From: Tom Lane <tgl@sss.pgh.pa.us>
Date: Sat, 21 Jul 2007 22:12:04 +0000 (+0000)
Subject: Fix elog.c to avoid infinite recursion (leading to backend crash) when
X-Git-Tag: REL8_3_BETA1~409
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=a0dab332a2e1961f45b38b23bd428859621e6f3c;p=postgresql

Fix elog.c to avoid infinite recursion (leading to backend crash) when
log_min_error_statement is active and there is some problem in logging the
current query string; for example, that it's too long to include in the log
message without running out of memory.  This problem has existed since the
log_min_error_statement feature was introduced.  No doubt the reason it
wasn't detected long ago is that 8.2 is the first release that defaults
log_min_error_statement to less than PANIC level.
Per report from Bill Moran.
---

diff --git a/src/backend/utils/error/elog.c b/src/backend/utils/error/elog.c
index 11231c90d6..3b86de353b 100644
--- a/src/backend/utils/error/elog.c
+++ b/src/backend/utils/error/elog.c
@@ -42,7 +42,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/utils/error/elog.c,v 1.189 2007/07/19 21:58:12 tgl Exp $
+ *	  $PostgreSQL: pgsql/src/backend/utils/error/elog.c,v 1.190 2007/07/21 22:12:04 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -240,10 +240,15 @@ errstart(int elevel, const char *filename, int lineno,
 
 		/*
 		 * If we recurse more than once, the problem might be something broken
-		 * in a context traceback routine.	Abandon them too.
+		 * in a context traceback routine.  Abandon them too.  We also
+		 * abandon attempting to print the error statement (which, if long,
+		 * could itself be the source of the recursive failure).
 		 */
 		if (recursion_depth > 2)
+		{
 			error_context_stack = NULL;
+			debug_query_string = NULL;
+		}
 	}
 	if (++errordata_stack_depth >= ERRORDATA_STACK_SIZE)
 	{