Previously, the logging sockets were asynchronous and if clogged we'd
lose messages. We did this to be extra careful given that PID 1 might
need to spawn the logging daemon as response to PID 1's own log messages
and we really should avoid a deadlock in that case.
As it turns out this causes loss of too many messages, hence make the
socket blocking again, however put a time limit on it to avoid unbounded
deadlocks in the unlikely case they happen.
Tom Gundersen [Mon, 22 Jul 2013 14:59:26 +0000 (16:59 +0200)]
systemd-tmpfiles-setup-dev: remain after exit
Without this, tmpfiles-setpu-dev would be re-run if any other service,
which pulls in basic.target, was started after setup-dev was finished
and before basic.target was active.
Kay Sievers [Sat, 20 Jul 2013 12:29:12 +0000 (14:29 +0200)]
rules: drivers - always call kmod, even when a driver is bound to the device
On Sat, Jul 20, 2013 at 12:56 PM, Rafael J. Wysocki <rjw@sisk.pl> wrote:
> After a recent change present in 3.11-rc1 there is a driver, called processor,
> that can be bound to the CPU devices whose sysfs directories are located under
> /sys/devices/system/cpu/. A side effect of this is that, after the driver has
> been bound to those devices, the kernel adds DRIVER=processor to ENV for CPU
> uevents and they don't match the default rule for autoloading modules matching
> MODALIAS:
>
> DRIVER!="?*", ENV{MODALIAS}=="?*", IMPORT{builtin}="kmod load $env{MODALIAS}"
>
> any more. However, there are some modules whose module aliases match specific
> CPU features through the modalias string and those modules should be loaded
> automatically if a compatible CPU is present. Yet, with the processor driver
> bound to the CPU devices the above rule is not sufficient for that, so we need
> a new default udev rule allowing those modules to be autoloaded even if the
> CPU devices have drivers.
make: Automake is complaining about .PRECIOUS being redefined
Yesterday I added test-suite.log as dependency to the .PRECIOUS
target. Automake is warning about this target being redefined
and from what I see there is no way I can stop the warning but
I can add the %MAKEFILE% as dependency.
automake warning:
Makefile.am:35: warning: user target '.PRECIOUS' defined here ...
/usr/share/automake-1.13/am/configure.am: ... overrides Automake target '.PRECIOUS' defined here
[zj: s/%MAKEFILE%/Makefile/ because %MAKEFILE% wasn't actually substituted properly.]
journal: Leave server_dispatch_message early when Storage is none
When using Storage=none there is no point in collecting all the
information just to throw them away. After this change journald
consumes a lot less CPU time when only forwarding messages.
test: Keep the test-suite.log around in case of a test failure
The addition of .DELETE_ON_ERROR will lead to the removal of the
test-suite.log in case of a test failure. Mark the rule as PRECIOUS
to keep that file around.
systemd,systemctl: export condition status and show failing condition
$ systemctl --user status hoohoo
hoohoo.service
Loaded: loaded (/home/zbyszek/.config/systemd/user/hoohoo.service; static)
Active: inactive (dead)
start condition failed at Tue 2013-06-25 18:08:42 EDT; 1s ago
ConditionPathExists=/tmp/hoo was not met
Full information is exported over D-Bus:
[(condition, trigger, negate, param, state),...]
where state is one of "failed" (<0), "untested" (0), "OK" (>0).
I've decided to use 0 for "untested", because it might be useful to
differentiate different types of failure later on, without breaking
compatibility.
systemctl shows the failing condition, if there was a non-trigger
failing condition, or says "none of the trigger conditions were met",
because there're often many trigger conditions, and they must all
fail for the condition to fail, so printing them all would consume
a lot of space, and bring unnecessary attention to something that is
quite low-level.
Instead of :-0, :1, :5, etc., use -0, 1 or +1, 5, etc. For BOOT_ID+OFFSET,
use BOOT_ID+offset or BOOT_ID-offset (either + or - is required).
Also make error handling a bit more robust and verbose.
Modify the man page to describe the most common case (-b) first,
and the second most common case (-b -1) second.
systemctl: also highlight a load state of "not-found" as red
"not-found" is a recently added load state and was previously just a
special case of "error". Since it also indicates a load error we should
also highlight it red, the same way as "error" was treated before.
Tom Gundersen [Sun, 7 Jul 2013 19:29:12 +0000 (21:29 +0200)]
logind: apply ACL's to "dead" device nodes
Based on a patch by Kay Sievers.
When a dead device nodes is tagged with "uaccess" using the static_node mechanism,
it's ACL's are managed by logind in the same way as "live" device nodes.
This allows in particular /dev/snd/{seq,timer} to cause modules to be loaded
on-demand when accessed by a non-privileged user.
test-tables: allow sparse tables and check mapping for -1
Jan: test-tables fails on my system. The one it's failing on is:
syscall: 222 → (null) → -1
... and indeed, our own tables should not have holes, but syscall
tables certainly might.
Jason St. John [Tue, 16 Jul 2013 08:20:03 +0000 (10:20 +0200)]
man: improve readability of --output options in journalctl(1)
The list and descriptions of valid output options was difficult to read,
so break up the long block of text into discrete man page list items to
improve readability.
Jan Janssen [Fri, 28 Jun 2013 15:26:30 +0000 (17:26 +0200)]
journalctl: Add support for showing messages from a previous boot
Hi,
I redid the boot ID look up to use enumerate_unique.
This is quite fast if the cache is warm but painfully slow if
it isn't. It has a slight chance of returning the wrong order if
realtime clock jumps around.
This one has to do n searches for every boot ID there is plus
a sort, so it depends heavily on cache hotness. This is in contrast
to the other way of look-up through filtering by a MESSAGE_ID,
which only needs about 1 seek + whatever amount of relative IDs
you want to walk.
I also have a linked-list + (in-place) mergesort version of this
patch, which has pretty much the same runtime. But since this one
is using libc sorting and armortized allocation, I prefer this
one.
To summarize: The MESSAGE_ID way is a *lot* faster but can be
incomplete due to rotation, while the enumerate+sort will find
every boot ID out there but will be painfully slow for large
journals and cold caches.