[pgpool-hackers: 781] Re: main process is terminated after reload when enabeld_pool_hba is on

[Yugo Nagata]

Hi Usama

Thank you for the quick response!
I confirmed it is fixed by the last commit.

Thanks again!

On Thu, 29 Jan 2015 23:27:09 +0500
Muhammad Usama <m.usama at gmail.com> wrote:

> Hi Yugo
> 
> I have found the problem in the code and pushed the fix in 3.4 and master
> branches.
> http://git.postgresql.org/gitweb/?p=pgpool2.git;a=commit;h=2b55c09cbfb1488ecd0fe0b2d7d5045febea57bd
> 
> Although I was getting the segmentation fault with and not the exact bus
> error as reportred, So can you please verify if the issue is fixed at your
> end.
> 
> Thanks
> Kind regards!
> Muhammad Usama
> 
> On Thu, Jan 29, 2015 at 5:51 AM, Yugo Nagata <nagata at sraoss.co.jp> wrote:
> 
> > Hi hackers,
> >
> > I found a bug about pool_hba on 3.4.x.
> >
> > When enalbed_pool_hba is on, after several times of "pgpool reload"
> > pgpool main process is terminated leaving children.
> > Any log messages isn't output.
> >
> > gdb said "Bus error" occured at free_lines (at pool_hba.c).
> > Here is backtrace result.
> >
> > Program received signal SIGBUS, Bus error.
> > free_lines (lines=0x72e3e8, line_nums=0x72e3f0) at auth/pool_hba.c:1189
> > 1189                                    pfree(lfirst(token));
> > (gdb) bt
> > #0  free_lines (lines=0x72e3e8, line_nums=0x72e3f0) at auth/pool_hba.c:1189
> > #1  0x00000000004202c3 in load_hba (hbapath=0x7410e0
> > "/home/yugo-n/work/git/pgpool2/src/test/heathcheck/etc/pool_hba.conf") at
> > auth/pool_hba.c:107
> > #2  0x00000000004055e2 in reload_config () at main/pgpool_main.c:2950
> > #3  0x00000000004073fd in pool_sleep (second=<value optimized out>) at
> > main/pgpool_main.c:2307
> > #4  0x0000000000409e6f in PgpoolMain (discard_status=<value optimized
> > out>, clear_memcache_oidmaps=<value optimized out>) at
> > main/pgpool_main.c:473
> > #5  0x0000000000404431 in main (argc=<value optimized out>, argv=<value
> > optimized out>) at main/main.c:319
> >
> > This problem occurs only on 3.4.0. (I used head of git repository.)
> > 3.3.4 has no problem.
> >
> > Is this a bug of memory-management of 3.4?
> >
> > --
> > Yugo Nagata <nagata at sraoss.co.jp>
> > _______________________________________________
> > pgpool-hackers mailing list
> > pgpool-hackers at pgpool.net
> > http://www.pgpool.net/mailman/listinfo/pgpool-hackers
> >


-- 
Yugo Nagata <nagata at sraoss.co.jp>