[systemd-devel] [PATCH] hostnamed: Fix the way that static and transient host names interact

[Stef Walter]

It is almost always incorrect to allow DHCP or other sources of
transient host names to override an explicitly configured static host
name.

This commit changes things so that if a static host name is set, this
will override the transient host name (eg: provided via DHCP). Transient
host names can still be used to provide host names for machines that have
not been explicitly configured with a static host name.

The exception to this rule is if the static host name is set to
localhost or matches localhost.*. In those cases we act as if no
static host name has been explicitly set.

As discussed elsewhere, systemd may want to have an fd based ownership
of the transient name. That part is not included in this commit.
---
 man/hostnamectl.xml  |  8 +++-
 src/hostname/hostnamed.c | 36 +++-
 2 files changed, 34 insertions(+), 10 deletions(-)

diff --git a/man/hostnamectl.xml b/man/hostnamectl.xml
index 4bc05f5..80e8cde 100644
--- a/man/hostnamectl.xml
+++ b/man/hostnamectl.xml
@@ -68,11 +68,9 @@
 (e.g. Lennart's Laptop), the static hostname which
 is used to initialize the kernel hostname at boot
 (e.g. lennarts-laptop), and the transient hostname
-which might be assigned temporarily due to network
-configuration and might revert back to the static
-hostname if network connectivity is lost and is only
-temporarily written to the kernel hostname
-(e.g. dhcp-47-11)./para
+which is a default received from network configuration.
+If a static hostname is set, and is valid (something other
+than localhost) then the transient hostname is not used./para
 
 paraNote that the pretty hostname has little
 restrictions on the characters used, while the static
diff --git a/src/hostname/hostnamed.c b/src/hostname/hostnamed.c
index e57891b..5174fa8 100644
--- a/src/hostname/hostnamed.c
+++ b/src/hostname/hostnamed.c
@@ -228,16 +228,36 @@ static char* context_fallback_icon_name(Context *c) {
 return strdup(computer);
 }
 
-static int context_write_data_hostname(Context *c) {
+static bool hostname_is_useful(const char *hn) {
+return !isempty(hn)  !streq(hn, localhost) 
+   !startswith(hn, localhost.);
+}
+
+static int context_update_kernel_hostname(Context *c) {
+const char *static_hn;
 const char *hn;
 
 assert(c);
 
-if (isempty(c-data[PROP_HOSTNAME]))
-hn = localhost;
-else
+static_hn = c-data[PROP_STATIC_HOSTNAME];
+
+/* /etc/hostname with something other than localhost
+ * has the highest preference ... */
+if (hostname_is_useful(static_hn))
+hn = static_hn;
+
+/* ... the transient host name, (ie: DHCP) comes next ...*/
+else if (!isempty(c-data[PROP_HOSTNAME]))
 hn = c-data[PROP_HOSTNAME];
 
+/* ... fallback to static localhost.* ignored above ... */
+else if (!isempty(static_hn))
+hn = static_hn;
+
+/* ... and the ultimate fallback */
+else
+hn = localhost;
+
 if (sethostname(hn, strlen(hn))  0)
 return -errno;
 
@@ -389,7 +409,7 @@ static int method_set_hostname(sd_bus *bus, sd_bus_message 
*m, void *userdata, s
 free(c-data[PROP_HOSTNAME]);
 c-data[PROP_HOSTNAME] = h;
 
-r = context_write_data_hostname(c);
+r = context_update_kernel_hostname(c);
 if (r  0) {
 log_error(Failed to set host name: %s, strerror(-r));
 return sd_bus_error_set_errnof(error, r, Failed to set 
hostname: %s, strerror(-r));
@@ -441,6 +461,12 @@ static int method_set_static_hostname(sd_bus *bus, 
sd_bus_message *m, void *user
 c-data[PROP_STATIC_HOSTNAME] = h;
 }
 
+r = context_update_kernel_hostname(c);
+if (r  0) {
+log_error(Failed to set host name: %s, strerror(-r));
+return sd_bus_error_set_errnof(error, r, Failed to set 
hostname: %s, strerror(-r));
+}
+
 r = context_write_data_static_hostname(c);
 if (r  0) {
 log_error(Failed to write static host name: %s, 
strerror(-r));
-- 
1.8.5.3

___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

[systemd-devel] threads and cgroups

[Allmeroth, Robert]

Hi,

I read a lot about the new cgroup concept of the kernel/systemd.
I understand the logic behind it and why systemd wants to be the MCP of it.
But I miss information how threads are handled.

Example:
Due to performance reasons we arrange components with various profiles into one 
executable.
Assume a splash-animation on startup, audio-management, CAN, key input etc. 
Some of them should be
protected by a realtime cgroup - others not.

Right now we are able to put the whole unit/process into one cgroup with 
systemd mechanisms (I like the slice concept :))
Named threads can be handled by extra tooling, e.g. invoked via ExecStartPost, 
managing the threads/cgroups directly 

Problems will arise if manual management is impossible in future 
implementations.
Is there a systemd concept if and/or how threads living inside of processes can 
be assigned to separate cgroups?

Best regards
  Robert

PS: resources I used:
http://www.freedesktop.org/wiki/Software/systemd/ControlGroupInterface/
http://lists.freedesktop.org/archives/systemd-devel/2013-June/011521.html



ROBERT ALLMEROTH
Harman Automotive Division

Becker-Göring-Straße 16
76307 Karlsbad
Phone: +49 (0) 7248-71-1156
Fax: +49 (0) 7248-71-2156
Email: robert.allmer...@harman.com
Web: www.harman.com




___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

[systemd-devel] [PATCH] rules: name Dell iDRAC virtual usb nic

[Charles_Rose]

From: Charles Rose charles_r...@dell.com

Dell servers contain a virtual usb nic for systems management.
Name the device 'idrac' to easily identify it.
---
  rules/75-net-description.rules | 3 +++
  1 file changed, 3 insertions(+)

diff --git a/rules/75-net-description.rules b/rules/75-net-description.rules
index 7e62f8b..d28c9ff 100644
--- a/rules/75-net-description.rules
+++ b/rules/75-net-description.rules
@@ -5,6 +5,9 @@ SUBSYSTEM!=net, GOTO=net_end

  IMPORT{builtin}=net_id

+# Dell iDRAC Virtual USB NIC
+SUBSYSTEMS==usb, ATTRS{idVendor}==413c, ATTRS{idProduct}==a102, 
NAME=idrac
+
  SUBSYSTEMS==usb, IMPORT{builtin}=usb_id, IMPORT{builtin}=hwdb 
--subsystem=usb
  SUBSYSTEMS==usb, GOTO=net_end

-- 
1.8.5.3

--
Charles Rose
Linux Engineering
Dell Inc.
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

Re: [systemd-devel] Undefined reference efi_loader_get_boot_usec

[Cristian Rodríguez]

El 11/02/14 09:12, Allmeroth, Robert escribió:


I helped myself to #ifdef the body of boot_timestamps() in 
shared/boot_timestamp.c


Oh btw.. since you figured out before I did.. the next time please send 
patches, it does not matter if they are wrong, ifdef is misplaced or 
whatever.. just try! the worst thing that could happen is that someone 
will reject it directing you to the correct solution.

___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

Re: [systemd-devel] [PATCH] rules: name Dell iDRAC virtual usb nic

[Kay Sievers]

On Wed, Feb 12, 2014 at 2:12 PM,  charles_r...@dell.com wrote:
 From: Charles Rose charles_r...@dell.com

 Dell servers contain a virtual usb nic for systems management.
 Name the device 'idrac' to easily identify it.

The 75-net-description.rules is doe describe interfaces, not to carry
out any policy.

Product specific matches do not really belong into default udev rules
shipped by systemd, we should not add more of them.

They might have a place in the hardware database files. I'm not
exactly sure though, how to apply policy from the hwdb to renaming
network devices, there is no user like that at the moment.

What tool is supposed to find that named interface? Can't it just ship
that rule in its own file?

Kay
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

[systemd-devel] [PATCH v2 1/2] test: add basic seccomp tests

[Ronny Chevalier]

---
 test/TEST-04-SECCOMP/Makefile   |  1 +
 test/TEST-04-SECCOMP/test-seccomp.sh| 13 +
 test/TEST-04-SECCOMP/test.sh| 79 +
 test/TEST-04-SECCOMP/will-fail.service  |  8 +++
 test/TEST-04-SECCOMP/will-fail2.service |  6 +++
 test/TEST-04-SECCOMP/will-not-fail.service  |  9 
 test/TEST-04-SECCOMP/will-not-fail2.service |  6 +++
 7 files changed, 122 insertions(+)
 create mode 12 test/TEST-04-SECCOMP/Makefile
 create mode 100755 test/TEST-04-SECCOMP/test-seccomp.sh
 create mode 100755 test/TEST-04-SECCOMP/test.sh
 create mode 100644 test/TEST-04-SECCOMP/will-fail.service
 create mode 100644 test/TEST-04-SECCOMP/will-fail2.service
 create mode 100644 test/TEST-04-SECCOMP/will-not-fail.service
 create mode 100644 test/TEST-04-SECCOMP/will-not-fail2.service

diff --git a/test/TEST-04-SECCOMP/Makefile b/test/TEST-04-SECCOMP/Makefile
new file mode 12
index 000..e9f93b1
--- /dev/null
+++ b/test/TEST-04-SECCOMP/Makefile
@@ -0,0 +1 @@
+../TEST-01-BASIC/Makefile
\ No newline at end of file
diff --git a/test/TEST-04-SECCOMP/test-seccomp.sh 
b/test/TEST-04-SECCOMP/test-seccomp.sh
new file mode 100755
index 000..efe5c11
--- /dev/null
+++ b/test/TEST-04-SECCOMP/test-seccomp.sh
@@ -0,0 +1,13 @@
+#!/bin/bash -x
+
+systemctl start will-fail.service
+systemctl start will-fail2.service
+systemctl start will-not-fail.service
+systemctl start will-not-fail2.service
+systemctl is-failed will-fail.service | grep failed || exit 1
+systemctl is-failed will-fail2.service | grep failed || exit 1
+systemctl is-failed will-not-fail.service | grep failed  exit 1
+systemctl is-failed will-not-fail2.service | grep failed  exit 1
+
+touch /testok
+exit 0
diff --git a/test/TEST-04-SECCOMP/test.sh b/test/TEST-04-SECCOMP/test.sh
new file mode 100755
index 000..a85b50c
--- /dev/null
+++ b/test/TEST-04-SECCOMP/test.sh
@@ -0,0 +1,79 @@
+#!/bin/bash
+# -*- mode: shell-script; indent-tabs-mode: nil; sh-basic-offset: 4; -*-
+# ex: ts=8 sw=4 sts=4 et filetype=sh
+TEST_DESCRIPTION=seccomp tests
+
+. $TEST_BASE_DIR/test-functions
+
+check_result_qemu() {
+ret=1
+mkdir -p $TESTDIR/root
+mount ${LOOPDEV}p1 $TESTDIR/root
+[[ -e $TESTDIR/root/testok ]]  ret=0
+[[ -f $TESTDIR/root/failed ]]  cp -a $TESTDIR/root/failed $TESTDIR
+cp -a $TESTDIR/root/var/log/journal $TESTDIR
+umount $TESTDIR/root
+[[ -f $TESTDIR/failed ]]  cat $TESTDIR/failed
+ls -l $TESTDIR/journal/*/*.journal
+test -s $TESTDIR/failed  ret=$(($ret+1))
+return $ret
+}
+
+test_run() {
+if run_qemu; then
+check_result_qemu || return 1
+else
+dwarn can't run QEMU, skipping
+fi
+if check_nspawn; then
+run_nspawn
+check_result_nspawn || return 1
+else
+dwarn can't run systemd-nspawn, skipping
+fi
+return 0
+}
+
+test_setup() {
+create_empty_image
+mkdir -p $TESTDIR/root
+mount ${LOOPDEV}p1 $TESTDIR/root
+
+# Create what will eventually be our root filesystem onto an overlay
+(
+LOG_LEVEL=5
+eval $(udevadm info --export --query=env --name=${LOOPDEV}p2)
+
+setup_basic_environment
+
+# setup the testsuite service
+cat $initdir/etc/systemd/system/testsuite.service EOF
+[Unit]
+Description=Testsuite service
+After=multi-user.target
+
+[Service]
+ExecStart=/test-seccomp.sh
+Type=oneshot
+EOF
+
+# copy the units used by this test
+cp {will-fail,will-fail2,will-not-fail,will-not-fail2}.service \
+$initdir/etc/systemd/system
+cp test-seccomp.sh $initdir/
+
+setup_testsuite
+)
+setup_nspawn_root
+
+ddebug umount $TESTDIR/root
+umount $TESTDIR/root
+}
+
+test_cleanup() {
+umount $TESTDIR/root 2/dev/null
+[[ $LOOPDEV ]]  losetup -d $LOOPDEV
+return 0
+}
+
+do_test $@
diff --git a/test/TEST-04-SECCOMP/will-fail.service 
b/test/TEST-04-SECCOMP/will-fail.service
new file mode 100644
index 000..c4e0be9
--- /dev/null
+++ b/test/TEST-04-SECCOMP/will-fail.service
@@ -0,0 +1,8 @@
+[Unit]
+Description=Will fail
+
+[Service]
+ExecStart=/bin/echo This should not be seen
+SystemCallFilter=ioperm
+SystemCallFilter=~ioperm
+SystemCallFilter=ioperm
diff --git a/test/TEST-04-SECCOMP/will-fail2.service 
b/test/TEST-04-SECCOMP/will-fail2.service
new file mode 100644
index 000..1bfaeaa
--- /dev/null
+++ b/test/TEST-04-SECCOMP/will-fail2.service
@@ -0,0 +1,6 @@
+[Unit]
+Description=Will fail 2
+
+[Service]
+ExecStart=/bin/echo This should not be seen
+SystemCallFilter=~write open execve exit_group close mmap munmap fstat 
DONOTEXIST
diff --git a/test/TEST-04-SECCOMP/will-not-fail.service 
b/test/TEST-04-SECCOMP/will-not-fail.service
new file mode 100644
index 000..5c1b594
--- /dev/null
+++ b/test/TEST-04-SECCOMP/will-not-fail.service
@@ -0,0 +1,9 @@
+[Unit]
+Description=Will not fail
+
+[Service]
+ExecStart=/bin/echo Foo bar
+SystemCallFilter=~read write open execve ioperm

Re: [systemd-devel] [PATCH v3 2/2] syscallfilter: port to libseccomp

[Lennart Poettering]

On Wed, 12.02.14 01:29, Ronny Chevalier (chevalier.ro...@gmail.com) wrote:

Commited this one! Thanks!

(I added another commit on top, to add SystemCallErrorNumber= to
optionally return an error from blacklisted syscalls instead of aborting
the process)

 ---
 Ok so I found the problem. I forgot a condition in SET_FOREACH that if e is 
 0, it will stop the foreach (and read have the id 0)
 
 Here is the new patch fixing this. Sorry again
 
  Makefile.am| 30 ++---
  README |  1 +
  TODO   |  1 -
  configure.ac   | 14 ++
  man/systemd.exec.xml   | 18 +++-
  src/core/build.h   |  8 +++-
  src/core/dbus-execute.c|  8 +---
  src/core/execute.c | 88 ++
  src/core/execute.h | 12 +-
  src/core/load-fragment.c   | 96 
 +++---
  src/shared/.gitignore  |  4 --
  src/shared/linux/seccomp-bpf.h | 76 -
  src/shared/linux/seccomp.h | 47 -
  src/shared/syscall-list.c  | 56 
  src/shared/syscall-list.h  | 41 --
  src/test/test-tables.c |  3 --
  16 files changed, 157 insertions(+), 346 deletions(-)
  delete mode 100644 src/shared/linux/seccomp-bpf.h
  delete mode 100644 src/shared/linux/seccomp.h
  delete mode 100644 src/shared/syscall-list.c
  delete mode 100644 src/shared/syscall-list.h
 
 diff --git a/Makefile.am b/Makefile.am
 index 6a3fd48..1b978e5 100644
 --- a/Makefile.am
 +++ b/Makefile.am
 @@ -653,8 +653,6 @@ noinst_LTLIBRARIES += \
  libsystemd_shared_la_SOURCES = \
   src/shared/linux/auto_dev-ioctl.h \
   src/shared/linux/fanotify.h \
 - src/shared/linux/seccomp.h \
 - src/shared/linux/seccomp-bpf.h \
   src/shared/ioprio.h \
   src/shared/missing.h \
   src/shared/initreq.h \
 @@ -762,8 +760,6 @@ libsystemd_shared_la_SOURCES = \
   src/shared/net-util.h \
   src/shared/errno-list.c \
   src/shared/errno-list.h \
 - src/shared/syscall-list.c \
 - src/shared/syscall-list.h \
   src/shared/audit.c \
   src/shared/audit.h \
   src/shared/xml.c \
 @@ -771,9 +767,7 @@ libsystemd_shared_la_SOURCES = \
  
  nodist_libsystemd_shared_la_SOURCES = \
   src/shared/errno-from-name.h \
 - src/shared/errno-to-name.h \
 - src/shared/syscall-from-name.h \
 - src/shared/syscall-to-name.h
 + src/shared/errno-to-name.h
  
  # 
 --
  noinst_LTLIBRARIES += \
 @@ -998,6 +992,7 @@ libsystemd_core_la_CFLAGS = \
   $(PAM_CFLAGS) \
   $(AUDIT_CFLAGS) \
   $(KMOD_CFLAGS) \
 + $(SECCOMP_CFLAGS) \
   -pthread
  
  libsystemd_core_la_LIBADD = \
 @@ -1012,6 +1007,7 @@ libsystemd_core_la_LIBADD = \
   $(PAM_LIBS) \
   $(AUDIT_LIBS) \
   $(CAP_LIBS) \
 + $(SECCOMP_LIBS) \
   $(KMOD_LIBS)
  
  src/core/load-fragment-gperf-nulstr.c: src/core/load-fragment-gperf.gperf
 @@ -1025,33 +1021,13 @@ CLEANFILES += \
   src/core/load-fragment-gperf.gperf \
   src/core/load-fragment-gperf.c \
   src/core/load-fragment-gperf-nulstr.c \
 - src/shared/syscall-list.txt \
 - src/shared/syscall-from-name.gperf \
   src/shared/errno-list.txt \
   src/shared/errno-from-name.gperf
  
  BUILT_SOURCES += \
 - src/shared/syscall-from-name.h \
 - src/shared/syscall-to-name.h \
   src/shared/errno-from-name.h \
   src/shared/errno-to-name.h
  
 -src/shared/syscall-list.txt:
 - $(AM_V_at)$(MKDIR_P) $(dir $@)
 - $(AM_V_GEN)$(CPP) $(CFLAGS) $(AM_CPPFLAGS) $(CPPFLAGS) -dM -include 
 sys/syscall.h -  /dev/null | $(AWK) '/^#define[ \t]+__NR_[^ ]+[ \t]+[0-9(]/ 
 { sub(/__NR_/, , $$2); if ($$2 !~ /SYSCALL_BASE/) print $$2; }'  $@
 -
 -src/shared/syscall-from-name.gperf: src/shared/syscall-list.txt
 - $(AM_V_at)$(MKDIR_P) $(dir $@)
 - $(AM_V_GEN)$(AWK) 'BEGIN{ print struct syscall_name { const char* 
 name; int id; };; print %null-strings; print %%;} { printf %s, 
 __NR_%s\n, $$1, $$1 }'  $  $@
 -
 -src/shared/syscall-from-name.h: src/shared/syscall-from-name.gperf
 - $(AM_V_at)$(MKDIR_P) $(dir $@)
 - $(AM_V_GPERF)$(GPERF) -L ANSI-C -t --ignore-case -N lookup_syscall -H 
 hash_syscall_name -p -C  $  $@
 -
 -src/shared/syscall-to-name.h: src/shared/syscall-list.txt
 - $(AM_V_at)$(MKDIR_P) $(dir $@)
 - $(AM_V_GEN)$(AWK) 'BEGIN{ print static const char* const 
 syscall_names[] = { } { printf [SYSCALL_TO_INDEX(__NR_%s)] = \%s\,\n, 
 $$1, $$1 } END{print };}'  $  $@
 -
  src/shared/errno-list.txt:
   $(AM_V_at)$(MKDIR_P) $(dir $@)
   $(AM_V_GEN)$(CPP) $(CFLAGS) $(AM_CPPFLAGS) $(CPPFLAGS) -dM -include 
 errno.h -  /dev/null | $(AWK) '/^#define[ \t]+E[^ _]+[ \t]+[0-9]/ { print 
 $$2; }'   $@
 diff --git a/README b/README
 index 509b45f..a237672 100644
 --- 

[systemd-devel] Misleading InterfaceStabilityPromise wiki page (due to formatting)

[Elijah Newren]

Hi,

At http://www.freedesktop.org/wiki/Software/systemd/InterfaceStabilityPromise
(linked to from LWN just recently), the stable, almost-stable, and
not-stable lists all run together in one long bulleted list making it
appear that all categories are part of the stability promise.

I suspect a newline is needed before the two The following
interfaces sentences, so that they don't get formatted as part of the
previous bullet item.  (I don't have perms to edit, or I'd just do
it.)

Elijah
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

Re: [systemd-devel] Misleading InterfaceStabilityPromise wiki page (due to formatting)

[Kay Sievers]

On Wed, Feb 12, 2014 at 6:59 PM, Elijah Newren new...@gmail.com wrote:
 At http://www.freedesktop.org/wiki/Software/systemd/InterfaceStabilityPromise
 (linked to from LWN just recently), the stable, almost-stable, and
 not-stable lists all run together in one long bulleted list making it
 appear that all categories are part of the stability promise.

 I suspect a newline is needed before the two The following
 interfaces sentences, so that they don't get formatted as part of the
 previous bullet item.  (I don't have perms to edit, or I'd just do
 it.)

Added the newlines.

Thanks,
Kay
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

[systemd-devel] https://bugzilla.redhat.com/show_bug.cgi?id=1047614

[Reindl Harald]

https://bugzilla.redhat.com/show_bug.cgi?id=1047614

Product:Red Hat Enterprise Linux 7
Component:  systemd (Show other bugs)
Version:7.0
Hardware:   Unspecified Unspecified
Priorityurgent Severity high

first reported more than 3 months ago
https://bugzilla.redhat.com/show_bug.cgi?id=1023788

maybe systemd-upstream should consider slow down development
and spend more energy in quality and stability



signature.asc
Description: OpenPGP digital signature
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

Re: [systemd-devel] https://bugzilla.redhat.com/show_bug.cgi?id=1047614

[Greg KH]

On Wed, Feb 12, 2014 at 08:05:47PM +0100, Reindl Harald wrote:
 https://bugzilla.redhat.com/show_bug.cgi?id=1047614
 
 Product:  Red Hat Enterprise Linux 7
 Component:systemd (Show other bugs)
 Version:  7.0
 Hardware: Unspecified Unspecified
 Priority  urgent Severity high
 
 first reported more than 3 months ago
 https://bugzilla.redhat.com/show_bug.cgi?id=1023788

This issue has been debugged already, see the mailing list archives.

 maybe systemd-upstream should consider slow down development
 and spend more energy in quality and stability

How exactly would you suggest something like this be done?

greg k-h
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

Re: [systemd-devel] https://bugzilla.redhat.com/show_bug.cgi?id=1047614

[Lennart Poettering]

On Wed, 12.02.14 20:05, Reindl Harald (h.rei...@thelounge.net) wrote:

 https://bugzilla.redhat.com/show_bug.cgi?id=1047614
 
 Product:  Red Hat Enterprise Linux 7
 Component:systemd (Show other bugs)
 Version:  7.0
 Hardware: Unspecified Unspecified
 Priority  urgent Severity high
 
 first reported more than 3 months ago
 https://bugzilla.redhat.com/show_bug.cgi?id=1023788
 
 maybe systemd-upstream should consider slow down development
 and spend more energy in quality and stability

Well, firstly, it's hardly your business how we spend our time.

Secondly, this bug is fixed upstream.

Thirdly, patches count more than complaining.

Lennart

-- 
Lennart Poettering, Red Hat
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

Re: [systemd-devel] https://bugzilla.redhat.com/show_bug.cgi?id=1047614

[Reindl Harald]

Am 12.02.2014 21:19, schrieb Lennart Poettering:
 On Wed, 12.02.14 20:05, Reindl Harald (h.rei...@thelounge.net) wrote:
 
 https://bugzilla.redhat.com/show_bug.cgi?id=1047614

 Product: Red Hat Enterprise Linux 7
 Component:   systemd (Show other bugs)
 Version: 7.0
 Hardware:Unspecified Unspecified
 Priority urgent Severity high

 first reported more than 3 months ago
 https://bugzilla.redhat.com/show_bug.cgi?id=1023788

 maybe systemd-upstream should consider slow down development
 and spend more energy in quality and stability
 
 Well, firstly, it's hardly your business how we spend our time

firstly it's my business to say my opinion if i get hit by the
result daily and as i woke up that morning i saw a free world
which permits that

your opinion: i do not care about downstream

my optinion:  bad attitude in case downstream is the public
testbed of your employer with the userbase giving you the
needed feedback for work in progress on real world setups

the truth is likely somewhere between

 Secondly, this bug is fixed upstream

and that is what i call release-and-forget strategy
it has not made it in any Fedora package right now

http://koji.fedoraproject.org/koji/packageinfo?packageID=10477

and no http://koji.fedoraproject.org/koji/buildinfo?buildID=491024
does not make things better, it breaks systemd-logind and floods
logs - i had installed it a few minutes after build and reverted

so finally there is no change downstream over months

 Thirdly, patches count more than complaining

that may be true, if i not would face the same bug and
it's friends in F20 day for day i would not need to complain



signature.asc
Description: OpenPGP digital signature
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

Re: [systemd-devel] https://bugzilla.redhat.com/show_bug.cgi?id=1047614

[Tomasz Torcz]

On Wed, Feb 12, 2014 at 08:05:47PM +0100, Reindl Harald wrote:
 https://bugzilla.redhat.com/show_bug.cgi?id=1047614
 
 Product:  Red Hat Enterprise Linux 7
 Component:systemd (Show other bugs)
 Version:  7.0

  That's some downstream distro bug.  And this is Fedora mailinglist.
RHEL issues are offtopic here.

-- 
Tomasz Torcz God, root, what's the difference?
xmpp: zdzich...@chrome.pl God is more forgiving.

___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

Re: [systemd-devel] https://bugzilla.redhat.com/show_bug.cgi?id=1047614

[Reindl Harald]

my last post for this topic before i get blocked again by
disturbing development in peace with real user problems

i thought so once per month it's woth to make notice

Am 12.02.2014 21:37, schrieb Tomasz Torcz:
 On Wed, Feb 12, 2014 at 08:05:47PM +0100, Reindl Harald wrote:
 https://bugzilla.redhat.com/show_bug.cgi?id=1047614

 Product: Red Hat Enterprise Linux 7
 Component:   systemd (Show other bugs)
 Version: 7.0
 
 That's some downstream distro bug

this is *the* downstream distro
this is *the* employer paing systemd project-leader

 And this is Fedora mailinglist
 RHEL issues are offtopic here

that's new, sadly not true, otherwise the bugreport
long before RHEL7-Beta would have prevented to make
it in Fedora 20 GA and stay there

first reported more than 3 months ago
https://bugzilla.redhat.com/show_bug.cgi?id=1023788



signature.asc
Description: OpenPGP digital signature
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

Re: [systemd-devel] https://bugzilla.redhat.com/show_bug.cgi?id=1047614

[David Timothy Strauss]

On Wed, Feb 12, 2014 at 12:37 PM, Tomasz Torcz to...@pipebreaker.pl wrote:
 That's some downstream distro bug.  And this is Fedora mailinglist.
 RHEL issues are offtopic here.

This is not a Fedora mailing list at all.
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

Re: [systemd-devel] https://bugzilla.redhat.com/show_bug.cgi?id=1047614

[Tomasz Torcz]

On Wed, Feb 12, 2014 at 12:45:50PM -0800, David Timothy Strauss wrote:
 On Wed, Feb 12, 2014 at 12:37 PM, Tomasz Torcz to...@pipebreaker.pl wrote:
  That's some downstream distro bug.  And this is Fedora mailinglist.
  RHEL issues are offtopic here.
 
 This is not a Fedora mailing list at all.

  Ouch.  I'm sorry.  I thought I was still reading fedora-devel.
My apologies to Harald and other readers of systemd-devel.

-- 
Tomasz Torcz God, root, what's the difference?
xmpp: zdzich...@chrome.pl God is more forgiving.



pgp8_mNKZFinb.pgp
Description: PGP signature
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

Re: [systemd-devel] https://bugzilla.redhat.com/show_bug.cgi?id=1047614

[Jóhann B. Guðmundsson]

On 02/12/2014 07:37 PM, Reindl Harald wrote:

these are all regressions from F19 to F20
https://bugzilla.redhat.com/show_bug.cgi?id=1023820
https://bugzilla.redhat.com/show_bug.cgi?id=1010572
https://bugzilla.redhat.com/show_bug.cgi?id=1057811
https://bugzilla.redhat.com/show_bug.cgi?id=1057618
https://bugzilla.redhat.com/show_bug.cgi?id=1023788

sometimes it feels like systemd-upstrearm has a release-and-forget-strategy
and don't care about the downstream distributions and issues there nor
how bad the impact of them for how long is


I'm not so sure you are aware of this but the major part of systemd 
maintenance in Fedora is and has been handled by Zbyszek for the past 
year which is doing that on his own free time and before him it was 
Michal who is a Red Hat employee but working on completely different 
thing then systemd and he was also doing systemd downstream distribution 
maintenance on his own free time.


Both of them have been doing good job of maintaining systemd in the 
distribution and been extremely conservative in their maintainership so 
I suggest you keep your tone down towards people contributing their own 
free time to the project.


With regards to those RHEL bugs I suggest you follow Red Hat's official 
procedure dealing with those and file a support case after all that's 
what you are paying them for.


Ranting on upstream mailinglist will get you nowhere regardless if that 
upstream is partially on Red Hat's payroll or not...


JBG
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

Re: [systemd-devel] https://bugzilla.redhat.com/show_bug.cgi?id=1047614

[Reindl Harald]

Am 12.02.2014 22:41, schrieb Jóhann B. Guðmundsson:
 On 02/12/2014 07:37 PM, Reindl Harald wrote:
 these are all regressions from F19 to F20
 https://bugzilla.redhat.com/show_bug.cgi?id=1023820
 https://bugzilla.redhat.com/show_bug.cgi?id=1010572
 https://bugzilla.redhat.com/show_bug.cgi?id=1057811
 https://bugzilla.redhat.com/show_bug.cgi?id=1057618
 https://bugzilla.redhat.com/show_bug.cgi?id=1023788

 sometimes it feels like systemd-upstrearm has a release-and-forget-strategy
 and don't care about the downstream distributions and issues there nor
 how bad the impact of them for how long is
 
 I'm not so sure you are aware of this but the major part of systemd 
 maintenance in 
 Fedora is and has been handled by Zbyszek for the past year which is doing 
 that on 
 his own free time and before him it was Michal who is a Red Hat employee but 
 working 
 on completely different thing then systemd and he was also doing systemd 
 downstream
 distribution maintenance on his own free time.

i am aware

the major point of my critism is systemd-upstream's release-and-forget
with no point releases at all and the state of systemd in F20 makes that
not looking like a working strategy overstraining downstream maintainers

that may work for a browser but not for a core-component of a distruibution
or only in case of luck and not hitted by bugs as we have them now

 Both of them have been doing good job of maintaining systemd in the 
 distribution 
 and been extremely conservative in their maintainership 

extremely conservative would have been keep Fedora 20 on systemd-204
which has none of that problems - systemd-208 obviously has *not*
release quality

extremely conservative with a stable release: fine
extremely conservative with a broken release: not fine, it needs fixes, timly

https://bugzilla.redhat.com/show_bug.cgi?id=1023788
reported:   2013-10-27
GA release: 2013-12-17
current:2014-02-12

 so I suggest you keep your tone down towards people contributing their 
 own free time to the project.

my tone is ok, if someone can't stand critism that's a different problem layer

i am contributing *my own free time* with testing and bugreports, the last
systemd build for F20 was tested, rolled back and commented with feedback
minutes after the build completed, so i deserve feedback in a timely manner
and not get ignored over months (and i am not the only one)

since you are the one fighting against RH on fedora-devel and claiming you are
Fedora QA more than once - where where you in case of the bugs above? how did
they pass QA if any random tester is facing two of them within minutes?

 With regards to those RHEL bugs I suggest you follow Red Hat's official 
 procedure dealing with those

RHEL bugs? maybe read them before make comments and get rid of your hate
against Redhat could help, the Fedora bugzilla is below redhat.com and they
are all *Fedora* bugs and *one* of them made it to RHEL-Beta *too*



signature.asc
Description: OpenPGP digital signature
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

Re: [systemd-devel] https://bugzilla.redhat.com/show_bug.cgi?id=1047614

[Greg KH]

On Wed, Feb 12, 2014 at 11:12:07PM +0100, Reindl Harald wrote:
 
 the major point of my critism is systemd-upstream's release-and-forget
 with no point releases at all and the state of systemd in F20 makes that
 not looking like a working strategy overstraining downstream maintainers
 
 that may work for a browser but not for a core-component of a distruibution
 or only in case of luck and not hitted by bugs as we have them now

I guess the Linux kernel is not a core-component of a distro then, by
your definition :)

My distro doesn't have this issue, if you have a problem with the distro
you are currently using, I suggest talking with them about this, or
change to a different one.  No one is forcing you to use Fedora...

good luck,

greg k-h
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

[systemd-devel] [PATCH 0/3] Password agent fixes

[David Härdeman]

The following series contains patch 1 - 3 of the previous patchset that
I've posted. Lennart wanted a single Id= instead of the two Purpose=
and Target= strings in the .ask files, so I've changed it accordingly.

The two following patches are just small fixes.

I've left out the binary protocol patch for now.

---

David Härdeman (3):
  Add more password agent information
  Fix keysize handling in cryptsetup (bits vs. bytes)
  Fix askpass buffer overflow


 src/ask-password/ask-password.c |   14 +++---
 src/cryptsetup/cryptsetup.c |   19 +++
 src/shared/ask-password-api.c   |   14 --
 src/shared/ask-password-api.h   |6 --
 4 files changed, 42 insertions(+), 11 deletions(-)

-- 
David Härdeman
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

[systemd-devel] [PATCH 3/3] Fix askpass buffer overflow

[David Härdeman]

Fix askpass overflow in reading a passphrase from a tty.
Doesn't seem security sensitive, but add a check for correctness.
---
 src/shared/ask-password-api.c |5 +
 1 file changed, 5 insertions(+)

diff --git a/src/shared/ask-password-api.c b/src/shared/ask-password-api.c
index 1c18274..045cff2 100644
--- a/src/shared/ask-password-api.c
+++ b/src/shared/ask-password-api.c
@@ -213,6 +213,11 @@ int ask_password_tty(
 loop_write(ttyfd, *, 1, false);
 
 dirty = true;
+
+   if (p = (sizeof(passphrase) - 1)) {
+   loop_write(ttyfd, \n, 1, false);
+   break;
+   }
 }
 }
 

___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

[systemd-devel] [PATCH 1/3] Add more password agent information

[David Härdeman]

Add an (optional) Id key in the password agent .ask files. The Id is
supposed to be a simple string in subsystem:target form which
is used to provide more information on what the requested passphrase
is to be used for (which e.g. allows an agent to only react to cryptsetup
requests).
---
 src/ask-password/ask-password.c |   14 +++---
 src/cryptsetup/cryptsetup.c |   11 +--
 src/shared/ask-password-api.c   |9 +++--
 src/shared/ask-password-api.h   |6 --
 4 files changed, 31 insertions(+), 9 deletions(-)

diff --git a/src/ask-password/ask-password.c b/src/ask-password/ask-password.c
index ea0c623..f939eab 100644
--- a/src/ask-password/ask-password.c
+++ b/src/ask-password/ask-password.c
@@ -43,6 +43,7 @@
 #include def.h
 
 static const char *arg_icon = NULL;
+static const char *arg_id = NULL;
 static const char *arg_message = NULL;
 static bool arg_use_tty = true;
 static usec_t arg_timeout = DEFAULT_TIMEOUT_USEC;
@@ -58,7 +59,8 @@ static int help(void) {
 --timeout=SEC   Timeout in sec\n
 --no-ttyAsk question via agent even on TTY\n
 --accept-cached Accept cached passwords\n
---multiple  List multiple passwords if available\n,
+--multiple  List multiple passwords if available\n
+--id=ID Query identifier (e.g. 
cryptsetup:/dev/sda5)\n,
program_invocation_short_name);
 
 return 0;
@@ -71,7 +73,8 @@ static int parse_argv(int argc, char *argv[]) {
 ARG_TIMEOUT,
 ARG_NO_TTY,
 ARG_ACCEPT_CACHED,
-ARG_MULTIPLE
+ARG_MULTIPLE,
+ARG_ID
 };
 
 static const struct option options[] = {
@@ -81,6 +84,7 @@ static int parse_argv(int argc, char *argv[]) {
 { no-tty,no_argument,   NULL, ARG_NO_TTY
},
 { accept-cached, no_argument,   NULL, ARG_ACCEPT_CACHED 
},
 { multiple,  no_argument,   NULL, ARG_MULTIPLE  
},
+{ id,required_argument, NULL, ARG_ID
},
 {}
 };
 
@@ -119,6 +123,10 @@ static int parse_argv(int argc, char *argv[]) {
 arg_multiple = true;
 break;
 
+   case ARG_ID:
+   arg_id = optarg;
+break;
+
 case '?':
 return -EINVAL;
 
@@ -162,7 +170,7 @@ int main(int argc, char *argv[]) {
 } else {
 char **l;
 
-if ((r = ask_password_agent(arg_message, arg_icon, timeout, 
arg_accept_cached, l)) = 0) {
+if ((r = ask_password_agent(arg_message, arg_icon, arg_id, 
timeout, arg_accept_cached, l)) = 0) {
 char **p;
 
 STRV_FOREACH(p, l) {
diff --git a/src/cryptsetup/cryptsetup.c b/src/cryptsetup/cryptsetup.c
index 033c0cd..f72cf9f 100644
--- a/src/cryptsetup/cryptsetup.c
+++ b/src/cryptsetup/cryptsetup.c
@@ -260,6 +260,7 @@ static int get_password(const char *name, usec_t until, 
bool accept_cached, char
 int r;
 char **p;
 _cleanup_free_ char *text = NULL;
+_cleanup_free_ char *id = NULL;
 
 assert(name);
 assert(passwords);
@@ -267,7 +268,10 @@ static int get_password(const char *name, usec_t until, 
bool accept_cached, char
 if (asprintf(text, Please enter passphrase for disk %s!, name)  0)
 return log_oom();
 
-r = ask_password_auto(text, drive-harddisk, until, accept_cached, 
passwords);
+   if (asprintf(id, cryptsetup:%s, name)  0)
+   return log_oom();
+
+r = ask_password_auto(text, drive-harddisk, id, until, 
accept_cached, passwords);
 if (r  0) {
 log_error(Failed to query password: %s, strerror(-r));
 return r;
@@ -281,7 +285,10 @@ static int get_password(const char *name, usec_t until, 
bool accept_cached, char
 if (asprintf(text, Please enter passphrase for disk %s! 
(verification), name)  0)
 return log_oom();
 
-r = ask_password_auto(text, drive-harddisk, until, false, 
passwords2);
+   if (asprintf(id, cryptsetup-verification:%s, name)  0)
+   return log_oom();
+
+r = ask_password_auto(text, drive-harddisk, id, until, 
false, passwords2);
 if (r  0) {
 log_error(Failed to query verification password: %s, 
strerror(-r));
 return r;
diff --git a/src/shared/ask-password-api.c b/src/shared/ask-password-api.c
index a328f14..1c18274 100644
--- a/src/shared/ask-password-api.c
+++ b/src/shared/ask-password-api.c
@@ -298,6 +298,7 @@ fail:
 int ask_password_agent(
 const char *message,

[systemd-devel] [PATCH 2/3] Fix keysize handling in cryptsetup (bits vs. bytes)

[David Härdeman]

The command line key-size is in bits but the libcryptsetup API expects bytes.
---
 src/cryptsetup/cryptsetup.c |8 ++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/src/cryptsetup/cryptsetup.c b/src/cryptsetup/cryptsetup.c
index f72cf9f..e7e8066 100644
--- a/src/cryptsetup/cryptsetup.c
+++ b/src/cryptsetup/cryptsetup.c
@@ -414,7 +414,7 @@ static int attach_luks_or_plain(struct crypt_device *cd,
 /* for CRYPT_PLAIN limit reads
  * from keyfile to key length, and
  * ignore keyfile-size */
-opt_keyfile_size = opt_key_size / 8;
+opt_keyfile_size = opt_key_size;
 
 /* In contrast to what the name
  * crypt_setup() might suggest this
@@ -577,7 +577,11 @@ int main(int argc, char *argv[]) {
 else
 until = 0;
 
-opt_key_size = (opt_key_size  0 ? opt_key_size : 256);
+if (opt_key_size % 8) {
+log_warning(Key size invalid (not a multiple of 8).);
+goto finish;
+}
+opt_key_size = (opt_key_size  0 ? opt_key_size : 256) / 8;
 
 if (key_file) {
 struct stat st;

___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

Re: [systemd-devel] How to prevent sleep during running oneshot units

[Kai Krakow]

Zbigniew Jędrzejewski-Szmek zbys...@in.waw.pl schrieb:

 No, you need to wrap your command with it. I.e. instead of

 ExecStart=/path/to/backup-command ...

 you use

 ExecStart=/usr/bin/systemd-inhibit /path/to/backup-command ...
 Right. Not too bad, but Inhibits=sleep sounds nice.

I probably try that, thou I'd prefer an extra configuration option for it 
because in the jobs monitor it would show me systemd-inhibit as the running 
process instead of my backup script.

 What a pity. Are there any deeper plans on that one? I think it should
 not be too difficult to implement and if I had some time I would try it.
 But I don't like inventing all the complex surrounding this may bring in,
 read as: Is there a list / brainstorming of what one had to consider and
 take care of? While documentation of systemd is very good and easy to
 follow, I find it difficult to find documentation of implementation
 details and feature plans (yes, there are the blogs but those seem to
 cover thoughts about already implemented features)...
 Yes, it seems that basically changing the timer type would do the trick.
 One thing which worries me, is the issue the laptop in the bag issue.
 But maybe we would just need to document the fact that setting the timer
 to need to make sure that the computer will not be in hostile surrounding.
 OTOH, it might make sense to add a new inhibition: 'alarms',
 which you could set *before* suspending, which would prevent
 CLOCK_REALTIME_ALARM timers from firing.

Your laptop in a bag example is that kind of brainstorming I was thinking 
of. Of course, such an option to resume from sleep should always be opt-in, 
never opt-out. So, if we had such a new timer type, systemd should have a 
switch to actually behave this way and default to fall-back behavior (not 
waking the system). That way, an energy profile in desktop environment could 
control this switch: Set your profile to desktop and it will opt-in to 
wake timers, set it to laptop or mobile, and it won't opt-in. If battery 
is below some threshold it sould also not opt-in. My use case is a desktop 
PC so I didn't think of such issues.

I don't like the fact to solely rely on an inhibition alarms because that 
would means it'd be opt-out only. In my opinion it needs to be opt-in.

-- 
Replies to list only preferred.

___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

Re: [systemd-devel] How to prevent sleep during running oneshot units

[Kai Krakow]

Mirco Tischler mircotisch...@gmx.net schrieb:

 It's late here so excuse me if I'm missing something. But wouldn't a
 simple Before=sleep.target suffice to implement b)?

I'm not sure... The unit is not enabled, only triggered by the timer - so it 
may work. I think I may try that as an alternative to systemd-inhibit. 
Systemd-inhibit just does not feel like the right tool for this.

 If your backup job is still running it blocks sleep.target's start until
 it finishes.

Then, what happens if I start using my system during backup, sleep.target 
had been triggered but is waiting for the backup job to finish? Will it 
simply go to sleep then while I am using the system? OTOH, will systemd-
inhibit cover that case?

Twisted around: What happens if sleep is triggered by the system due to idle 
time while sleep is inhibited? Will it trigger as soon as the inhibit is 
removed again? Or will the idle timer for my system just trigger sleep again 
after it has passed a second time? 

 Though for the grace period some ugly hack such as
 ExecStartPost=/usr/bin/sleep n would be required.

My primary concern here is btrfs doing a lot of background work when 
handling snapshots. My backup medium is a btrfs with rsync copying the 
source as in-place deltas to a scratch area, then taking a snapshot of it. 
Later, I'd like to automatically cleanup old snapshots - but that means I 
should wait before going back to sleep until btrfs finished that job. A 
simple sleep probably won't do. I suppose it would require an option like 
--wait to btrfs subvolume delete, then wait another 5 minutes to give 
all pending write commits a chance to make it to disk.

-- 
Replies to list only preferred.

___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

Re: [systemd-devel] How to prevent sleep during running oneshot units

[Mirco Tischler]

Am 13.02.2014 00:14 schrieb Kai Krakow hurikha...@gmail.com:

 Mirco Tischler mircotisch...@gmx.net schrieb:

  It's late here so excuse me if I'm missing something. But wouldn't a
  simple Before=sleep.target suffice to implement b)?

 I'm not sure... The unit is not enabled, only triggered by the timer - so
it
 may work. I think I may try that as an alternative to systemd-inhibit.
 Systemd-inhibit just does not feel like the right tool for this.

  If your backup job is still running it blocks sleep.target's start until
  it finishes.

 Then, what happens if I start using my system during backup, sleep.target
 had been triggered but is waiting for the backup job to finish? Will it
 simply go to sleep then while I am using the system? OTOH, will systemd-
 inhibit cover that case?

Yeah, the job to start sleep.target will still be active, so this should
only work for overnight stuff.
 Twisted around: What happens if sleep is triggered by the system due to
idle
 time while sleep is inhibited? Will it trigger as soon as the inhibit is
 removed again? Or will the idle timer for my system just trigger sleep
again
 after it has passed a second time?

Systemd-inhibit has the options block and delay for this. But when and if
the idle time gets restarted is up to the idle timer ;-)
  Though for the grace period some ugly hack such as
  ExecStartPost=/usr/bin/sleep n would be required.

 My primary concern here is btrfs doing a lot of background work when
 handling snapshots. My backup medium is a btrfs with rsync copying the
 source as in-place deltas to a scratch area, then taking a snapshot of it.
 Later, I'd like to automatically cleanup old snapshots - but that means I
 should wait before going back to sleep until btrfs finished that job. A
 simple sleep probably won't do. I suppose it would require an option
like
 --wait to btrfs subvolume delete, then wait another 5 minutes to give
 all pending write commits a chance to make it to disk.
Sounds like you want systemd-inhibit for both your backup service and the
btrfs subvolume delete 'service'. And the kernel will call sync() before
suspending, shouldn't that suffice?

 --
 Replies to list only preferred.
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

Re: [systemd-devel] [PATCH 1/2] test: add basic seccomp tests

[Dave Reisner]

On Thu, Jan 23, 2014 at 01:34:57AM +0100, Ronny Chevalier wrote:
 ---
  test/TEST-04-SECCOMP/Makefile   |  1 +
  test/TEST-04-SECCOMP/test-seccomp.sh| 11 
  test/TEST-04-SECCOMP/test.sh| 79 
 +
  test/TEST-04-SECCOMP/will-fail.service  |  6 +++
  test/TEST-04-SECCOMP/will-not-fail.service  |  6 +++
  test/TEST-04-SECCOMP/will-not-fail2.service |  6 +++
  6 files changed, 109 insertions(+)
  create mode 12 test/TEST-04-SECCOMP/Makefile
  create mode 100755 test/TEST-04-SECCOMP/test-seccomp.sh
  create mode 100755 test/TEST-04-SECCOMP/test.sh
  create mode 100644 test/TEST-04-SECCOMP/will-fail.service
  create mode 100644 test/TEST-04-SECCOMP/will-not-fail.service
  create mode 100644 test/TEST-04-SECCOMP/will-not-fail2.service
 
 diff --git a/test/TEST-04-SECCOMP/Makefile b/test/TEST-04-SECCOMP/Makefile
 new file mode 12
 index 000..e9f93b1
 --- /dev/null
 +++ b/test/TEST-04-SECCOMP/Makefile
 @@ -0,0 +1 @@
 +../TEST-01-BASIC/Makefile
 \ No newline at end of file
 diff --git a/test/TEST-04-SECCOMP/test-seccomp.sh 
 b/test/TEST-04-SECCOMP/test-seccomp.sh
 new file mode 100755
 index 000..fef334e
 --- /dev/null
 +++ b/test/TEST-04-SECCOMP/test-seccomp.sh
 @@ -0,0 +1,11 @@
 +#!/bin/bash -x
 +
 +systemctl start will-fail.service
 +systemctl start will-not-fail.service
 +systemctl start will-not-fail2.service
 +systemctl is-failed will-fail.service | grep failed || exit 1
 +systemctl is-failed will-not-fail.service | grep failed  exit 1
 +systemctl is-failed will-not-fail2.service | grep failed  exit 1

This is weird. You should be able to rely on the exit code rather than
parsing the output, but it seems this was broken in e3e0314b.

 +
 +touch /testok
 +exit 0
 diff --git a/test/TEST-04-SECCOMP/test.sh b/test/TEST-04-SECCOMP/test.sh
 new file mode 100755
 index 000..c29192e
 --- /dev/null
 +++ b/test/TEST-04-SECCOMP/test.sh
 @@ -0,0 +1,79 @@
 +#!/bin/bash
 +# -*- mode: shell-script; indent-tabs-mode: nil; sh-basic-offset: 4; -*-
 +# ex: ts=8 sw=4 sts=4 et filetype=sh
 +TEST_DESCRIPTION=seccomp tests
 +
 +. $TEST_BASE_DIR/test-functions
 +
 +check_result_qemu() {
 +ret=1
 +mkdir -p $TESTDIR/root
 +mount ${LOOPDEV}p1 $TESTDIR/root
 +[[ -e $TESTDIR/root/testok ]]  ret=0
 +[[ -f $TESTDIR/root/failed ]]  cp -a $TESTDIR/root/failed $TESTDIR
 +cp -a $TESTDIR/root/var/log/journal $TESTDIR
 +umount $TESTDIR/root
 +[[ -f $TESTDIR/failed ]]  cat $TESTDIR/failed
 +ls -l $TESTDIR/journal/*/*.journal
 +test -s $TESTDIR/failed  ret=$(($ret+1))
 +return $ret
 +}
 +
 +test_run() {
 +if run_qemu; then
 +check_result_qemu || return 1
 +else
 +dwarn can't run QEMU, skipping
 +fi
 +if check_nspawn; then
 +run_nspawn
 +check_result_nspawn || return 1
 +else
 +dwarn can't run systemd-nspawn, skipping
 +fi
 +return 0
 +}
 +
 +test_setup() {
 +create_empty_image
 +mkdir -p $TESTDIR/root
 +mount ${LOOPDEV}p1 $TESTDIR/root
 +
 +# Create what will eventually be our root filesystem onto an overlay
 +(
 +LOG_LEVEL=5
 +eval $(udevadm info --export --query=env --name=${LOOPDEV}p2)
 +
 +setup_basic_environment
 +
 +# setup the testsuite service
 +cat $initdir/etc/systemd/system/testsuite.service EOF
 +[Unit]
 +Description=Testsuite service
 +After=multi-user.target
 +
 +[Service]
 +ExecStart=/test-seccomp.sh
 +Type=oneshot
 +EOF
 +
 +# copy the units used by this test
 +cp {will-fail,will-not-fail,will-not-fail2}.service \
 +$initdir/etc/systemd/system
 +cp test-seccomp.sh $initdir/
 +
 +setup_testsuite
 +)
 +setup_nspawn_root
 +
 +ddebug umount $TESTDIR/root
 +umount $TESTDIR/root
 +}
 +
 +test_cleanup() {
 +umount $TESTDIR/root 2/dev/null
 +[[ $LOOPDEV ]]  losetup -d $LOOPDEV
 +return 0
 +}
 +
 +do_test $@
 diff --git a/test/TEST-04-SECCOMP/will-fail.service 
 b/test/TEST-04-SECCOMP/will-fail.service
 new file mode 100644
 index 000..18e034e
 --- /dev/null
 +++ b/test/TEST-04-SECCOMP/will-fail.service
 @@ -0,0 +1,6 @@
 +[Unit]
 +Description=Will fail
 +
 +[Service]
 +ExecStart=/bin/echo This should not be seen
 +SystemCallFilter=ioperm
 diff --git a/test/TEST-04-SECCOMP/will-not-fail.service 
 b/test/TEST-04-SECCOMP/will-not-fail.service
 new file mode 100644
 index 000..c56797f
 --- /dev/null
 +++ b/test/TEST-04-SECCOMP/will-not-fail.service
 @@ -0,0 +1,6 @@
 +[Unit]
 +Description=Will not fail
 +
 +[Service]
 +ExecStart=/bin/echo Foo bar
 +SystemCallFilter=~ioctl
 diff --git a/test/TEST-04-SECCOMP/will-not-fail2.service 
 b/test/TEST-04-SECCOMP/will-not-fail2.service
 new file mode 100644
 index 000..2df05e3
 --- /dev/null
 +++ b/test/TEST-04-SECCOMP/will-not-fail2.service
 @@ -0,0 +1,6 @@
 +[Unit]
 +Description=Reset SystemCallFilter
 +
 +[Service]
 +ExecStart=/bin/echo Foo bar