[pgpool-general: 7697] Re: Segmentation fault in watchdog child process
m.usama at gmail.com
Fri Sep 10 15:49:53 JST 2021
Hi Ishii San,
Thanks for pointing that out. Unfortunately, GCC version on my MacOS does
support '-Wmisleading-indentation' so I didn't get any warning.
I will take care of the indentation before committing the patch.
On Fri, Sep 10, 2021 at 7:19 AM Tatsuo Ishii <ishii at sraoss.co.jp> wrote:
> Hi Usama,
> Thank you for the patch.
> You might want to fix below before committing the patch:
> wd_json_data.c: In function ‘parse_wd_exec_cluster_command_json’:
> wd_json_data.c:934:2: warning: this ‘if’ clause does not guard...
> 934 | if (*args_list)
> | ^~
> wd_json_data.c:936:3: note: ...this statement, but the latter is
> misleadingly indented as if it were guarded by the ‘if’
> 936 | *args_list = NULL;
> | ^
> Tatsuo Ishii
> SRA OSS, Inc. Japan
> English: http://www.sraoss.co.jp/index_en.php
> > Hi,
> > It was a silly coding mistake that was causing the crash. I have cooked
> > a fix and currently testing it.
> > Attached are the patches for master and 4.2 branches, If you want to try
> > these to verify if the fix works for your scenario as well.
> > Thanks
> > Best Regards
> > On Thu, Sep 9, 2021 at 1:15 PM Tatsuo Ishii <ishii at sraoss.co.jp> wrote:
> >> >> Do you use watchdog and enable follow_primary_command? If so, You are
> >> >> right. We identified the issue with 4.2.4 and one of Pgpool-II
> >> >> developers is working on it. Sorry for the inconvenience.
> >> >>
> >> >
> >> > Yes, we the watchdog with follow_primary_command, so this might be the
> >> > issue you identified. Do you already have a patch I might try?
> >> Not yet. But Usama, one of Pgpool-II developers, (he is an expert
> >> on watchdog) is working on it now.
> >> --
> >> Tatsuo Ishii
> >> SRA OSS, Inc. Japan
> >> English: http://www.sraoss.co.jp/index_en.php
> >> Japanese:http://www.sraoss.co.jp
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the pgpool-general