rpms/autofs
6
0
Fork 0
Code Issues Pull Requests Releases Activity

- add current upstream fixes.

Browse Source
This commit is contained in:
Ian Kent 2012-10-15 15:52:07 +08:00
parent 3462367b2e
commit 66b5fc9ab5
9 changed files with 452 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

56
autofs-5.0.7-README-update-mailing-list-subscription-info.patch Normal file
View File

@ -0,0 +1,56 @@
autofs-5.0.7 - README: update mailing list subscription info
From: Chris Packham <chris.packham@alliedtelesis.co.nz>
Following the kernel.org compromise the mailing list was moved to
vger.kernel.org. Update the subscription info and add URLs for the gmane
mailing list archive.
Signed-off-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
---
CHANGELOG | 1 +
README | 17 ++++++++++++++---
2 files changed, 15 insertions(+), 3 deletions(-)
diff --git a/CHANGELOG b/CHANGELOG
index fe801e8..44c9fb2 100644
--- a/CHANGELOG
+++ b/CHANGELOG
@@ -10,6 +10,7 @@
- fix null map entry order handling.
- make description of default MOUNT_WAIT setting clear.
- configure.in: allow cross compilation.
+- README: update mailing list subscription info.
25/07/2012 autofs-5.0.7
=======================
diff --git a/README b/README
index cef16a9..9024e64 100644
--- a/README
+++ b/README
@@ -43,9 +43,20 @@ Fitzhardinge's <jeremy@goop.org> work on autofs 3. Further enhancements
have been made by Ian Kent <raven@themaw.net>.
If you use or want to help develop autofs, please join the autofs
-mailing list by visiting:
+mailing list by sending an email to:
- http://linux.kernel.org/mailman/listinfo/autofs
+ majordomo@vger.kernel.org
-and folling the instructions there to subscribe to the autofs mailing list.
+With the body text:
+
+ subscribe autofs
+
+Once subscribed you can send patches to:
+
+ autofs@vger.kernel.org
+
+The autofs mailing list archive can be viewed on gmane:
+
+ http://news.gmane.org/gmane.linux.kernel.autofs
+ http://blog.gmane.org/gmane.linux.kernel.autofs

42
autofs-5.0.7-add-timeout-option-description-to-man-page.patch Normal file
View File

@ -0,0 +1,42 @@
autofs-5.0.7 - add timeout option description to man page
From: Ian Kent <ikent@redhat.com>
The pseudo option used t set the timeout for map entries is one of
the most most frequently used autofs options but is not mentioned
in auto.master(5).
---
CHANGELOG | 1 +
man/auto.master.5.in | 5 +++++
2 files changed, 6 insertions(+), 0 deletions(-)
diff --git a/CHANGELOG b/CHANGELOG
index 93b9c26..7b8d185 100644
--- a/CHANGELOG
+++ b/CHANGELOG
@@ -6,6 +6,7 @@
- fix parse buffer initialization.
- fix typo in automount(8).
- dont wait forever to restart.
+- add timeout option description to man page.
25/07/2012 autofs-5.0.7
=======================
diff --git a/man/auto.master.5.in b/man/auto.master.5.in
index 54269f8..21d7544 100644
--- a/man/auto.master.5.in
+++ b/man/auto.master.5.in
@@ -167,6 +167,11 @@ server is specified in the map entry. If no server weights are given
then each available server will be tried in the order listed, within
proximity.
.TP
+.I "\-t, \-\-timeout <seconds>"
+Set the expire timeout for map entries. This option can be used to
+override the global default given either on the command line
+or in the configuration.
+.TP
.I "\-n, \-\-negative\-timeout <seconds>"
Set the timeout for caching failed key lookups. This option can be
used to override the global default given either on the command line

69
autofs-5.0.7-allow-non-root-user-to-check-status.patch Normal file
View File

@ -0,0 +1,69 @@
autofs-5.0.7 - allow non root user to check status
From: Ian Kent <ikent@redhat.com>
---
CHANGELOG | 1 +
redhat/autofs.init.in | 20 +++++++++++++-------
2 files changed, 14 insertions(+), 7 deletions(-)
diff --git a/CHANGELOG b/CHANGELOG
index 44c9fb2..936c9ab 100644
--- a/CHANGELOG
+++ b/CHANGELOG
@@ -11,6 +11,7 @@
- make description of default MOUNT_WAIT setting clear.
- configure.in: allow cross compilation.
- README: update mailing list subscription info.
+- allow non root user to check status.
25/07/2012 autofs-5.0.7
=======================
diff --git a/redhat/autofs.init.in b/redhat/autofs.init.in
index cd5cb34..fe18b3e 100644
--- a/redhat/autofs.init.in
+++ b/redhat/autofs.init.in
@@ -167,6 +167,19 @@ function usage_message() {
RETVAL=0
+# allow non-root users to read status / usage
+
+case "$1" in
+ status)
+ status -p @@autofspiddir@@/autofs.pid -l autofs $prog
+ exit 0;
+ ;;
+ usage)
+ usage_message
+ exit 0;
+ ;;
+esac
+
# Only the root user may change the service status
if [ `id -u` -ne 0 ] && [ "$1" != "status" ]; then
echo "insufficient privilege to change service status"
@@ -184,9 +197,6 @@ case "$1" in
stop)
stop
;;
- status)
- status -p @@autofspiddir@@/autofs.pid -l autofs $prog
- ;;
restart|force-reload)
restart
;;
@@ -202,10 +212,6 @@ case "$1" in
restart
fi
;;
- usage)
- usage_message
- exit 0
- ;;
*)
usage_message
exit 2

41
autofs-5.0.7-configure-in-allow-cross-compilation.patch Normal file
View File

@ -0,0 +1,41 @@
autofs-5.0.7 - configure.in: allow cross compilation
From: Chris Packham <chris.packham@alliedtelesis.co.nz>
The default behaviour of AC_RUN_IFELSE is to stop with an error if cross
compiling. Avoid this by providing the optional 4th argument to set
gcc_supports_pie=no if support for PIE cannot be detected.
Signed-off-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
---
CHANGELOG | 1 +
configure.in | 2 +-
2 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/CHANGELOG b/CHANGELOG
index 961e340..fe801e8 100644
--- a/CHANGELOG
+++ b/CHANGELOG
@@ -9,6 +9,7 @@
- add timeout option description to man page.
- fix null map entry order handling.
- make description of default MOUNT_WAIT setting clear.
+- configure.in: allow cross compilation.
25/07/2012 autofs-5.0.7
=======================
diff --git a/configure.in b/configure.in
index 1a24e34..90bda62 100644
--- a/configure.in
+++ b/configure.in
@@ -307,7 +307,7 @@ DAEMON_CFLAGS=
DAEMON_LDFLAGS=
AC_MSG_CHECKING([whether gcc -fPIE works])
AC_RUN_IFELSE([AC_LANG_PROGRAM([[]], [[int main(void) {return 0;}]])],
- [gcc_supports_pie=yes], [gcc_supports_pie=no])
+ [gcc_supports_pie=yes], [gcc_supports_pie=no], [gcc_supports_pie=no])
AC_MSG_RESULT([$gcc_supports_pie])
if test $gcc_supports_pie = yes ; then
DAEMON_CFLAGS="-fPIE"

54
autofs-5.0.7-dont-wait-forever-to-restart.patch Normal file
View File

@ -0,0 +1,54 @@
autofs-5.0.7 - dont wait forever to restart
From: Ian Kent <ikent@redhat.com>
When restarting autofs the daemon must be stopped before it is started
again if it is to function properly. At the moment the init script waits
forever which is not ok if the daemon won't exit for some reason.
So, if the daemon is still running after the stop, run stop() again, wait
a bit longer and if it still hasn't stopped kill it with a SIGKILL to clear
the way for the startup.
---
CHANGELOG | 1 +
redhat/autofs.init.in | 13 ++++++++++---
2 files changed, 11 insertions(+), 3 deletions(-)
diff --git a/CHANGELOG b/CHANGELOG
index 6051723..93b9c26 100644
--- a/CHANGELOG
+++ b/CHANGELOG
@@ -5,6 +5,7 @@
- fix ipv6 proximity calculation.
- fix parse buffer initialization.
- fix typo in automount(8).
+- dont wait forever to restart.
25/07/2012 autofs-5.0.7
=======================
diff --git a/redhat/autofs.init.in b/redhat/autofs.init.in
index ec6d5d6..cd5cb34 100644
--- a/redhat/autofs.init.in
+++ b/redhat/autofs.init.in
@@ -129,9 +129,16 @@ function restart() {
status autofs > /dev/null 2>&1
if [ $? -eq 0 ]; then
stop
- while [ -n "`pidof $prog`" ] ; do
- sleep 5
- done
+ if [ -n "`pidof $prog`" ]; then
+ # If we failed to stop, try at least one more time
+ # after waiting a little while
+ sleep 20
+ stop
+ auto_pid=`pidof $prog`
+ if [ -n "$auto_pid" ]; then
+ kill -9 $auto_pid
+ fi
+ fi
fi
start
}

81
autofs-5.0.7-fix-null-map-entry-order-handling.patch Normal file
View File

@ -0,0 +1,81 @@
autofs-5.0.7 - fix null map entry order handling
From: Ian Kent <ikent@redhat.com>
If a null map entry appears after a corresponding indirect map entry
autofs doesn't handle it properly.
Since it appears after the map entry it should'nt affect it but autofs
doesn't account for this case and assumes the map entry is already
mounted and tries to shut it down causing attempted access to facilities
that don't exist.
---
CHANGELOG | 1 +
lib/master.c | 32 +++++++++++++++++++++++++++++---
2 files changed, 30 insertions(+), 3 deletions(-)
diff --git a/CHANGELOG b/CHANGELOG
index 7b8d185..79cf673 100644
--- a/CHANGELOG
+++ b/CHANGELOG
@@ -7,6 +7,7 @@
- fix typo in automount(8).
- dont wait forever to restart.
- add timeout option description to man page.
+- fix null map entry order handling.
25/07/2012 autofs-5.0.7
=======================
diff --git a/lib/master.c b/lib/master.c
index 904b13d..a0e62f2 100644
--- a/lib/master.c
+++ b/lib/master.c
@@ -1179,9 +1179,35 @@ int master_mount_mounts(struct master *master, time_t age, int readall)
cache_readlock(nc);
ne = cache_lookup_distinct(nc, this->path);
- if (ne && this->age > ne->age) {
+ /*
+ * If this path matched a nulled entry the master map entry
+ * must be an indirect mount so the master map entry line
+ * number may be obtained from this->maps.
+ */
+ if (ne) {
+ int lineno = ne->age;
cache_unlock(nc);
- st_add_task(ap, ST_SHUTDOWN_PENDING);
+
+ /* null entry appears after map entry */
+ if (this->maps->master_line < lineno) {
+ warn(ap->logopt,
+ "ignoring null entry that appears after "
+ "existing entry for %s", this->path);
+ goto cont;
+ }
+ if (ap->state != ST_INIT) {
+ st_add_task(ap, ST_SHUTDOWN_PENDING);
+ continue;
+ }
+ /*
+ * The map entry hasn't been started yet and we've
+ * seen a preceeding null map entry for it so just
+ * delete it from the master map entry list so it
+ * doesn't get in the road.
+ */
+ list_del_init(&this->list);
+ master_free_mapent_sources(ap->entry, 1);
+ master_free_mapent(ap->entry);
continue;
}
nested = cache_partial_match(nc, this->path);
@@ -1194,7 +1220,7 @@ int master_mount_mounts(struct master *master, time_t age, int readall)
cache_delete(nc, nested->key);
}
cache_unlock(nc);
-
+cont:
st_mutex_lock();
state_pipe = this->ap->state_pipe[1];

25
autofs-5.0.7-include-usage-in-usage-message.patch Normal file
View File

@ -0,0 +1,25 @@
autofs-5.0.7 - include usage in usage message
From: Ian Kent <ikent@redhat.com>
Since usage is a case entry we should also nclude it in the usage
message.
---
redhat/autofs.init.in | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/redhat/autofs.init.in b/redhat/autofs.init.in
index 8e355da..ec6d5d6 100644
--- a/redhat/autofs.init.in
+++ b/redhat/autofs.init.in
@@ -155,7 +155,7 @@ function reload() {
}
function usage_message() {
- echo $"Usage: $0 {start|forcestart|stop|status|restart|force-reload|forcerestart|reload|condrestart|try-restart}"
+ echo $"Usage: $0 {start|forcestart|stop|status|restart|force-reload|forcerestart|reload|condrestart|try-restart|usage}"
}
RETVAL=0

57
autofs-5.0.7-make-description-of-default-MOUNT_WAIT-setting-clear.patch Normal file
View File

@ -0,0 +1,57 @@
autofs-5.0.7 - make description of default MOUNT_WAIT setting clear
From: Ian Kent <ikent@redhat.com>
---
CHANGELOG | 1 +
redhat/autofs.sysconfig.in | 5 +++--
samples/autofs.conf.default.in | 5 +++--
3 files changed, 7 insertions(+), 4 deletions(-)
diff --git a/CHANGELOG b/CHANGELOG
index 79cf673..961e340 100644
--- a/CHANGELOG
+++ b/CHANGELOG
@@ -8,6 +8,7 @@
- dont wait forever to restart.
- add timeout option description to man page.
- fix null map entry order handling.
+- make description of default MOUNT_WAIT setting clear.
25/07/2012 autofs-5.0.7
=======================
diff --git a/redhat/autofs.sysconfig.in b/redhat/autofs.sysconfig.in
index 36b924d..a8992c4 100644
--- a/redhat/autofs.sysconfig.in
+++ b/redhat/autofs.sysconfig.in
@@ -18,8 +18,9 @@ TIMEOUT=300
# Setting this timeout can cause problems when
# mount would otherwise wait for a server that
# is temporarily unavailable, such as when it's
-# restarting. The defailt of waiting for mount(8)
-# usually results in a wait of around 3 minutes.
+# restarting. The default setting (-1) of waiting
+# for mount(8) usually results in a wait of around
+# 3 minutes.
#
#MOUNT_WAIT=-1
#
diff --git a/samples/autofs.conf.default.in b/samples/autofs.conf.default.in
index ac2f63b..1da89cf 100644
--- a/samples/autofs.conf.default.in
+++ b/samples/autofs.conf.default.in
@@ -18,8 +18,9 @@ TIMEOUT=300
# Setting this timeout can cause problems when
# mount would otherwise wait for a server that
# is temporarily unavailable, such as when it's
-# restarting. The defailt of waiting for mount(8)
-# usually results in a wait of around 3 minutes.
+# restarting. The default setting (-1) of waiting
+# for mount(8) usually results in a wait of around
+# 3 minutes.
#
#MOUNT_WAIT=-1
#

28
autofs.spec
View File

@ -8,7 +8,7 @@
Summary: A tool for automatically mounting and unmounting filesystems
Name: autofs
Version: 5.0.7
Release: 3%{?dist}
Release: 4%{?dist}
Epoch: 1
License: GPLv2+
Group: System Environment/Daemons
@ -18,6 +18,14 @@ Patch2: autofs-5.0.7-fix-use-cache-entry-after-free-mistake.patch
Patch3: autofs-5.0.7-fix-ipv6-proximity-calculation.patch
Patch4: autofs-5.0.7-fix-parse-buffer-initialization.patch
Patch5: autofs-5.0.7-fix-typo-in-automount-8.patch
Patch6: autofs-5.0.7-include-usage-in-usage-message.patch
Patch7: autofs-5.0.7-dont-wait-forever-to-restart.patch
Patch8: autofs-5.0.7-add-timeout-option-description-to-man-page.patch
Patch9: autofs-5.0.7-fix-null-map-entry-order-handling.patch
Patch10: autofs-5.0.7-make-description-of-default-MOUNT_WAIT-setting-clear.patch
Patch11: autofs-5.0.7-configure-in-allow-cross-compilation.patch
Patch12: autofs-5.0.7-README-update-mailing-list-subscription-info.patch
Patch13: autofs-5.0.7-allow-non-root-user-to-check-status.patch
Buildroot: %{_tmppath}/%{name}-%{version}-%{release}-root-%(%{__id_u} -n)
%if %{with_systemd}
BuildRequires: systemd-units
@ -79,6 +87,14 @@ echo %{version}-%{release} > .version
%patch3 -p1
%patch4 -p1
%patch5 -p1
%patch6 -p1
%patch7 -p1
%patch8 -p1
%patch9 -p1
%patch10 -p1
%patch11 -p1
%patch12 -p1
%patch13 -p1
%build
#CFLAGS="$RPM_OPT_FLAGS" ./configure --prefix=/usr --libdir=%{_libdir}
@ -170,6 +186,16 @@ fi
%dir /etc/auto.master.d
%changelog
* Mon Sep 15 2012 Ian Kent <ikent@redhat.com> - 1:5.0.7-4
- include usage in usage message.
- dont wait forever to restart.
- add option description to man page.
- fix null map entry order handling.
- make description of default MOUNT_WAIT setting clear.
- configure.in: allow cross compilation.
- README: update mailing list subscription info.
- allow non root user to check status.
* Mon Sep 12 2012 Ian Kent <ikent@redhat.com> - 1:5.0.7-3
- fix nobind sun escaped map entries.
- fix use cache entry after free mistake.