<![CDATA[Bug in logwatch subagent]]>

<![CDATA[Bug in logwatch subagent]]>If there's a eventlog entry bigger than the default BUFFER_SIZE then the logwatch enters an endless loop.
If the buffer isn't big enough then there's a realloc with size "bytesNeeded" but the buffersize stays "BUFFER_SIZE" in stead of bufferSize.

diff:
--- C:/Temp/eventlog.cpp-revBASE.svn001.tmp.cpp Tue Aug 14 19:44:38 2012
+++ D:/Source/netxms/trunk/src/agent/subagents/logwatch/eventlog.cpp Wed Feb 27 19:56:50 2013
@@ -341,7 +341,7 @@
{
retry_read:
success = ReadEventLog(hLog, EVENTLOG_SEQUENTIAL_READ | EVENTLOG_FORWARDS_READ, 0,
- buffer, BUFFER_SIZE, &bytes, &bytesNeeded);
+ buffer, bufferSize, &bytes, &bytesNeeded);
                     error = GetLastError();
                     if (!success && (error == ERROR_INSUFFICIENT_BUFFER))
                     {

]]>https://forum.netxms.org/topic/2218/bug-in-logwatch-subagentRSS for NodeFri, 13 Nov 2026 06:23:03 GMTWed, 27 Feb 2013 19:04:00 GMT60<![CDATA[Reply to Bug in logwatch subagent on Thu, 28 Feb 2013 07:15:44 GMT]]>Hi!

Thanks for the fix, just committed it into trunk.

Best regards,
Victor

]]>https://forum.netxms.org/post/9791https://forum.netxms.org/post/9791Thu, 28 Feb 2013 07:15:44 GMT