Extra metrics #34
Extra metrics #34
Conversation
| } | ||
|
|
||
| module.exports = function() { | ||
| var gauge = new Gauge('node_eventloop_lag_milliseconds', 'Lag of event loop in milliseconds.'); |
There was a problem hiding this comment.
could report this in nanoseconds instead?
|
Thanks, as always, a test for this would be great - i will probably mess stuff up sooner rather than later :) |
| var gauge = new Gauge('node_active_handles_total', 'Number of active requests.'); | ||
|
|
||
| return function() { | ||
| gauge.set(process._getActiveRequests().length); |
There was a problem hiding this comment.
_getActiveRequests and _getActiveHandles are new stuff to me, tried googling it and didn't find much info in node documentation. Is it a good practice to depend on "private" functions?
There was a problem hiding this comment.
It's the current recommended way. nodejs/node#1128 (comment)
There was a problem hiding this comment.
Could add an if guard like for process.cpuUsage in case it disappears in later nodes, if you want?
There was a problem hiding this comment.
That would be great! Just to as you said to safe guard us for the future
|
@siimon Should be GTG now. Took way more time than it should because timers didn't work. Fix for that included in its own commit. |
|
Lets just add a safeguard and I'll merge right away! |
|
@siimon Done! |
|
Merged and published to npm with a new major version! |
Builds on top of #25. I'll rebase after that's merged (or you can just merge this).
See the last commit in this PR for the diff in the meantime.
Ideas for extra metrics very much welcome!
At work we show more of
memoryUsage(but I think those are covered in #25, although just for Linux) andos.loadAvg[0], but I'm not sure if those are actually useful?