diff options
author | Rainer Gerhards <rgerhards@adiscon.com> | 2004-12-08 11:53:00 +0000 |
---|---|---|
committer | Rainer Gerhards <rgerhards@adiscon.com> | 2004-12-08 11:53:00 +0000 |
commit | fb7ed6d2ca2546b0a3ec851276c4521f4d26497b (patch) | |
tree | 2c6949f3178a3ecb87b3143ff1e844de3aaf3451 | |
parent | af41d91b65bccc5a210926b7f5fbd0f7f5349bd4 (diff) | |
download | rsyslog-fb7ed6d2ca2546b0a3ec851276c4521f4d26497b.tar.gz rsyslog-fb7ed6d2ca2546b0a3ec851276c4521f4d26497b.tar.xz rsyslog-fb7ed6d2ca2546b0a3ec851276c4521f4d26497b.zip |
getting closer to initial (beta) release
-rw-r--r-- | BUGS | 31 | ||||
-rw-r--r-- | INSTALL | 1 | ||||
-rw-r--r-- | README.linux | 2 | ||||
-rw-r--r-- | syslogd.c | 20 |
4 files changed, 49 insertions, 5 deletions
@@ -5,3 +5,34 @@ REPEATED LOG LINES the duplicates are NOT suppressed. This is done by sysklogd in all cases. We plan to add this as an optional feature, but as of now every line is logged. + +ERRORS ON STARTUP + Error reporting on startup is not particular good. We handle all + errors correctly, but sometimes a message is only reported if + running in debug mode. If you experience some strange behaviour + I suggest to run the syslogd interactively with debugging enabled. + +VERY LARGE LOG FILES + We have done nothing to support log files larger than 2GB. As such, it + behaves just like sysklogd in this regard. And, as far as I recall, this + means that - at least with certain glibc versions - no more than 2GB + will be written. This will be checked and supported at a later time, for + now just be careful ;). + +EQUALLY-NAMED TEMPLATES + If multiple templates with the SAME name are created, all but the + first definition is IGNORED. So you can NOT (yet) replace a + template definition. + +MEMORY LEAK ON HUPING + Currently, the template memory is NOT freed when SIGHUP is received. + This causes a small memory leak on each HUP. Will be fixed soon and + should not cause real trouble unless you HUP syslogd quite frequently. + +Updates to the bug lists (and potential fixes...) can be found at the +official home page at + +http://www.monitorware.com/rsyslog + +List as of 2004-12-08 +Rainer Gerhards @@ -26,7 +26,6 @@ may be security concerns with running it as root. Please repeat step 1 if you are unsure of why this may be the case. -5.) The make file currently doe 5.) YOU NEED TO DECIDE IF YOU WOULD LIKE TO REPLACE THE STANDARD SYSLOGD OR NOT! The is important as it influences the names used. If you call make install diff --git a/README.linux b/README.linux index 1192f3f7..16718cd9 100644 --- a/README.linux +++ b/README.linux @@ -52,4 +52,4 @@ on sourceforge.net. Best regards, Rainer Gerhards Adiscon -2004-11-26 +2004-12-08 @@ -9,6 +9,7 @@ * is specified in an action line (hint: keep action type * to F_UNUSED unless a proper template could be found) * ONLY TO BE DONE FOR MySQL logging, rest is fixed rgerhards 2004-12-02 + * - implement the escape-cc property replacer option * * \brief This is what will become the rsyslogd daemon. * @@ -3825,10 +3826,12 @@ void die(sig) fprintlog(f, 0); } - Initialized = was_initialized; + Initialized = was_initialized; /* we restore this so that the logmsgInternal() + * below can work ... and keep in mind we need the + * filed structure still intact (initilzed) for the below! */ if (sig) { dprintf("syslogd: exiting on signal %d\n", sig); - (void) snprintf(buf, sizeof(buf), "rsyslogd: exiting on signal %d", sig); + (void) snprintf(buf, sizeof(buf), "syslogd: exiting on signal %d", sig); errno = 0; logmsgInternal(LOG_SYSLOG|LOG_INFO, buf, LocalHostName, ADDDATE); } @@ -3839,6 +3842,8 @@ void die(sig) if (f->f_type == F_MYSQL) closeMySQL(f); } + + /* now clean up the listener part */ /* Close the UNIX sockets. */ for (i = 0; i < nfunix; i++) @@ -4011,7 +4016,7 @@ void init() Files = NULL; } - /* TODO: we need to free the templates! */ + /* TODO: we need to free the templates! TODO-URGENT */ #ifdef SYSV lognum = 0; @@ -4981,6 +4986,15 @@ int checkDBErrorState(register struct filed *f) still valid, we need no reInit. Of course, if an unkown error appeared, we should reInit. */ + /* rgerhards 2004-12-08: I think it is pretty unlikely + * that we can re-use a connection after the error. So I guess + * the connection must be closed and re-opened in all cases + * (as it is done currently). When we come back to optimize + * this code, we should anyhow see if there are cases where + * we could keep it open. I just doubt this won't be the case. + * I added this comment (and did not remove Michaels) just so + * that we all know what we are looking for. + */ f->f_timeResumeOnError = 0; f->f_iLastDBErrNo = 0; reInitMySQL(f); |