| Commit message (Collapse) | Author | Age | Files | Lines |
|\
| |
| |
| |
| |
| |
| |
| | |
Conflicts:
Makefile.am
plugins/imfile/imfile.c
runtime/stream.c
tests/tcpflood.c
|
| | |
|
| |\
| | |
| | |
| | |
| | |
| | | |
Conflicts:
Makefile.am
tests/tcpflood.c
|
| | |\
| | | |
| | | |
| | | |
| | | |
| | | | |
Conflicts:
ChangeLog
runtime/stream.c
|
| | | | |
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
- bugfix: a couple of problems that imfile had on some platforms, namely
Ubuntu (not their fault, but occured there)
- bugfix: imfile utilizes 32 bit to track offset. Most importantly,
this problem can not experienced on Fedora 64 bit OS (which has
64 bit long's!)
|
| | | |
| | | |
| | | |
| | | | |
Signed-off-by: Rainer Gerhards <rgerhards@adiscon.com>
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
either one or two bugs fixed ;)
Definitely a problem where no state file is written when working with
relative pathes. Also, some problems with offsets should be fixed for
very large files. However, I could not yet experimentally show the issue
so it probably needs more verification.
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
by doing math a little bit more optimal in hash table code. Also reduced
memory requirement for imuxsock hash tables (expected number of connections
was set too high -- table can be extended dynamically).
|
| | | |
| | | |
| | | |
| | | |
| | | | |
When the rsyslog service is enabled via
"systemctl enable rsyslog.service" also enable the socket unit.
|
| | | | |
|
| | | | |
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Note: we do rate-limiting by the pid and we obtain the pid via SCM_CREDENTIALS
socket options. So this patch effectively disables the (new) ratelimiting
capability. In a later patch, I'll see that I can provide a global ratelimiting
capability, which could always be used (an alternative may be using the tag, will
check this out as well).
|
| | | | |
|
|\ \ \ \ |
|
| | | | | |
|
| | | | | |
|
| | | | | |
|
| | | | | |
|
| | | | | |
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
finally this looks almost production ready for files where no
directory path needs to be created
|
| | | | |
| | | | |
| | | | |
| | | | | |
Note:compiles, but not yet tested
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
... that I could only see on my execution environment (I right
now have two envs, a local one where I can compile, but not run
and a remote one where I can do both, but this is a bit less
convenient to use).
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
now things look much better, also done some prep in order to
support a file cache (we need this for multiple selectors writing
to the same file).
|
| | | | |
| | | | |
| | | | |
| | | | | |
but still pretty ugly. Any java folks out there to help clean it up?
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
very crude implementation, but probably good enough to gather some
early performance data and experience with the module. No real
error handling done, if something breaks, the whole thing will
be blown up ;)
|
| | | | | |
|
| | | | | |
|
| | | | |
| | | | |
| | | | |
| | | | | |
I left a testing setting in, resulting in /dev/log never being deleted...
|
| | | | | |
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
This adds a systemd socket and service unit file to the default install
if systemd is found or explicitly enabled in ./configure.
Patch is against current git v5-devel.
|
|/ / / /
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
related bug tracker: http://bugzilla.adiscon.com/show_bug.cgi?id=200
Note that this may not be the ultimate fix. I could not definitely
identify the root cause of the problem and so I did some changes
that I *think* (but could not verify) that resolved the issue.
|
| | | |
| | | |
| | | |
| | | | |
also improved test tool
|
| | | | |
|
| | | | |
|
| | | | |
|
| | | |
| | | |
| | | |
| | | |
| | | | |
rate limiting now applies only to messages with a given severity or above. By
default, emergency messages are NOT rate-limited.
|
| | | | |
|
| | | |
| | | |
| | | |
| | | | |
found at http://www.cl.cam.ac.uk/~cwc22/hashtable/
|
| | | |
| | | |
| | | |
| | | | |
works, but at a global level, need to go down to pid or cgroup
|
| | | | |
|
| | | |
| | | |
| | | |
| | | |
| | | | |
still not great, but far better readable (and extendable) than what we had
before.
|
| | | | |
|
| | | | |
|
| | | | |
|
| | | | |
|
| | | | |
|
| | | | |
|
| | | | |
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
... well, actually this is a first real implementation of this subsystem.
I have added a counter registry, a way to access the countres (as readable
string) and a way to define and maintem them. Also, module impstats has
been updated to utilize the new system. Finally, I added some counters. I
hope that this sets the baseline for useful future enhancements.
|