JMC-5506 Rule that detects fatal errors in the end of the recording
marcus.hirt at oracle.com
Mon Nov 5 19:46:11 UTC 2018
Just one small nit left:
Could you change core/org.openjdk.jmc.flightrecorder/src/main/resources/org/openjdk/jmc/flightrecorder/jdk/messages/internal/messages.properties to say JVM instead of VM?
No need for a new webrev.
From: Joshua Matsuoka <jmatsuok at redhat.com>
Date: Monday, 5 November 2018 at 18:07
To: Marcus Hirt <marcus.hirt at oracle.com>
Cc: <jmc-dev at openjdk.java.net>
Subject: Re: JMC-5506 Rule that detects fatal errors in the end of the recording
Here's an updated webrev:
On Sun, Nov 4, 2018 at 11:18 AM, Marcus Hirt <mailto:marcus.hirt at oracle.com> wrote:
Perfect. Josh, I've reviewed the code and it looks fine. Once you've updated
the texts to say JVM instead of VM, you're good to go. If you want to add code
for reason "No remaining non-daemon Java threads" (25), then that would be
great. Please let me know if you would like me to sponsor the change.
On 2018-11-04, 16:54, "Erik Gahlin" <mailto:erik.gahlin at oracle.com> wrote:
It is stated in the event. For details, look at the patch which is linked in the bug for the event.
> On 4 Nov 2018, at 12:43, Marcus Hirt <mailto:marcus.hirt at oracle.com> wrote:
> Hi Erik,
> Will Shutdown Reason state that the reason for the shutdown was due to no more
> remaining non-daemon threads? If so, the reason will be shown on the page.
> If we want it to be even more apparent, we could also add it to the rule
> as Severity.INFO (25). Either way is fine with me.
> If the Shutdown Reason doesn't state this explicitly, that should
> probably be fixed in the event.
> Kind regards,
> On 2018-11-02, 22:28, "jmc-dev on behalf of Erik Gahlin" <mailto:jmc-dev-bounces at openjdk.java.net on behalf of mailto:erik.gahlin at oracle.com> wrote:
> Hi Joshua,
> I think the text should say "JVM" instead of "VM"
> Calling the event "VM" was a mistake in JVM implementation and we intend
> to change the label of the event to "JVM", similar to other events, such
> as "JVM Information"
> One of the reason the event was added was to detect if there were no
> more remaining non-daemon Java threads  . Maybe that should be
> called out? A user may wonder why their application exited abruptly.
> It's not an error per se, but it would still be good to see somehow.
>  https://bugs.openjdk.java.net/browse/JDK-8041626
>  http://hg.openjdk.java.net/jdk/hs/rev/f413e471a6ab
>> The following patch adds support for the recently added VM Shutdown event
>> by creating a rule to detect if the JVM terminated in a VM Error, as well
>> as adding shutdown information to the JVM internals page.
>> - Josh
More information about the jmc-dev