Hello everybody,
I have wondered this several times before and searched the mailing list
to no avail, so I thought I'd just go ahead and ask:
Can somebody shed some more light on how to read the ttime metric that
haproxy supplies?
I am confident I understand qtime, ctime and rtime, but at least ttime I
don't understand. The docs say:
> 61. ttime [..BS]: the average total session time in ms over the 1024 >
last requests
Now, we are not using any session tracking at all. When I read this
value for a backend, how are the 1024 last requests mapped to session
times? Does one request map to one session?
If so, I am very concerned about how high the ttime values are compared
to rtime for some of our backend. What other factors, besides [qcr]time
are making up a sessions ttime?
If not, wouldn't some of the last 1024 requests belong to sessions that
are not yet over? If so, is any value for these sessions used to
calculate the final metric?
Any insights greatly appreciated,
Conrad
--
Conrad Hoffmann
Traffic Engineer
SoundCloud Ltd. | Rheinsberger Str. 76/77, 10115 Berlin, Germany
Managing Director: Alexander Ljung | Incorporated in England & Wales
with Company No. 6343600 | Local Branch Office | AG Charlottenburg |
HRB 110657B
