Your message dated Tue, 14 Jun 2011 21:02:24 +0200
with message-id <[email protected]>
and subject line Re: [Pkg-openssl-devel] Bug#624254: Bug#624254: libssl1.0.0:
segfault when attempting a secured PostgreSQL connection
has caused the Debian Bug report #624254,
regarding libssl1.0.0: segfault when attempting a secured PostgreSQL connection
to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact [email protected]
immediately.)
--
624254: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=624254
Debian Bug Tracking System
Contact [email protected] with problems
--- Begin Message ---
Package: libssl1.0.0
Version: 1.0.0d-2
Severity: important
Coin,
In a program, i'm connecting to a remote PostgreSQL server using TLS,
which now gives the following result:
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7f384f495700 (LWP 14122)]
0x0000000000000000 in ?? ()
(gdb) bt
#0 0x0000000000000000 in ?? ()
#1 0x00007f38499c41d5 in int_update (ctx=<value optimized out>,
data=<value optimized out>, count=<value optimized out>) at
hm_pmeth.c:144
#2 0x00007f3849d3114b in tls1_mac (ssl=0x1118eb0,
md=0x1826f81
"\307+\225\336\070\240\004\210~\322]\345\026\245\341\325\354\016\034\024ZP\r3$\351R\257\277Q[\033\312ƾn\254~\242I2O\006\065\221yƀ\202\252\275<\352Xf\235\277\332\321s", <incomplete sequence \306>, send=1) at
t1_enc.c:932
#3 0x00007f3849d28e86 in do_ssl3_write (s=0x1118eb0, type=22,
buf=0x197b5b0 "\024", len=16, create_empty_fragment=0) at s3_pkt.c:771
#4 0x00007f3849d28fe6 in ssl3_write_bytes (s=0x1118eb0, type=22,
buf_=0x197b5b0, len=<value optimized out>) at s3_pkt.c:603
#5 0x00007f3849d2a422 in ssl3_do_write (s=0x1118eb0, type=22) at
s3_both.c:132
#6 0x00007f3849d24fc4 in ssl3_connect (s=0x1118eb0) at s3_clnt.c:456
#7 0x00007f3849f7f8b3 in open_client_SSL (conn=0x126e670) at fe-secure.c:1161
#8 0x00007f3849f7df19 in pqsecure_open_client (conn=0x126e670) at
fe-secure.c:284
#9 0x00007f3849f689a3 in PQconnectPoll (conn=0x126e670) at fe-connect.c:1926
#10 0x00007f3849f67bc5 in connectDBComplete (conn=0x126e670) at
fe-connect.c:1359
#11 0x00007f3849f661e3 in PQconnectdb (conninfo=0x126e5a0 "host='xxx'
port='5432' dbname='xxx' user='xxx' password='xxx'") at fe-connect.c:400
[…]
The program as not changed, and worked using libpq5 9.0.3-1. What made
me think it is a libssl bug and not a postgresql is it appeared right
after upgrading libpq5 to 9.0.3-1+b1, and the reason for this binary
rebuild was (according to wb): Rebuild against libssl1.0.0.
Btw, as the debian/rules does not honour the nocheck build option, and
the upstream build system is custom, i had no quick way to disable
optimizations.
Regards.
--
Marc Dequènes (Duck)
pgpydQ6OXspvk.pgp
Description: PGP Digital Signature
--- End Message ---
--- Begin Message ---
Coin,
Sorry for the delay.
Quoting Kurt Roeckx <[email protected]>:
I can't reproduce this using psql. I get a working ssl
connection.
I was not able to reproduce this problem with psql either. Rebuilding
libpgsql-ruby for PG9 did not help and i did not find any clue.
I just noticed #624106 while searching, and it seems to have revolved
my problem.
I'm closing this issue. Thanks.
--
Marc Dequènes (Duck)
pgpJ46BdxPeIS.pgp
Description: PGP Digital Signature
--- End Message ---
