summaryrefslogtreecommitdiff
path: root/src/journal
AgeCommit message (Collapse)Author
2013-06-10journal: letting (interleaved) seqnums goZbigniew Jędrzejewski-Szmek
In the following scenario: server creates system.journal server creates user-1000.journal both journals share the same seqnum_id. Then server writes to user-1000.journal first, and server writes to system.journal a bit later, and everything is fine. The server then terminates (crash, reboot, rsyslog testing, whatever), and user-1000.journal has entries which end with a lower seqnum than system.journal. Now server is restarted server opens user-1000.journal and writes entries to it... BAM! duplicate seqnums for the same seqnum_id. Now, we usually don't see that happen, because system.journal is closed last, and opened first. Since usually at least one message is written during boot and lands in the system.journal, the seqnum is initialized from it, and is set to a number higher than than anything found in user journals. Nevertheless, if system.journal is corrupted and is rotated, it can happen that an entry is written to the user journal with a seqnum that is a duplicate with an entry found in the corrupted system.journal~. When browsing the journal, journalctl can fall into a loop where it tries to follow the seqnums, and tries to go the next location by seqnum, and is transported back in time to to the older duplicate seqnum. There is not way to find out the maximum seqnum used in a multiple files, without actually looking at all of them. But we don't want to do that because it would be slow, and actually it isn't really possible, because a file might e.g. be temporarily unaccessible. Fix the problem by using different seqnum series for user journals. Using the same seqnum series for rotated journals is still fine, because we know that nothing will write to the rotated journal anymore. Likely related: https://bugs.freedesktop.org/show_bug.cgi?id=64566 https://bugs.freedesktop.org/show_bug.cgi?id=59856 https://bugs.freedesktop.org/show_bug.cgi?id=64296 https://bugs.archlinux.org/task/35581 https://bugzilla.novell.com/show_bug.cgi?id=817778 Possibly related: https://bugs.freedesktop.org/show_bug.cgi?id=64293
2013-06-10tests: add testcase for duplicate seqnumsZbigniew Jędrzejewski-Szmek
2013-06-10tests: add testcase for skipping-entries-on-direction-change-bugMarius Vollmer
This test case failed until a3e6f050de8. Taken from https://bugs.freedesktop.org/show_bug.cgi?id=65255.
2013-06-10journal: change direction tests to use the same convention (cp </> np)Zbigniew Jędrzejewski-Szmek
The order was different in various places, which makes it harder to read to code. Also consistently use ternany for all direction checks. Remove one free(NULL).
2013-06-10journal: remember last direction of search and keep offset cacheZbigniew Jędrzejewski-Szmek
The fields in JournalFile are moved around to avoid wasting 7 bytes because of alignment.
2013-06-10journalctl: allow the user to specify the file(s) to useZbigniew Jędrzejewski-Szmek
This is useful for debugging and feels pretty natural. For example answering the question "is this big .journal file worth keeping?" is made easier.
2013-06-10journal: add sd_journal_open_filesZbigniew Jędrzejewski-Szmek
This allows the caller to explicitly specify which journal files should be opened. The same functionality could be achieved before by creating a directory and playing around with symlinks. It is useful to debug stuff and explore the journal, and has been requested before. Waiting is supported, the journal will notice modifications on the files supplied when opening the journal, but will not add any new files.
2013-06-10tests: add test for empty journal filesZbigniew Jędrzejewski-Szmek
The headers are currently not printed properly: some "(null)"s appear.
2013-06-10journalctl: print monotonic timestamp in --headerZbigniew Jędrzejewski-Szmek
2013-06-10journalctl: print proper IDs with --headerZbigniew Jędrzejewski-Szmek
The same buffer was used for two different IDs, messing up the output.
2013-06-10Use stdint.h macros instead of casts to print uint64_t valuesZbigniew Jędrzejewski-Szmek
Casts are visually heavy, and can obscure unwanted truncations.
2013-06-10journalctl: no color for --reboot-- when not on ttyZbigniew Jędrzejewski-Szmek
2013-06-10journal: loop less in MATCH_AND_TERM conditionalsZbigniew Jędrzejewski-Szmek
AND term usually don't have many subterms (4 seems to be the maximum sensible number, e.g. _BOOT_ID && _SYSTEMD_UNIT && _PID && MESSAGE_ID). Nevertheless, the cost of checking each subterm can be relatively high, especially when the nested terms are compound, and it makes sense to minimize the number of checks. Instead of looping to the end and then again over the whole list once again after at least one term changed the offset, start the loop at the term which caused the change. This way ½ terms in the AND match are not checked unnecessarily again.
2013-06-10journalctl: add --system/--user flagsZbigniew Jędrzejewski-Szmek
--user basically gives messages from your own systemd --user services. --system basically gives messages from PID 1, kernel, and --system services. Those two options are not exahustive, because a priviledged user might be able to see messages from other users, and they will not be shown with either or both of those flags.
2013-06-10journal: add ability to filter by current userZbigniew Jędrzejewski-Szmek
This is the just the library part. SD_JOURNAL_CURRENT_USER flags is added to sd_j_open(), to open files from current user. SD_JOURNAL_SYSTEM_ONLY is renamed to SD_JOURNAL_SYSTEM, and changed to mean to (also) open system files. This way various flags can be combined, which gives them nicer semantics, especially if other ones are added later. Backwards compatibility is kept, because SD_JOURNAL_SYSTEM_ONLY is equivalent to SD_JOURNAL_SYSTEM if used alone, and before there we no other flags.
2013-06-10journal: simplify match_free_if_emptyZbigniew Jędrzejewski-Szmek
2013-06-09journalctl: fix verbose output when no logs are foundZbigniew Jędrzejewski-Szmek
$ journalctl -o verbose _EXE=/quiet/binary -f -- Logs begin at Sun 2013-03-17 17:28:22 EDT. -- Failed to get realtime timestamp: Cannot assign requested address JOURNAL_FOREACH_DATA_RETVAL is added, which allows the caller to get the return value from sd_journal_enumerate_data. I think we might want to expose this macro like SD_JOURNAL_FOREACH_DATA, but for now it is in journal-internal.h. There's a change in behaviour for output_*, not only in output_verbose, that errors in sd_j_enumerate_data are not silently ignored anymore. https://bugs.freedesktop.org/show_bug.cgi?id=56459
2013-06-04systemctl: limit logs in status to current bootZbigniew Jędrzejewski-Szmek
Also reworded a few debug messages for brevity, and added a log statement which prints out the filter at debug level: Journal filter: (((UNIT=sys-module-configfs.device AND _PID=1) OR (COREDUMP_UNIT=sys-module-configfs.device AND MESSAGE_ID=fc2e22bc6ee647b6b90729ab34a250b1) OR _SYSTEMD_UNIT=sys-module-configfs.device) AND _BOOT_ID=4e3c518ab0474c12ac8de7896fe6b154)
2013-06-01journal: take KeepFree into account when reporting maximum sizeDaniel Albers
When reporting the maximum journal size add a hint if it's limited by KeepFree.
2013-05-30journald: DO recalculate the ACL mask, but only if it doesn't existJan Alexander Steffens (heftig)
Since 11ec7ce, journald isn't setting the ACLs properly anymore if the files had no ACLs to begin with: acl_set_fd fails with EINVAL. An ACL with ACL_USER or ACL_GROUP entries but no ACL_MASK entry is invalid, so make sure a mask exists before trying to set the ACL.
2013-05-21man: Document missing optionsJan Janssen
2013-05-15journal: correctly convert usec_t to timespec.Michał Bartoszkiewicz
Use timespec_store instead of (incorrectly) doing it inline.
2013-05-14journalctl: add -k/--dmesgZbigniew Jędrzejewski-Szmek
2013-05-14sd-journal: check if the pointers passed are the sameThomas Hindoe Paaboel Andersen
2013-05-13journald-stream: typo in error message.Auke Kok
2013-05-08systemd-python: add __version__ stringsZbigniew Jędrzejewski-Szmek
2013-05-08Rearrange a few fields to reduce holesZbigniew Jędrzejewski-Szmek
2013-05-07hostname: only suppress setting of pretty hostname if it is non-equal to the ↵Lennart Poettering
static hostname and if the static hostname is set, too https://bugzilla.redhat.com/show_bug.cgi?id=957814
2013-05-07journald: don't recalculate the ACL maskLennart Poettering
Otherwise we might end up with executable files if some default ACL is set for the journal directory.
2013-05-07journald: be more careful when we try to flush the runtime journal to disk ↵systemd/v203Lennart Poettering
and the disk is close to being full Bump the minimal size of the journal so that we can be sure creating the journal file will always succeed. Previously the minimum size was smaller than a empty jounral file...
2013-05-03journal: as a µ-opt, use sizeof instead of strlenZbigniew Jędrzejewski-Szmek
2013-05-03journal: Set the default keep free value to 15% (up from 5%)Colin Guthrie
As some SSDs are still seeing performance degredation when reaching 85% usage the default value of 5% seems a little low. Set this to 15% by default.
2013-05-02Add __attribute__((const, pure, format)) in various placesZbigniew Jędrzejewski-Szmek
I'm assuming that it's fine if a _const_ or _pure_ function calls assert. It is assumed that the assert won't trigger, and even if it does, it can only trigger on the first call with a given set of parameters, and we don't care if the compiler moves the order of calls.
2013-04-25coredump: use realloc() loop instead of malloc(768M)Colin Walters
I typically run VMs with 1024MiB allocated; systemd is unable to write coredumps in this scenario at all because the default kernel configuration will only overcommit 50% of available RAM. Avoid this failure by using a realloc() loop. See: http://lists.freedesktop.org/archives/systemd-devel/2013-April/010709.html
2013-04-25Add some extra __attribute__ ((format)) sCristian Rodríguez
2013-04-25conf-parser: restrict .include usageLennart Poettering
Disallow recursive .include, and make it unavailable in anything but unit files.
2013-04-25journal: remove build warning when SELinux is disabledGreg Kroah-Hartman
A small patch to remove a build warnining when SELinux is disabled.
2013-04-24Add set_consume which always takes ownershipZbigniew Jędrzejewski-Szmek
Freeing in error path is the common pattern with set_put().
2013-04-22cgroup: make sure all our cgroup objects have a suffix and are properly escapedLennart Poettering
Session objects will now get the .session suffix, user objects the .user suffix, nspawn containers the .nspawn suffix. This also changes the user cgroups to be named after the numeric UID rather than the username, since this allows us the parse these paths standalone without requiring access to the cgroup file system. This also changes the mapping of instanced units to cgroups. Instead of mapping foo@bar.service to the cgroup path /user/foo@.service/bar we will now map it to /user/foo@.service/foo@bar.service, in order to ensure that all our objects are properly suffixed in the tree.
2013-04-19Reintroduce f_type comparison macroHarald Hoyer
This reverts commit 4826f0b7b5c0aefa08b8cc7ef64d69027f84da2c. Because statfs.t_type can be int on some architecures, we have to cast the const magic to the type, otherwise the compiler warns about signed/unsigned comparison, because the magic can be 32 bit unsigned. statfs(2) man page is also wrong on some systems, because f_type is not __SWORD_TYPE on some architecures. The following program: int main(int argc, char**argv) { struct statfs s; statfs(argv[1], &s); printf("sizeof(f_type) = %d\n", sizeof(s.f_type)); printf("sizeof(__SWORD_TYPE) = %d\n", sizeof(__SWORD_TYPE)); printf("sizeof(long) = %d\n", sizeof(long)); printf("sizeof(int) = %d\n", sizeof(int)); if (sizeof(s.f_type) == sizeof(int)) { printf("f_type = 0x%x\n", s.f_type); } else { printf("f_type = 0x%lx\n", s.f_type); } return 0; } executed on s390x gives for a btrfs: sizeof(f_type) = 4 sizeof(__SWORD_TYPE) = 8 sizeof(long) = 8 sizeof(int) = 4 f_type = 0x9123683e
2013-04-18journal: when iterating through a file we might lose messages when changing ↵Marius Vollmer
direction. https://bugs.freedesktop.org/show_bug.cgi?id=63672
2013-04-18Revert f_type fixupsHarald Hoyer
This reverts commit a858b64dddf79177e12ed30f5e8c47a1471c8bfe. This reverts commit aea275c43194b6ac519ef907b62c5c995050fde0. This reverts commit fc6e6d245ee3989c222a2a8cc82a33475f9922f3. This reverts commit c4073a27c555aeceac87a3b02a83141cde641a1e. This reverts commit cddf148028f525be8176e7f1cbbf4f862bd287f6. This reverts commit 8c68a70170b31f93c287f29fd06c6c17edaf19ad. The constants are now casted to __SWORD_TYPE, which should resolve the compiler warnings about signed vs unsigned. After talking to Kay, we concluded: This should be fixed in the kernel, not worked around in userspace tools. Architectures cannot use int and expect magic constants lager than INT_MAX to work correctly. The kernel header needs to be fixed. Even coreutils cannot handle it: #define RAMFS_MAGIC 0x858458f6 # stat -f -c%t / ffffffff858458f6 #define BTRFS_SUPER_MAGIC 0x9123683E # stat -f -c%t /mnt ffffffff9123683e Although I found the perfect working macro to fix the thing :) __extension__ ({ \ bool _ret = false; \ switch(f) { case c: _ret=true; }; \ ( _ret ); \ })
2013-04-18move _cleanup_ attribute in front of the typeHarald Hoyer
http://lists.freedesktop.org/archives/systemd-devel/2013-April/010510.html
2013-04-18rename CMP_F_TYPE to F_TYPE_CMPHarald Hoyer
2013-04-18Add ugly CMP_F_TYPE() macroHarald Hoyer
On some architectures (like s390x) the kernel has the type int for f_type, but long in userspace. Assigning the 32 bit magic constants from linux/magic.h to the 31 bit signed f_type in the kernel, causes f_type to be negative for some constants. glibc extends the int to long for those architecures in 64 bit mode, so the negative int becomes a negative long, which cannot be simply compared to the original magic constant, because the compiler would automatically cast the constant to long. To workaround this issue, we also compare to the (int)MAGIC value in a macro. Of course, we could do #ifdef with the architecure, but it has to be maintained, and the magic constants are 32 bit anyway. Someday, when the int is unsigned or long for all architectures, we can remove this macro again. Until then, keep it as simple as it can be.
2013-04-18build-sys: add sd_journal_add_conjunction() to symbol versioning fileLennart Poettering
2013-04-17fixup for cddf148028f52Harald Hoyer
Instead of making a type up, just use __SWORD_TYPE, after reading statfs(2). Too bad, this does not fix s390x because __SWORD_TYPE is (long int) and the kernel uses (int) to fill in the field!!!!!!
2013-04-17fixup 8c68a7017 and cast to (unsigned long)Harald Hoyer
2013-04-17fixed statfs.f_type signed vs unsigned comparisonsHarald Hoyer
statfs.f_type is signed but the filesystem magics are unsigned. Casting the magics to signed will not make the signed. Problem seen on big-endian 64bit s390x with __fsword_t 8 bytes. Casting statfs.f_type to unsigned on the other hand will get us what we need. https://bugzilla.redhat.com/show_bug.cgi?id=953217
2013-04-17journalctl: specify "--unit=" and "--user-unit" multiple timesHarald Hoyer
Previously only one "--unit=" or "--user-unit" could be specified. With this patch, journalcrtl can show multiple units. $ journalctl -u systemd-udevd.service -u sshd.service -u crond.service -b -- Logs begin at Sa 2013-03-23 11:08:45 CET, end at Fr 2013-04-12 09:10:22 CEST. -- Apr 12 08:41:37 lenovo systemd[1]: Started udev Kernel Device Manager. Apr 12 08:41:37 lenovo systemd[1]: Stopped udev Kernel Device Manager. Apr 12 08:41:38 lenovo systemd[1]: Started udev Kernel Device Manager. Apr 12 08:41:38 lenovo crond[291]: (CRON) INFO (Syslog will be used instead of sendmail.) Apr 12 08:41:38 lenovo crond[291]: (CRON) INFO (running with inotify support) Apr 12 08:41:39 lenovo systemd[1]: Starting OpenSSH server daemon... Apr 12 08:41:39 lenovo systemd[1]: Started OpenSSH server daemon. Apr 12 08:41:39 lenovo sshd[355]: Server listening on 0.0.0.0 port 22. Apr 12 08:41:39 lenovo sshd[355]: Server listening on :: port 22. Apr 12 08:41:39 lenovo mtp-probe[373]: checking bus 1, device 8: "/sys/devices/pci0000:00/0000:00:1a.0/usb1/1-1/1-1.5/1-1.5.6/1-1.5.6.2/1-1.5.6.2.1"
generated by cgit v1.2.3-54-g00ecf (git 2.45.2) at 2024-12-23 00:28:34 +0000