openafs-modules-source, we need to change the package name in the
changelog as well or dpkg-gencontrol explodes. Thanks, Aaron M.
Ucko. (Closes: #549140)

18. By Russ Allbery on 2009-08-21

* Add armv5tel to debian/sysname, mapping it to arm_linux26. Patch from
  Daniel Bayer. Also make the similar change for the module source
  package. (Closes: #542315)
* Also install restorevol in the openafs-client package.
* Swap maintainer and uploaders in the packaging files for the module
  source package, matching the main package.
* Update standards version to 3.8.3.
  - Change section of packages generated by openafs-modules-source to
    kernel.
  - Add Homepage to packages generated by openafs-modules-source.

17. By Russ Allbery on 2009-07-10

New upstream release.

16. By Russ Allbery on 2009-04-06

* New upstream release.
  - OPENAFS-SA-2009-001: Avoid a potential kernel memory overrun if more
    items than requested are returned from an InlineBulk or BulkStatus
    message. (CVE-2009-1251)
  - OPENAFS-SA-2009-002: Avoid converting negative errors into invalid
    kernel memory pointers. (CVE-2009-1250)
  - Preliminary support for 2.6.30 kernels.
  - Dynamic vcache allocation support to deal with inotify vcache
    pinning.
  - Do appropriate locking for CellServDB in /proc.
  - Use +dfsg instead of .dfsg for saner version sorting.
* Debian's 2.6.29 packages no longer include symlinks from the
  architecture-specific header tree to the common header tree and
  instead overlay both header trees using kbuild. Change the Autoconf
  probes to always use kbuild and generate stub headers in the paths
  that OpenAFS expects that include the linux headers. Patch from Aaron
  M. Ucko. (Closes: #521745)
* Build PIC versions of libafsauthent and libafsrpc and install them in
  libopenafs-dev for use when AFS code should be embedded into shared
  libraries. Patch from Garrett Wollman.
* Update CellServDB to 2008-11-07 version. (Closes: #522451)
* Update debian/watch for +dfsg naming instead of .dfsg.
* Update standards version to 3.8.1 (no changes required).
* Translation updates:
  - Japanese, thanks Hideki Yamane. (Closes: #521518)

15. By Russ Allbery on 2009-03-27

* Apply upstream CVS deltas:
  - STABLE14-cbr-free-what-you-alloc-20090325: dequeue items in the same
    way they were allocated.
  - STABLE14-shutdown-vcache-avoid-null-deref-20090324: avoid oops on
    shutdown. (LP: #333197)
  - STABLE14-uphys-invalidate-returns-void-20081130: fix apparent Ubik
    synchronization errors due to incorrect use of a void return value.
* Update package sections for the new archive organization.

14. By Russ Allbery on 2009-02-18

* Make openafs-modules-source architecture-specific (arch: any instead
  of arch: all). The new method for generating the kernel source tree
  only generates correct build machinery for the architecture on which
  it was built, making the package architecture-specific. The generated
  tree is so much smaller that this still only uses about 1MB of archive
  space more than the old method and makes each openafs-modules-source
  package much smaller. This is a workaround until we can fix
  libafs_tree to include enough build machinery to handle all Linux
  architectures from one source tree. (Closes: #516023)
* Explicitly remove *-stamp files in the clean target of the module
  build until the module build is upgraded to debhelper 7.

13. By Russ Allbery on 2008-09-22

* Apply upstream patch to free /proc entries in the correct order.
  Thanks, Marc Dionne. (Closes: #493914)
* Apply upstream deltas to support 2.6.27 kernels and to stop using
  COMMON_KERN_CFLAGS for all 2.6 kernels uniformly, which fixes
  problems on amd64 with newer kernels. Thanks, Björn Torkelsson.
  (LP: #267504)
* Translation updates:
  - Swedish, thanks Martin Bagge. (Closes: #493120)

12. By Russ Allbery on 2006-12-17

* Update documentation to note that enabling v4 support or running
  krb524d is no longer required for aklog to work.
* Translation updates:
  - German, thanks Henrik Kroeger. (Closes: #400800)

11. By Russ Allbery on 2006-11-20

* No longer pass explicit cache tuning options to afsd and instead let
  OpenAFS automatically choose tuning based on the cache size.
* Accept trailing whitespace in ThisCell.
* Fix path canonicalization on the server, enabling bos getlog with a
  simple log name to work properly against a Debian bosserver.
* Change the documentation of afsd -shutdown to be less dire and more
  accurate. Thanks, Daniel J. Priem. (Closes: #394990)
* Document (at least partially) AFS's mapping of Kerberos v5 principal
  names to Kerberos v4 format in the aklog man page. Thanks, Daniel
  J. Priem. (Closes: #394832)
* Document that aklog -setpag may not always work.