[wildfly-dev] WFCORE-4807 - Log Routing

[James Perkins]

On Tue, Aug 11, 2020 at 11:57 AM Brian Stansberry <
brian.stansberry at redhat.com> wrote:

>
>
> On Tue, Aug 11, 2020 at 12:13 PM James Perkins <jperkins at redhat.com>
> wrote:
>
>>
>>
>> On Tue, Aug 11, 2020 at 9:32 AM Brian Stansberry <
>> brian.stansberry at redhat.com> wrote:
>>
>>> FWIW there are 3 options:
>>>
>>> 1) Default on.
>>> 2) Default off, not specified in standard configs.
>>> 3) Default off, but set to on in standard configs.
>>>
>>> The distinction between 1 and 3 is people bringing an existing config do
>>> not see a behavior change.
>>>
>>
>> Yes. What I have locally is a config option. I'm starting to think 3 is
>> the best option.
>>
>>
>>>
>>> No matter which we do we should aim to get this in very early in a
>>> release dev cycle so we have time to notice any odd impacts.
>>>
>>> Do you have any thoughts on how we can look for odd impacts?  IIRC the
>>> odd impact would be stuff we expect to appear in the server.log no longer
>>> appearing there.
>>>
>>
>> The only thing I can think of would be creating a deployment with it's
>> own logging configuration and examining the logs. Essentially any time
>> anything is logged with the deployments class loader it would end up in the
>> deployments log configuration. Assuming of course the deployment is using
>> some sort of per-deployment logging.
>>
>
> I suppose things like the kitchen-sink QS app, maybe a few others, would
> touch enough areas to get a good sense of what would happen.
>

Ah good call. I could also probably hard-code something and run the test
suite to see what happens as well which is probably not a bad idea.


>
>
>> It's definitely a change in behavior which is why I'm starting to think
>> we should default to off.
>>
>
> Input from people who support WildFly would be good, particularly if we
> have a sense of what messages might move. People doing support know best
> how much hassle it would be if messages were no longer in the server.log.
>

Yes. That's who I'd really like to hear from. I've seen forum
posts/questions a lot over the years around "why does this not log to my
file" so it would be good to know if what we end up with is sufficient or
not.


>
>
>>
>>>
>>> On Mon, Aug 10, 2020 at 5:50 PM James Perkins <jperkins at redhat.com>
>>> wrote:
>>>
>>>> Hello All,
>>>> I've got an open proposal for some changes with how log messages may be
>>>> routed [1]. There is some discussion on there with some concerns about
>>>> messages that the user may not expect to be routed to their own specific
>>>> log context. Feedback from a wider audience would be appreciated.
>>>>
>>>> Currently I've got this enabled by default which would be a change in
>>>> behavior. However, as I've stated in a comment on the proposal, maybe it's
>>>> better to opt-in than opt-out. Either way any feedback would be appreciated.
>>>>
>>>> Thanks in advance.
>>>>
>>>> [1]: https://github.com/wildfly/wildfly-proposals/pull/281
>>>>
>>>> --
>>>> James R. Perkins
>>>> JBoss by Red Hat
>>>> _______________________________________________
>>>> wildfly-dev mailing list
>>>> wildfly-dev at lists.jboss.org
>>>> https://lists.jboss.org/mailman/listinfo/wildfly-dev
>>>
>>>
>>>
>>> --
>>> Brian Stansberry
>>> Manager, Senior Principal Software Engineer
>>> Red Hat
>>>
>>
>>
>> --
>> James R. Perkins
>> JBoss by Red Hat
>>
>
>
> --
> Brian Stansberry
> Manager, Senior Principal Software Engineer
> Red Hat
>


-- 
James R. Perkins
JBoss by Red Hat
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.jboss.org/pipermail/wildfly-dev/attachments/20200811/1af0c391/attachment.html