Path: utzoo!mnetor!tmsoft!torsqnt!news-server.csri.toronto.edu!cs.utexas.edu!tut.cis.ohio-state.edu!snorkelwacker.mit.edu!bloom-beacon!eru!hagbard!sunic!mcsun!piet
From: piet@NIC.EU.net (Piet Beertema)
Newsgroups: comp.mail.sendmail
Subject: Re: sendmail 5.6[45] problems; clues anyone?
Message-ID: <2060@mcsun.eu.net>
Date: 22 Dec 90 13:17:57 GMT
References: <2059@mcsun.eu.net>
Organization: EUUG
Lines: 51

Here's a patch to main.c that sofar indeed seems to
have cured the problem with 5.6[45] that I described
in my previous article.

In main.c the environment is reset and the preceding
comment says that this is done after the thaw. However,
the code is such that it is *always* done, whether or
not the thaw() succeeded or was even done at all.
The result is that the environment also ends up in the
frozen config file, which could lead to problems when
the frozen config is used.

*** o.main.c	Sat Dec 22 03:29:37 1990
--- main.c	Sat Dec 22 05:50:34 1990
***************
*** 254,268 ****
  	OutChannel = stdout;
  
  #ifdef _PATH_SENDMAILFC
! 	if (!nothaw)
! 		readconfig = !thaw(FreezeFile);
  #endif /* _PATH_SENDMAILFC */
- 
- 	/* reset the environment after the thaw */
- 	for (i = 0; i < MAXUSERENVIRON && envp[i] != NULL; i++)
- 		UserEnviron[i] = newstr(envp[i]);
- 	UserEnviron[i] = NULL;
- 	environ = UserEnviron;
  
  #ifdef NAMED_BIND
  	/*
--- 254,267 ----
  	OutChannel = stdout;
  
  #ifdef _PATH_SENDMAILFC
! 	if (!nothaw && !(readconfig = !thaw(FreezeFile))) {
! 		/* reset the environment after successful thaw */
! 		for (i = 0; i < MAXUSERENVIRON && envp[i] != NULL; i++)
! 			UserEnviron[i] = newstr(envp[i]);
! 		UserEnviron[i] = NULL;
! 		environ = UserEnviron;
! 	}
  #endif /* _PATH_SENDMAILFC */
  
  #ifdef NAMED_BIND
  	/*

-------

-- 
	Piet Beertema, EUnet-NIC, Amsterdam   (piet@mcsun.EU.net)