weird stats tail behavior

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

weird stats tail behavior

carlhoerberg
We have a "constant spike" at the end of our charts:

https://www.dropbox.com/s/8n7p2ih0rhm6nre/Screenshot%202013-12-12%2017.45.41.png
https://www.dropbox.com/s/y637pm0w66lskm0/Screenshot%202013-12-12%2017.46.08.png
https://www.dropbox.com/s/g17tna9d390oncn/Screenshot%202013-12-12%2017.46.59.png

There's always a spike at the end of the 10min chart, no matter how long you look at it.

Also note that the current msgs rates shows 0, while in fact the pub rate is 400msg/s.  
Reply | Threaded
Open this post in threaded view
|

Re: weird stats tail behavior

Tim Watson-6
Hi Carl,

So sorry for the delayed response - this slipped through the xmas cracks I'm afraid.

On 12 Dec 2013, at 10:51, carlhoerberg wrote:
> We have a "constant spike" at the end of our charts:
>
> https://www.dropbox.com/s/8n7p2ih0rhm6nre/Screenshot%202013-12-12%2017.45.41.png
> https://www.dropbox.com/s/y637pm0w66lskm0/Screenshot%202013-12-12%2017.46.08.png
> https://www.dropbox.com/s/g17tna9d390oncn/Screenshot%202013-12-12%2017.46.59.png
>
> There's always a spike at the end of the 10min chart, no matter how long you
> look at it.
>

Hmn, interesting. Can you tell us anything more about the environment/configuration? Do you have logs from these time periods to share?

> Also note that the current msgs rates shows 0, while in fact the pub rate is
> 400msg/s.  
>

There have been a few bug fixes to the management plugin since 3.1.5 - do you observe the same behaviour with more recent versions? (I do realise that your production environments don't always make it easy/possible to answer that kind of question though!).

Cheers,
Tim
_______________________________________________
rabbitmq-discuss mailing list
[hidden email]
https://lists.rabbitmq.com/cgi-bin/mailman/listinfo/rabbitmq-discuss