Re: SEGFAULT in CREATE EXTENSION related pg_init_privs - Mailing list pgsql-hackers

From Pavel Stehule
Subject Re: SEGFAULT in CREATE EXTENSION related pg_init_privs
Date
Msg-id CAFj8pRCNsJ3szUfsHZHy+EyrfZy13_h44J1o6wNhXXbdR1i+Uw@mail.gmail.com
Whole thread Raw
In response to Re: SEGFAULT in CREATE EXTENSION related pg_init_privs  (Peter Geoghegan <pg@bowt.ie>)
List pgsql-hackers


2016-04-14 10:49 GMT+02:00 Peter Geoghegan <pg@bowt.ie>:
On Thu, Apr 14, 2016 at 1:27 AM, Pavel Stehule <pavel.stehule@gmail.com> wrote:
>  (gdb) bt
> #0  heap_deform_tuple (tuple=tuple@entry=0x1d87e90,
> tupleDesc=tupleDesc@entry=0x7f1dab9525b0, values=values@entry=0x1d87a28,
>     isnull=isnull@entry=0x1d882d8 "\177\177\177\177\177~\177\177") at
> heaptuple.c:881

The memory appears to be freed -- CLOBBER_FREED_MEMORY stuff does this.

You might try using Valgrind, with --track-origins, which will show
where the memory was first allocated.

I did it, but there is nothing interesting

 ==00:00:00:03.519 17221== For counts of detected and suppressed errors, rerun with: -v
==00:00:00:03.519 17221== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0)
==00:00:01:09.357 17228== Invalid read of size 2
==00:00:01:09.357 17228==    at 0x477A15: heap_deform_tuple (heaptuple.c:881)
==00:00:01:09.357 17228==    by 0x477ED9: heap_modify_tuple (heaptuple.c:817)
==00:00:01:09.357 17228==    by 0x516DDA: recordExtensionInitPriv.part.2 (aclchk.c:5305)
==00:00:01:09.357 17228==    by 0x51954F: recordExtensionInitPriv (aclchk.c:5262)
==00:00:01:09.357 17228==    by 0x51954F: ExecGrant_Relation (aclchk.c:1965)
==00:00:01:09.357 17228==    by 0x51A324: ExecGrantStmt_oids (aclchk.c:564)
==00:00:01:09.357 17228==    by 0x51BDFE: ExecuteGrantStmt (aclchk.c:549)
==00:00:01:09.357 17228==    by 0x6F88F8: ProcessUtilitySlow.isra.5 (utility.c:1514)
==00:00:01:09.357 17228==    by 0x6F76BA: standard_ProcessUtility (utility.c:907)
==00:00:01:09.357 17228==    by 0x58ED9C: execute_sql_string (extension.c:748)
==00:00:01:09.357 17228==    by 0x58ED9C: execute_extension_script.isra.5 (extension.c:908)
==00:00:01:09.357 17228==    by 0x58FC71: CreateExtensionInternal.isra.6 (extension.c:1500)
==00:00:01:09.357 17228==    by 0x5901D7: CreateExtension (extension.c:1558)
==00:00:01:09.357 17228==    by 0x6F8378: ProcessUtilitySlow.isra.5 (utility.c:1296)
==00:00:01:09.357 17228==  Address 0x7f7f7f7f7f7f7f91 is not stack'd, malloc'd or (recently) free'd
==00:00:01:09.357 17228==.
{
   <insert_a_suppression_name_here>
   Memcheck:Addr2
   fun:heap_deform_tuple
   fun:heap_modify_tuple
   fun:recordExtensionInitPriv.part.2
   fun:recordExtensionInitPriv
   fun:ExecGrant_Relation
   fun:ExecGrantStmt_oids
   fun:ExecuteGrantStmt
   fun:ProcessUtilitySlow.isra.5
   fun:standard_ProcessUtility
   fun:execute_sql_string
   fun:execute_extension_script.isra.5
   fun:CreateExtensionInternal.isra.6
   fun:CreateExtension
   fun:ProcessUtilitySlow.isra.5
}

Regards

Pavel

--
Peter Geoghegan

pgsql-hackers by date:

Previous
From: Kyotaro HORIGUCHI
Date:
Subject: Re: Support for N synchronous standby servers - take 2
Next
From: Christoph Berg
Date:
Subject: Re: \crosstabview fixes