[lttng-dev] Using lttng for traces and (in-process) logging

[nolange79@gmail.com (Norbert Lange)]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1624 bytes --]

I think you understood, I would want to use lttng for tracing as-is,
but have another "consumer daemon" reading the buffer from the process
and piping (the filtered log-) events back.
After spending some more time, I think lttng has too big of a scope
(system level tracing) to add this per-process feature painless.
Am Fr., 10. Aug. 2018 um 18:24 Uhr schrieb Jérémie Galarneau
<jeremie.galarneau at efficios.com>:
>
> Hi,
>
> I can see how this could be approximated by configuring a "live" tracing
> session (i.e. you would see the log as they are produced), but there is
> no way, currently, to serialize the events to stdout/stderr from the
> application itself.
>
> Maybe I misunderstood your question. Let me know.
>
> Jérémie
>
> On 6 August 2018 at 18:42, Norbert Lange <nolange79 at gmail.com> wrote:
> > Hello,
> >
> > since lttng would make for a decent logger, is it possible to create a
> > own tracepoint provider and channel and acess it in the very same
> > process thats traced?
> > the log entries would need to be accessible directly to support some
> > existing schemes for manipulating (interpolating to a different
> > clock-source), forwarding and storing them.
> >
> > eg. I would like to access the logs and other traces with the normal
> > lttng tools, but also be able to get a copy of the logs directly from
> > the process generating them.
> > _______________________________________________
> > lttng-dev mailing list
> > lttng-dev at lists.lttng.org
> > https://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
>
>
>
> --
> Jérémie Galarneau
> EfficiOS Inc.
> http://www.efficios.com