[Linux-cluster] How does log option work in RHEL5 cluster

[Lon Hohberger]

On Wed, Jul 25, 2007 at 12:59:24PM -0400, siman hew wrote:
> It "works", because you did not change the default log level.
> How about change the log level, like me set it to 3, then run the exactly
> same commands, see what it will happen.

Yup, you're right.  This fixes it.

-- 
Lon Hohberger - Software Engineer - Red Hat, Inc.
-------------- next part --------------
? clubufflush
? clufindhostname
? clulog
? clunfslock
? clustat
? clusvcadm
? foo.out
? foo.out.diff
? foo.out2
? foo.out3
? foo.out4
Index: clulog.c
===================================================================
RCS file: /cvs/cluster/cluster/rgmanager/src/utils/clulog.c,v
retrieving revision 1.1.2.3
diff -u -r1.1.2.3 clulog.c
--- clulog.c	3 May 2007 15:02:47 -0000	1.1.2.3
+++ clulog.c	25 Jul 2007 17:58:05 -0000
@@ -93,9 +93,6 @@
     if (argc < 4)
 	usage(argv[0]);
 
-    /* ../daemons/log.c */
-    configure_logging(-1);
-
     while ((opt = getopt(argc, argv, "f:l:s:hp:n:")) != -1) {
 	switch (opt) {
 	case 'l':
@@ -134,9 +131,10 @@
     if (!cmdline_loglevel) {
 	/*
 	 * Let's see what loglevel the SM is running at.
-	 * TODO Get rgmgr log level
+	 * If ccsd's not available, use default.
 	 */
-	clu_set_loglevel(LOGLEVEL_DFLT);
+    	if (configure_logging(-1) < 0)
+		clu_set_loglevel(LOGLEVEL_DFLT);
     }
     result = clulog_pid(severity, pid, progname, logmsg);
     free(progname);