Russ Allbery [Fri, 27 Mar 2009 14:59:25 +0000 (07:59 -0700)]
Apply upstream delta to fix Ubik synchronization
* Apply upstream CVS deltas:
- STABLE14-uphys-invalidate-returns-void-20081130: fix apparent Ubik
synchronization errors due to incorrect use of a void return value.
Russ Allbery [Fri, 27 Mar 2009 14:48:21 +0000 (07:48 -0700)]
Apply upstream deltas to fix shutdown problems
* 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)
Russ Allbery [Wed, 18 Feb 2009 22:20:52 +0000 (14:20 -0800)]
Make openafs-modules-source arch: all
* 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)
Russ Allbery [Fri, 13 Feb 2009 03:40:45 +0000 (19:40 -0800)]
Fix creation of the openafs-modules-* packages
Also remove server documentation from openafs-modules-source and add
the README to openafs-modules-dkms, and remove DH_OPTIONS from
debian/rules since it's no longer needed.
Russ Allbery [Fri, 13 Feb 2009 02:02:30 +0000 (18:02 -0800)]
General improvements to debconf configuration handling
* General improvements to debconf configuration handling.
- Do not abort the config script if we have no default realm and
dnsdomainname fails. (LP: #274886, #318132)
- Do not abort postinst if we can't resolve any host name for the
cell. Just continue in that case, omitting any configuration for
the local cell. (LP: #322054)
- Correctly set the default for AFS_CRYPT from the existing config.
- The cache size is high, not critical, since we have a default.
- Don't create the client ThisCell if no local cell was specified.
- Improve the leading comment in afs.conf.client. Users can edit it,
but any changes other than the setting values will be lost.
- Remove the openafs-fileserver/bosconfig_moved template, which would
only be shown on upgrades from pre-1.2.0 versions.
- Standardize coding style and always quote strings.
Russ Allbery [Fri, 13 Feb 2009 01:27:03 +0000 (17:27 -0800)]
Don't exit successfully if sourcing the debconf module fails
* Abort the config script rather than exiting successfully if the
debconf shell module is not available. The config script should never
be run in the absence of debconf, so if this fails, something deeper
is wrong and we shouldn't blindly continue.
Russ Allbery [Fri, 13 Feb 2009 01:24:57 +0000 (17:24 -0800)]
Add initial DKMS support and compress module source with bzip2
* Add initial DKMS support via an openafs-modules-dkms package, based on
work by Robert Gerlach. (LP: #288743)
* Compress the source in openafs-modules-source with bzip2 instead of
gzip to reduce the package size.
Russ Allbery [Thu, 12 Feb 2009 21:23:20 +0000 (13:23 -0800)]
First cut at debhelper 7 support
* Update to debhelper compatibility level V7.
- Use dh where possible in debian/rules.
- Use dh_lintian to install Lintian overrides.
- Eliminate the configure target and do all the work in build.
- Do more in the install target and less in binary, matching the
default command stacks in dh.
- Use a debian/clean configuration file to let dh clean do more of the
work rather than using rm commands in debian/rules.
Russ Allbery [Sun, 1 Feb 2009 17:49:25 +0000 (09:49 -0800)]
Make dynroot the default for new installs
* Make dynroot the default for new installations. It works much better
with systems that don't bring up their network until late in the boot
process, such as wireless laptops. (LP: #318605)
Russ Allbery [Mon, 1 Dec 2008 03:01:39 +0000 (19:01 -0800)]
Improve README.modules
* Suggest module-assistant auto-install in README.modules and be clearer
that building one's own kernel is not required. Thanks, Anders
Kaseorg and Lutz Goehring for suggestions. (LP: #295897)
Russ Allbery [Mon, 6 Oct 2008 18:10:23 +0000 (11:10 -0700)]
Mention current packages in libpam-openafs-kaserver description
* Mention libpam-afs-session instead of libpam-openafs-session in the
description of libpam-openafs-kaserver, and mention libpam-krb5 by
name. Thanks, Philipp Kaluza. (Closes: #501308)
* 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)
Use ARCH in preference to uname -m when building kernel modules
* When building the kernel module, use the setting of the ARCH
environment or make variable, if present, in preference to uname -m to
allow cross-compiles. Thanks, Jose Calhariz. (Closes: #492232)
We only have a param file for a 2.4 kernel, we were building with UCONTEXT
which is stubbed out in glibc for hppa, and builds without UCONTEXT fail.
The binaries we were building before didn't work and no one reported a
bug, indicating no one was attempting to use OpenAFS on hppa.
Undo the move of up.pod in clean to not FTBFS on the second build
Undo the move of up.pod to afs-up.pod in debian/rules clean rather
than just deleting the file and causing the second build to fail.
Thanks, Dominic Hargreaves.
Russ Allbery [Mon, 30 Jun 2008 20:32:18 +0000 (13:32 -0700)]
Don't try to use UCONTEXT on hppa/parisc
glibc doesn't implement the ucontext functions on hppa/parisc, so don't
try to use them. This probably means that OpenAFS user space is broken
on hppa right now. That's not a surprise, given that we don't have a
2.6 kernel param file either.
Russ Allbery [Sat, 28 Jun 2008 06:52:31 +0000 (23:52 -0700)]
Ship fewer debug symbols
Don't include debug symbols for /sbin, /usr/bin, and /usr/sbin binaries or
PAM modules in openafs-dbg, just the servers in /usr/lib/openafs, since
those are the most likely cases where backtraces are important and a 10MB
debug symbol package seems like overkill.
Russ Allbery [Sat, 28 Jun 2008 06:25:50 +0000 (23:25 -0700)]
Add lpia as an alias for i386
Enable support for lpia by making it synonymous with i386. This means
that lpia systems will get a sysname of i386_linux26, which isn't ideal
but should at least work.
Russ Allbery [Sat, 28 Jun 2008 06:01:22 +0000 (23:01 -0700)]
Exit successfully in init script if openafs isn't in modules.dep
If the openafs module isn't listed in modules.dep, exit with a zero
status from the init script even if modprobe fails. Otherwise, if a
module package and openafs-client are installed at the same time and
openafs-client is set up before the module package, the init script can
find the module, fail the modprobe (because the dependency map hasn't
yet been updated), and fail the whole installation. Exiting with a zero
status still doesn't get OpenAFS started, but at least it doesn't leave
the system in a partly installed state. Thanks to Durk Strooisma for the
analysis.