Instead of log parsing, I wrote an in-house plugin to collect runtime stats (e.g. req count, traffic) for each channel (map) weeks ago. It doesn't care about bandwidth, but simply records total transferred bytes for each channel and exposes the stats in json with a http interface. (Actually it's a combination of simple global plugin and official stats_over_http plugin). Then you could calculate the bandwidth yourself by retrieving the traffic every 10s, 5min or whatever you want with external program. If someone thought it useful, I'll try to put it somewhere if possible. On Nov 24, 2012 4:06 AM, "Ron Tsoref" <[email protected]> wrote:
> Hi. > > I'm using ATS for a couple of weeks now as a reverse proxy for 2 different > websites, and I'm trying to get some general statistics regarding the use > of it. > > I already parse the squid-formatted log file to get an overall picture of > requests, total GB served, etc. > > Another important thing I'm trying to get right now is the bandwidth > (Mbits/sec) used for each remap rule at any given time. > > Is there any way to get this information? Maybe a solution is already > built into ATS or some kind of an add-on to ATS is available to handle this? > > It seems like something trivial to look for, but it's very hard to find a > solution for this kind of stats, and not just for ATS, but rather to all > other caching solutions. > > Thanks, > > Ron >
