NetBSD Problem Report #46564

From Wolfgang.Stukenbrock@nagler-company.com  Fri Jun  8 12:23:17 2012
Return-Path: <Wolfgang.Stukenbrock@nagler-company.com>
Received: from mail.netbsd.org (mail.netbsd.org [149.20.53.66])
	by www.NetBSD.org (Postfix) with ESMTP id A8FFF63B882
	for <gnats-bugs@gnats.NetBSD.org>; Fri,  8 Jun 2012 12:23:17 +0000 (UTC)
Message-Id: <20120608122253.EA61E97110@s011.nagler-company.com>
Date: Fri,  8 Jun 2012 14:22:53 +0200 (CEST)
From: Wolfgang.Stukenbrock@nagler-company.com
Reply-To: Wolfgang.Stukenbrock@nagler-company.com
To: gnats-bugs@gnats.NetBSD.org
Subject: yppush pushes map to himself triggering an error message each time
X-Send-Pr-Version: 3.95

>Number:         46564
>Category:       bin
>Synopsis:       yppush pushes map to himself triggering an error message each time
>Confidential:   no
>Severity:       serious
>Priority:       high
>Responsible:    bin-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Fri Jun 08 12:25:00 +0000 2012
>Last-Modified:  Fri Jun 08 13:35:01 +0000 2012
>Originator:     Dr. Wolfgang Stukenbrock
>Release:        NetBSD 5.1.2
>Organization:
Dr. Nagler & Company GmbH
>Environment:


System: NetBSD s011 5.1.2 NetBSD 5.1.2 (NSW-S011) #2: Tue Jun 5 15:02:21 CEST 2012 wgstuken@s011:/usr/src/sys/arch/amd64/compile/NSW-S011 amd64
Architecture: x86_64
Machine: amd64
>Description:
	In a setup with one master and multiple slave-YP-servers the ypservers map is used to find the slave servers
	during yppush operation to inform them about map updates.
	It is common to have the master yp-server hostname in this map too.
	The implementation of yppush in NetBSD accedently does not check for the own hostname as other implementations do.
	(E.g. in OpenSolaris a simple strcmp() with the name returned from sysctl() is done)
	When trying to send the just build new map to ourself, the system detects that the map that should be transfered
	is not newver than the current one and yppush prints an error message.
	This is ugly, because this message will be seem by the caller of the make that updates some maps.
	It is stupid too, to send the map from the master to the master again.
>How-To-Repeat:
	setup a host with ypserv, add some host to ypservers map, including this host
	and try to update a map. You will get an error message that the new map is
	not newer than the current one.
>Fix:
	The following patch to /usr/src/usr.sbin/ypserv/yppush/yppush.c will fix this issue.
	The check is integrated by calling gethostbyname() in order to get the full calified name and to allow alias
	name expansion - e.g. an alias name is specified in the ypservers map.
	The check is omitted, if the internal variables cannot hold the data or the hostname lookup fails in order
	to preserve previous behaviour as much as possible (on "strange" input data).

--- yppush.c	2012/06/08 11:23:50	1.1
+++ yppush.c	2012/06/08 12:05:54
@@ -56,6 +56,8 @@
 #include <rpcsvc/yp_prot.h>
 #include <rpcsvc/ypclnt.h>

+#include <netdb.h>
+
 #include "ypdb.h"
 #include "ypdef.h"
 #include "yplib_host.h"
@@ -89,6 +91,13 @@
 int     verbo = 0;		/* verbose */

 /*
+ * remark: do not use YPMAXPEER here to allow longer canonical host names
+ *         as the relative short YP-limit.
+ */
+#define MY_MAX_HOST 256
+static char my_hostname[MY_MAX_HOST];
+
+/*
  * prototypes
  */

@@ -118,6 +127,27 @@
 	struct timeval tv;
 	enum clnt_stat retval;
 	struct ypall_callback ypallcb;
+	struct hostent *he;
+
+	/*
+	 * setup own host name
+	 */
+
+	my_hostname[sizeof(my_hostname) - 1] = '\0';
+	if (gethostname(my_hostname, sizeof(my_hostname) - 1) != 0)
+	  {
+	    errx(1, "cannot determine own host name");
+	  }
+	if ((he = gethostbyname(my_hostname)) == NULL)
+	  {
+	    errx(1, "cannot determine canonical host name for this host");
+	  }
+	if (strlen(he->h_name) >= sizeof(my_hostname))
+	  {
+	    errx(1, "canonical host name to long (%lu - limit %lu)",
+	      (unsigned long)strlen(he->h_name), sizeof(my_hostname) - 1);
+	  }
+	strcpy(my_hostname, he->h_name); // save canonical name for later compare ...

 	/*
          * parse command line
@@ -298,6 +328,28 @@
        int invallen, char *indata)
 {
 	struct yppush_info *ypi = (struct yppush_info *) indata;
+	struct hostent *he;
+	char tmp_hname[MY_MAX_HOST];
+
+	/*
+	 * check if for own host name and do not push to ourself ...
+	 */
+	if (inkeylen < MY_MAX_HOST) {
+		/*
+		 * perform the check only if we are able to store it
+		 * preserv previous semantic for oversized host names ...
+		 */
+		memcpy(tmp_hname, inkey, inkeylen);
+		tmp_hname[inkeylen] = '\0';
+		if ((he = gethostbyname(tmp_hname)) != NULL) {
+			if (!strcmp(my_hostname, he->h_name)) {
+				/*
+				 * skip own host name ....
+				 */
+				return 0;
+			}
+		}
+	}

 	push(inkey, inkeylen, ypi);		/* do it! */
 	return (0);

>Audit-Trail:
From: Wolfgang Stukenbrock <wolfgang.stukenbrock@nagler-company.com>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: bin/46564: yppush pushes map to himself triggering an error message each time
Date: Fri, 08 Jun 2012 14:41:08 +0200

 Hi again,

 I've just detected that my patch will not compile on all architectures 
 due to a problem with printf() and sizeof() data types.
 Around line 148 the statement:
 errx(1, .... long (%lu - limit %lu)", (unsigned long)(strlen(...)), 
 sizeof(...) - 1);
 should be changed into
 errx(1, .... long (%lu - limit %lu)", (unsigned long)(strlen(...)), 
 (unsigned long)(sizeof(...) - 1));

 Sorry for this.

 W. Stukenbrock

 gnats-admin@NetBSD.org wrote:

 > Thank you very much for your problem report.
 > It has the internal identification `bin/46564'.
 > The individual assigned to look at your
 > report is: bin-bug-people. 
 > 
 > 
 >>Category:       bin
 >>Responsible:    bin-bug-people
 >>Synopsis:       yppush pushes map to himself triggering an error message each time
 >>Arrival-Date:   Fri Jun 08 12:25:00 +0000 2012
 >>
 > 
 > 


 -- 


 Dr. Nagler & Company GmbH
 Hauptstraße 9
 92253 Schnaittenbach

 Tel. +49 9622/71 97-42
 Fax +49 9622/71 97-50

 Wolfgang.Stukenbrock@nagler-company.com
 http://www.nagler-company.com


 Hauptsitz: Schnaittenbach
 Handelregister: Amberg HRB
 Gerichtsstand: Amberg
 Steuernummer: 201/118/51825
 USt.-ID-Nummer: DE 273143997
 Geschäftsführer: Dr. Martin Nagler, Dr. Dr. Karl-Kuno Kunze


From: Martin Husemann <martin@duskware.de>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: bin/46564: yppush pushes map to himself triggering an error message each time
Date: Fri, 8 Jun 2012 15:34:45 +0200

 On Fri, Jun 08, 2012 at 12:45:03PM +0000, Wolfgang Stukenbrock wrote:
 >  should be changed into
 >  errx(1, .... long (%lu - limit %lu)", (unsigned long)(strlen(...)), 
 >  (unsigned long)(sizeof(...) - 1));

 Please use %zu instead and don't cast.

 Martin

>Unformatted:

Home	PR Database Search