Question about benchmark finish and non daemon threads

[Aleksey Shipilev]

Got it, thanks. I assumed JMH process is stuck waiting for your
intervention anyway. But now I remember we kill off the JMH process once
the timeout is reached.

Submitted:
 https://bugs.openjdk.java.net/browse/CODETOOLS-7901456

-Aleksey

On 06/25/2015 06:05 PM, Zoltan Farkas wrote:
> I run benchmarks continuously, by the time I look at the results the
> process is long gone.
> 
> When this happens Java flight recorder data is not persisted, which
> is inconvenient.
> 
> To help with this, I am printing out all non daemon threads on
> benchmark teardown...
> 
> Thought it would be useful of jmh would provide this detail in case
> the benchmark is not finishing properly...
> 
> --z
> 
> 
>> On Jun 25, 2015, at 9:18 AM, Aleksey Shipilev
>> <aleksey.shipilev at oracle.com> wrote:
>> 
>> Hi Zoltan,
>> 
>>> On 06/25/2015 03:29 PM, Zoltan Farkas wrote: I see at the end of
>>> the benchmark:
>>> 
>>> <JMH had finished, but forked VM did not exit, are there stray 
>>> running threads? Waiting 14 seconds more…>
>>> 
>>> Would it be possible before to add some code to list the non
>>> daemon threads still running? would help with troubleshooting…
>> 
>> Is there a reason you can't use "jstack" against the forked VM to
>> figure this out?
>> 
>> -Aleksey.
>>