diff options
Diffstat (limited to 'ChangeLog')
-rw-r--r-- | ChangeLog | 128 |
1 files changed, 128 insertions, 0 deletions
@@ -1,4 +1,132 @@ --------------------------------------------------------------------------- +Version 4.5.4 [v4-beta] (rgerhards), 2009-09-29 +- bugfix: potential segfault in stream writer on destruction + Most severely affected omfile. The problem was that some buffers were + freed before the asynchronous writer thread was shut down. So the + writer thread accessed invalid data, which may even already be + overwritten. Symptoms (with omfile) were segfaults, grabled data + and files with random names placed around the file system (most + prominently into the root directory). Special thanks to Aaron for + helping to track this down. +- bugfix: potential race in object loader (obj.c) during use/release + of object interface +- bugfixes: potential problems in out file zip writer. Problems could + lead to abort and/or memory leak. The module is now hardened in a very + conservative way, which is sub-optimal from a performance point of view. + This should be improved if it has proven reliable in practice. +--------------------------------------------------------------------------- +Version 4.5.3 [v4-beta] (rgerhards), 2009-09-17 +- bugfix: repeated messages were incorrectly processed + this could lead to loss of the repeated message content. As a side- + effect, it could probably also be possible that some segfault occurs + (quite unlikely). The root cause was that some counters introduced + during the malloc optimizations were not properly duplicated in + MsgDup(). Note that repeated message processing is not enabled + by default. +- bugfix: message sanitation had some issues: + - control character DEL was not properly escaped + - NUL and LF characters were not properly stripped if no control + character replacement was to be done + - NUL characters in the message body were silently dropped (this was + a regeression introduced by some of the recent optimizations) +- bugfix: strings improperly reused, resulting in some message properties + be populated with strings from previous messages. This was caused by + an improper predicate check. [backported from v5] +- fixed some minor portability issues +- bugfix: reverse lookup reduction logic in imudp do DNS queries too often + [imported from 4.4.2] +--------------------------------------------------------------------------- +Version 4.5.2 [v4-beta] (rgerhards), 2009-08-21 +- legacy syslog parser changed so that it now accepts date stamps in + wrong case. Some devices seem to create them and I do not see any harm + in supporting that. +- added $InputTCPMaxListeners directive - permits to specify how many + TCP servers shall be possible (default is 20). +- bugfix: memory leak with some input modules. Those inputs that + use parseAndSubmitMsg() leak two small memory blocks with every message. + Typically, those process only relatively few messages, so the issue + does most probably not have any effect in practice. +- bugfix: if tcp listen port could not be created, no error message was + emitted +- bugfix: potential segfault in output file writer (omfile) + In async write mode, we use modular arithmetic to index the output + buffer array. However, the counter variables accidently were signed, + thus resulting in negative indizes after integer overflow. That in turn + could lead to segfaults, but was depending on the memory layout of + the instance in question (which in turn depended on a number of + variables, like compile settings but also configuration). The counters + are now unsigned (as they always should have been) and so the dangling + mis-indexing does no longer happen. This bug potentially affected all + installations, even if only some may actually have seen a segfault. +- bugfix: hostnames with dashes in them were incorrectly treated as + malformed, thus causing them to be treated as TAG (this was a regression + introduced from the "rfc3164 strict" change in 4.5.0). +--------------------------------------------------------------------------- +Version 4.5.1 [DEVEL] (rgerhards), 2009-07-15 +- CONFIG CHANGE: $HUPisRestart default is now "off". We are doing this + to support removal of restart-type HUP in v5. +- bugfix: fromhost-ip was sometimes truncated +- bugfix: potential segfault when zip-compressed syslog records were + received (double free) +- bugfix: properties inputname, fromhost, fromhost-ip, msg were lost when + working with disk queues +- performance enhancement: much faster, up to twice as fast (depending + on configuration) +- bugfix: abort condition when RecvFrom was not set and message reduction + was on. Happend e.g. with imuxsock. +- added $klogConsoleLogLevel directive which permits to set a new + console log level while rsyslog is active +- bugfix: message could be truncated after TAG, often when forwarding + This was a result of an internal processing error if maximum field + sizes had been specified in the property replacer. +- added ability for the TCP output action to "rebind" its send socket after + sending n messages (actually, it re-opens the connection, the name is + used because this is a concept very similiar to $ActionUDPRebindInterval). + New config directive $ActionSendTCPRebindInterval added for the purpose. + By default, rebinding is disabled. This is considered useful for load + balancers. +- testbench improvements +--------------------------------------------------------------------------- +Version 4.5.0 [DEVEL] (rgerhards), 2009-07-02 +- activation order of inputs changed, they are now activated only after + privileges are dropped. Thanks to Michael Terry for the patch. +- greatly improved performance +- greatly reduced memory requirements of msg object + to around half of the previous demand. This means that more messages can + be stored in core! Due to fewer cache misses, this also means some + performance improvement. +- improved config error messages: now contain a copy of the config line + that (most likely) caused the error +- reduced max value for $DynaFileCacheSize to 1,000 (the former maximum + of 10,000 really made no sense, even 1,000 is very high, but we like + to keep the user in control ;)). +- added capability to fsync() queue disk files for enhanced reliability + (also add's speed, because you do no longer need to run the whole file + system in sync mode) +- more strict parsing of the hostname in rfc3164 mode, hopefully + removes false positives (but may cause some trouble with hostname + parsing). For details, see this bug tracker: + http://bugzilla.adiscon.com/show_bug.cgi?id=126 +- added configuration commands (see doc for explanations) + * $OMFileZipLevel + * $OMFileIOBufferSize + * $OMFileFlushOnTXEnd + * $MainMsgQueueSyncQueueFiles + * $ActionQueueSyncQueueFiles +- done some memory accesses explicitely atomic +- bugfix: subtle (and usually irrelevant) issue in timout processing + timeout could be one second too early if nanoseconds wrapped +- set a more sensible timeout for shutdow, now 1.5 seconds to complete + processing (this also removes those cases where the shutdown message + was not written because the termination happened before it) +- internal bugfix: object pointer was only reset to NULL when an object + was actually destructed. This most likely had no effect to existing code, + but it may also have caused trouble in remote cases. Similarly, the fix + may also cause trouble... +- bugfix: missing initialization during timestamp creation + This could lead to timestamps written in the wrong format, but not to + an abort +--------------------------------------------------------------------------- Version 4.4.2 [v4-stable] (rgerhards), 2009-09-?? - bugfix: invalid handling of zero-sized messages, could lead to mis- addressing and potential memory corruption/segfault |