RFR (S): 8237777 "Dumping core ..." is shown despite claiming that "# No core dump will be written."

gerard ziemski gerard.ziemski at oracle.com
Thu Apr 23 15:33:13 UTC 2020

hi Yasumasa,

Thank you for your feedback!

On 4/22/20 7:45 PM, Yasumasa Suenaga wrote:
> Hi Gerard,
> os_linux.cpp:
>      if (core_pattern[0] == '|') {
> +      // on some systems, PID is not included by default, so don't 
> add it here either
>        written = jio_snprintf(buffer, bufferSize,
> -                             "\"%s\" (or dumping to %s/core.%d)",
> -                             &core_pattern[1], p, current_process_id());
> +                             "\"%s\", or dumping to %s/core",
> +                             &core_pattern[1], p);
> I think it is better to "%s" instead of "%s/core".
> I added this code is for the behavior of ABRT. If the handler does not 
> add extension to core file, it is better to show default location 
> (directory).

I like your approach better than mine, but in the end I decided to leave 
this particular file alone.

>>> I have filed JDK-8243196 to address related issues next.
>> I don't think we have an issue to fix. We do not want to be parsing 
>> the core pattern to go and determine what program is going to be used 
>> to process core files and then whether that program is installed or 
>> not, or then enabled or not.
> In Fedora / RHEL / CentOS (maybe Oracle Linux) has different coredump 
> handler such as systemd-coredump, and it is enabled by default.
> I left my comment to JBS about them.

Thank you for the info.


More information about the hotspot-runtime-dev mailing list