From: dgoulet@efficios.com (David Goulet)
Subject: [lttng-dev] LTTng Tools 2.1 streaming commands
Date: Fri, 05 Oct 2012 09:13:37 -0400 [thread overview]
Message-ID: <506EDD01.60704@efficios.com> (raw)
In-Reply-To: <506EC818.7050708@ericsson.com>
Right.
I guess this discussion can continue and we can come up with a community
consensus.
This is a matter of UI naming so it should be decided now before we
release stable since for the complete 2.x cycle we'll have to support this.
Anyone wants to ship in here?
Thanks!
David
eamcs/eedbhu:
> Hi David
>
> thanks for the detailed explanation.
>
> I'm not debating the purpose of the command or that a 2 step approach is
> supported for enabling the consumer. As you described it makes sense to
> have this possibility.
>
> However I'm debating mainly the name of the command "enable-consumer".
> The command suggests that the consumer will be enabled when executing
> the command. I foresee that a lot of users expect that the consumer is
> enabled after the command execution. I think, a different name would
> improve the user experience.
>
> Best Regards
> Bernd
>
> On 10/03/2012 02:27 PM, David Goulet wrote:
>> Hi Bernd,
>>
>> The enable-consumer, by default, always enable a consumer. We added the
>> "extended options" such as the -U/-C/-D/-e to control each part of the
>> API. So let say,
>>
>> $ lttng enable-consumer -k net://localhost
>>
>> This command does two API calls underneath which are
>> lttng_set_consumer_uri and lttng_enable_consumer.
>>
>> However, the set_consumer_uri can be arbitrary long because it has to
>> connect to the relayd (if remote) and set the session. It adds "unknown"
>> latency to the command. So, for someone willing to control the full time
>> window of the streaming setup using the API, it is divided in two calls.
>>
>> This is why we added the extended options so the command line UI could
>> also be controlled on a per API call basis.
>>
>> Is it clear enough?
>>
>> Reply continues below:
>>
>> Bernd Hufmann:
>>> Hello
>>>
>>> For the support of LTTng Tools 2.1 in Eclipse, I'm currently trying to
>>> understand how to use the configuration for network streaming with the
>>> updated "lttng create"-command and new "enable-consumer"-command.
>>>
>>> a) lttng enable-consumer
>>> I find this command confusing because this command does not always
>>> enables the consumer, even if the command name implieeees so. The
>>> enabling
>>> actually depends on how the command is executed.
>>> Examples:
>>>
>>> * "lttng enable-consumer -k -U net://<remote_addr>" or "lttng
>>> enable-consumer -k -C tcp://<remote_addr> -D tcp://<remote_addr>"
>>> don't enable the consumer. You need to either add option
>>> --enable or
>>> execute subsequently "lttng enable-consumer --enable"
>>> * lttng enable-consumer -k net://<remote_addr> does enable the
>>> consumer. I took me a while to figure out the difference to the
>>> example above: The option -U is omitted.
>>>
>>>
>>> What the command actually provides, is 2 features: A way to configure
>>> streaming (e.g. remote_addr) and a way to enable the consumer. Would it
>>> be better to name it to "lttng configure-consumer"? Also, remove the
>>> support of the possibility to not specify -U, -C or -D. The following
>>> variants of this command should be enough:
>>> lttng configure-consumer -k -U<remote_addr> [--enable]
>>> lttng configure-consumer -k -C<remote_addr> -D<remote_addr> [--enable]
>>> lttng configure-consumer -k --enable
>>> lttng configure-consumer -u -U<remote_addr> [--enable]
>>> lttng configure-consumer -u -C<remote_addr> -D<remote_addr> [--enable]
>>> lttng configure-consumer -u --enable
>>>
>>> Please let me know what you think.
>>>
>>> b) lttng create [-U<remote_addr>] | [-C<remote_addr> -D<remote_addr>]
>>> [--no-consumer] [--disable-consumer]
>>>
>>> * Are options --no-consumer or --disable-consumer only applicable for
>>> streaming?
>> No, also for local consumer.
>>
>>> * I'm not sure what is the purpose of the options --no-consumer or
>>> --disable-consumer. Could you please explain the use cases for
>>> using
>>> --no-consumer or --disable-consumer?
>> This basically disable the consumer for a tracing session. It's not very
>> useful for now but for upcoming snapshots and live tracing, it will make
>> way more sense! :).
>>
>> Again, same idea, the API can control the consumer "state"
>> (enable/disable), so we added these options for the UI.
>>
>> Cheers!
>> David
>>
>>>
>>> Thanks
>>> Bernd
>>>
>>> This Communication is Confidential. We only send and receive email on
>>> the basis of the terms set out at _www.ericsson.com/email_disclaimer_
>>>
>>>
>>>
>>>
>>> This body part will be downloaded on demand.
>
>
> _______________________________________________
> lttng-dev mailing list
> lttng-dev at lists.lttng.org
> http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
next prev parent reply other threads:[~2012-10-05 13:13 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-10-03 18:17 Bernd Hufmann
2012-10-03 18:27 ` David Goulet
2012-10-05 11:44 ` eamcs/eedbhu
2012-10-05 13:13 ` David Goulet [this message]
2012-10-09 19:08 ` David Goulet
2012-10-09 21:18 ` Matthew Khouzam
2012-10-10 13:20 ` Bernd Hufmann
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=506EDD01.60704@efficios.com \
--to=dgoulet@efficios.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox