ARI method of posting log messages to Asterisk

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Wed, 2013-10-16 at 13:49 -0500, Mark Michelson wrote:
> Hey folks,
> 
> Have a look at ASTERISK-22699 [1]
> 
> The issue pertains to allowing for ARI to write content to Asterisk's 
> logs. I'd like to propose my ideas for how logging from ARI could be 
> done. First, let's consider what an Asterisk log message looks like:
> 
> [Oct 16 17:20:23] ERROR[20023] foo.c:666 I AM ERROR.
> 
> The message consists of 6 parts: the timestamp, the level, the thread 
> ID, the file name, the line number, and the message. Of these, the 
> timestamp and thread ID should be filled in by Asterisk. The other four 
> parts could potentially be fed into Asterisk by an ARI application.
> 
> Moving on to the ARI method for logging to the Asterisk log, I think it 
> should be something like:
> 
> POST /asterisk/logMessage
> 
> At a minimum, the parameters for this would be
> 
> level: One of the Asterisk log levels (such as "error", "warning", 
> "debug", etc.)
> message: The string to print to the log
> 
> The minimum is nice in that it keeps the API call simple, but it has the 
> downside of making your messages look like this:
> 
> [Oct 16 17:20:23] WARNING[20023] ?:0 You're tearing me apart, Lisa!
> 
> We could expand the method a bit by adding in a couple more parameters:
> 
> fileName: The file name to place in the log message.
> lineNumber: The line number to place in the log message.
> 
> This way, the previous log message would look like this instead:
> 
> [Oct 16 17:20:23] WARNING[20023] cool_app.py:143 You're tearing me 
> apart, Lisa!
> 
> Expanding further, consider that log levels such as debug and verbose 
> also have an additional integer value associated with them so that they 
> are only printed depending on what you've set core debug or core verbose 
> to. We could just make the assumption that all verbose and debug 
> messages logged through ARI have a 0 for this value, or this integer 
> value could be exposed as part of the ARI method as well.
> 
> So what do you thing of this proposal? What degree of control would you 
> prefer to have for writing log messages to Asterisk's logger?
> 
> 
> As a second discussion, consider that Asterisk, since 1.8 I believe, has 
> had the concept of dynamic log levels. What this means is that a module 
> can register a log level called "FOO" for instance, and log messages 
> directed at that level would look like:
> 
> [Oct 16 17:20:23] FOO[20023] bar.c:717 They're eating her! And then 
> they're going to eat me! OH MY GOOOOOOOOOOOOOOOOOOD!
> 
> This could be a useful concept for ARI in one of two possible ways.
> 
> 1) The ARI implementation in Asterisk could register a log level at 
> startup (called "ARI" possibly) that ARI app developers could write to 
> in order to keep their messages at a different level from any of the 
> other messages in the system.
> 
> 2) We could expose an ARI call that would allow app developers to create 
> their own dynamic log levels on the fly. For instance, you could have
> 
> POST /asterisk/logLevel
> 
> with the parameter
> 
> level: The name of the level to register to Asterisk
> 
> This way you could record different log messages at different dynamic 
> log levels if you desired.
> 
> Would dynamic log levels be useful for you in your applications? If so, 
> would you prefer to have a dedicated "ARI" level to write to or would 
> you prefer the ability to create your own levels to write to?
> 
> _______________________________________________
> asterisk-app-dev mailing list
> asterisk-app-dev at lists.digium.com
> http://lists.digium.com/cgi-bin/mailman/listinfo/asterisk-app-dev


I don't personally see a need for logging functionality within ARI.
There are so many ways in any language you can think of to log to
files/database/etc. that complicating the ARI in order to pollute
asterisk's logs more seems unnecessary.

- Brad


[Index of Archives]     [Asterisk SS7]     [Asterisk Announcements]     [Asterisk Users]     [PJ SIP]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Security]     [Bugtraq]     [Linux]     [Linux OMAP]     [Linux MIPS]     [ECOS]     [Linux API]

  Powered by Linux