summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)Author
2015-08-16Merge pull request #908 from richardmaw-codethink/nspawn-path-escapes-v3Lennart Poettering
Allow arbitrary file paths to be passed to nspawn (v3)
2015-08-16Merge pull request #955 from poettering/resolved-localhost-xyzTom Gundersen
resolved: enable synthesizing of A, AAAA and PTR records for "localhost"
2015-08-16Merge pull request #954 from poettering/nss-myhostname-loopbackTom Gundersen
nss-myhostname: use LOOPBACK_IFINDEX instead of if_nametoindex("lo")
2015-08-16resolve-host: enable auto start of resolvedLennart Poettering
There's no reason to explicitly turn off bus activation for resolved here. The reason this was done before was that the code was copied from nss-resolve, which has a fallback to glibc's nss-dns if resolved is not reachable. However, such a logic makes no sense for resolve-host since such a fallback doesn't make sense here, which means we can actually turn on activation. Let's do it hence.
2015-08-16Merge pull request #932 from kaysievers/busLennart Poettering
sd-bus: do not connect to dbus-1 socket when kdbus is available
2015-08-14resolved: never allow routing of "localhost" queries to DNS or LLMNRLennart Poettering
We should never allow leaking of "localhost" queries onto the network, even if there's an explicit domain rotue set for this.
2015-08-14resolved: locally synthesize replies for "localhost"Lennart Poettering
Let's make sure that clients querying resolved via the bus for A, AAAA or PTR records for "localhost" get a synthesized, local reply, so that we do not hit the network. This makes part of nss-myhostname redundant, if used in conjunction. However, given that nss-resolve shall be optional we need to keep this code in both places for now.
2015-08-14nss-myhostname: use LOOPBACK_IFINDEX instead of if_nametoindex("lo")Lennart Poettering
Given that we already hardocde the loopback ifindex, following the kernel's own logic, we can replace the invocation of if_nametoindex("lo") with LOOPBACK_IFINDEX.
2015-08-13logind: actually fail on OOMThomas Hindoe Paaboel Andersen
Since dacd6cee76a08331b8c8616c5f30f70ee49aa2f9 the two OOM's are ignored as the value of r will be overwritten and we only log in the fail section anyway. This patch jumps to fail on OOM. Note that this is different behavior compared to both the current code and previous to dacd6cee76a08331b8c8616c5f30f70ee49aa2f9. Before that commit we would log that saving the inhibit data failed, but still write the file, though without the WHO/WHY section. CID# 1313545
2015-08-13Merge pull request #907 from keszybz/sd-daemon-badfLennart Poettering
sd-daemon: return EBADF for invalid fd numbers
2015-08-12Merge pull request #938 from tblume/fix-kexec-force-rebootKay Sievers
support reboot -f for kexec kernel
2015-08-12Merge pull request #939 from karelzak/smatchTom Gundersen
treewide: trivial issues detected by smatch
2015-08-12support reboot -f for kexec kernelThomas Blume
Fix error message: -->-- Code should not be reached 'Unknown action.' at src/systemctl/systemctl.c:6382, function halt_now(). Aborting. Aborted --<-- when executing 'reboot -f' from a system running a kexec kernel.
2015-08-11gpt-auto-generator: warn on ambiguous blkid probeTom Gundersen
2015-08-11gpt-auto-generator: don't warn on !ENABLE_EFITom Gundersen
add_automount() was only used on EFI systems, compile it conditionally to avoid the warning.
2015-08-11 sd-bus: do not connect to dbus-1 socket when kdbus is availableKay Sievers
We should not fall back to dbus-1 and connect to the proxy when kdbus returns an error that indicates that kdbus is running but just does not accept new connections because of quota limits or something similar. Using is_kdbus_available() in libsystemd/ requires it to move from shared/ to libsystemd/. Based on a patch from David Herrmann: https://github.com/systemd/systemd/pull/886
2015-08-11gpt-auto-generator: apply partition-type flags only to specific partition-typesKay Sievers
The partition-type flags are defined independently for every partition-type. Apply them only to the types where they are defined, and not to the ESP, which does not appear to share the same set of flags. https://github.com/systemd/systemd/issues/920
2015-08-08libsystemd-network: fix memory leakreverendhomer
2015-08-08Merge pull request #914 from reverendhomer/patch-2Daniel Mack
Coverity #1299013
2015-08-08Coverity #1299013reverendhomer
event cannot be NULL due to assert
2015-08-08Coverity #1299015reverendhomer
bus can never be NULL due to assert
2015-08-07nspawn: Allow : characters in overlay pathsRichard Maw
: characters can be entered with the \: escape sequence.
2015-08-07nspawn: escape paths in overlay mount optionsRichard Maw
Overlayfs uses , as an option separator and : as a list separator. These characters are both valid in file paths, so overlayfs allows file paths which contain these characters to backslash escape these values.
2015-08-07strv: Add strv_shell_escapeRichard Maw
This modifies the strv in-place, replacing strings with their escaped version. It's mostly just a convenience function for when you need to join a strv together because it's passed as a string to something, and the separator needs escaping.
2015-08-07util: Add shell_escapeRichard Maw
This is for shell-style \ escaping rather than quoting, which while it has the same effect in produced shell commands, is not exclusively useful for shell commands. shell_escape would be useful for producing sed commands, as you would be able to \ escape the normal special characters, plus whichever argument separator was chosen; or it could be used to escape arguments passed to the overlayfs mount command.
2015-08-07nspawn: Allow : characters in nspawn --bind pathsRichard Maw
: characters in bind paths can be entered as the \: escape sequence.
2015-08-07strv: convert strv_split_quotes into a generic strv_split_extractRichard Maw
strv_split_extract is to strv_split_quotes as extract_first_word was to unquote_first_word. Now there's extract_first_word for extracting a single argument, extract_many_words for extracting a bounded number of arguments, and strv_split_extract for extracting an arbitrary number of arguments.
2015-08-07nspawn: Allow : characters in --tmpfs pathRichard Maw
This now accepts : characters with the \: escape sequence. Other escape sequences are also interpreted, but having a \ in your file path is less likely than :, so this shouldn't break anyone's existing tools.
2015-08-07util: Allow non-separator coalescing parsing in extract_first_wordRichard Maw
If EXTRACT_DONT_COALESCE_SEPARATORS is passed, then leading separators, trailing separators and spans of multiple separators aren't skipped, and empty arguments from before, after or between separators may be extracted.
2015-08-07util: Don't interpret quotes by default in extract_first_wordRichard Maw
This adds an EXTRACT_QUOTES option to allow the previous behaviour, of not interpreting any character inside ' or " quotes as separators.
2015-08-07util: change unquote_*_word to extract_*_wordRichard Maw
It now takes a separators argument, which defaults to WHITESPACE if NULL is passed.
2015-08-07unquote_first_word: set *p=NULL on terminationRichard Maw
To add a flag to allow an empty string to be parsed as an argument, we need to be able to distinguish between the end of the string, and after the end of the string, so when we *do* reach the end, let's set *p to this state.
2015-08-07Convert unquote_*_word users to expect isempty(p) after the last entryRichard Maw
This is so that, when called in a loop, unquote_first_word can distinguish between reaching the end of a string because it has consumed all the input before the end, and consuming all the input. This is important because we later add a flag that allows char *in = ""; char *out; unquote_first_word(&in, &out, flags); To put "" in out, and set in = NULL, so the trailing empty string of the input can be consumed, and mark that the input has been consumed.
2015-08-07sd-daemon: return EBADF for invalid fd numbersZbigniew Jędrzejewski-Szmek
This matches what open(2) and other system functions do.
2015-08-07core: s/reexection/reexecution/ typo fixJan Pokorný
Signed-off-by: Jan Pokorný <jpokorny@redhat.com>
2015-08-07load-fragment: fix segv on parse errorZbigniew Jędrzejewski-Szmek
https://bugzilla.redhat.com/show_bug.cgi?id=1251334
2015-08-06Merge pull request #898 from poettering/machined-fix-reloadDaniel Mack
Fix for #376, plus some other fixes
2015-08-06machined: rework state tracking logic for machinesLennart Poettering
This splits up the stopping logic for machines into two steps: first on machine_stop() we begin with the shutdown of a machine by queuing the stop method call for it. Then, in machine_finalize() we actually remove the rest of its runtime context. This mimics closely how sessions are handled in logind. This also reworks the GC logic to strictly check the current state of the machine unit, rather than shortcutting a few cases, like for example assuming that UnitRemoved really means a machine is gone (which it isn't since Reloading might trigger it, see #376). Fixes #376.
2015-08-06machine: drop state variable from Machine objectLennart Poettering
We never made use of it, let's get rid of it.
2015-08-06logind,machined: various smaller cleanupsLennart Poettering
Use mfree() where we can. Drop unnecessary {}. Drop unnecessary variable declarations. Cast syscall invocations where explicitly don't care for the return value to (void). Reword a comment.
2015-08-06machined,logind: don't generate errors on signal match functionsLennart Poettering
If we get a weird signal, then we should log about it, but not return an error, since sd-bus will not call us again then anymore, but for these signals we match here we actually do want to be called on the next invocation.
2015-08-06sd-bus: ignore BLOOM_FILTER kdbus itemsLennart Poettering
The kernel nowadays sends these along, and that's OK, hence don't even debug log about it, but completely ignore it.
2015-08-06Merge pull request #896 from poettering/runtimedir-execTom Gundersen
execute: don't fail if we create the runtime directory from two proce…
2015-08-06bootctl: fix path in log messageKarel Zak
2015-08-06execute: don't fail if we create the runtime directory from two processes ↵Lennart Poettering
simultaneously If a service has both ExecStart= and ExecStartPost= set with Type=simple, then it might happen that we have two children create the runtime directory of a service (as configured with RuntimeDirectory=) at the same time. Previously we did this with mkdir_safe() which will create the dir only if it is missing, but if it already exists will at least verify the access mode and ownership to match the right values. This is problematic in this case, since it creates and then adjusts the settings, thus it might happen that one child creates the directory with root owner, another one then verifies it, and only afterwards the directory ownership is fixed by the original child, while the second child already failed. With this change we'll now always adjust the access mode, so that we know that it is right. In the worst case this means we adjust the mode/ownership even though its unnecessary, but this should have no negative effect. https://bugzilla.redhat.com/show_bug.cgi?id=1226509
2015-08-06udev: variable dereferenced before check 'cmsg'Karel Zak
2015-08-06bus-proxy: add comment to remove -EPERM checkKay Sievers
2015-08-06libsystemd: fix RTNL_CONTAINER_DEPTH assertKarel Zak
The m->n_containers is index and has to be smaller than the array size.
2015-08-06core: unit: remove bus slot after calling unit_done()Daniel Mack
The ->done callback in the unit's vtable might call into unit_unwatch_bus_name() and corrupt memory by that. Move the call down, and clean up the bus slot in case it hasn't been done yet.
2015-08-06bus-proxy: fix error code for invalid reply-slotsDavid Herrmann
The kernel may return EBADSLT if a reply slot cannot be found. Make sure to ignore it just like we ignore EPERM (the comment still applies).