Rajkumar S

Duplicate of #54

After running about 36 hours, the graphs trends are the same. Throughput decreases steadily. Not sure what could be wrong here.

That sounds good. I will run this tonight and update tomorrow. Eventually, I think, some sort of time series db like rrd would be a nice addition to tackle the...

IMHO all these should be outside snabb, separate daemon can query the `/var/run/snabb/$pid` and do stuff like populate in RRD(and other time series db like influxdb) , provide SNMP etc......

I have captured 15 hours worth of shm in tar.xz files using the script.

Put the vmprofile data in a spread sheet. I have selected only the _interesting_ data here, and removed all which do not show much variation or big enough value. ![screenshot...

@lukego Thanks for the fine docs, I will follow the docs and see if I can identify it. One thing that I find strange is that the percentage of the...

Slightly different error here: ```lua module(..., package.seeall) local lib = require("core.lib") local logger = lib.logger_new({ rate = 32, module = 'lpm'}); local lpm4 = require("lib.lpm.lpm4_dxr").LPM4_dxr Route = {} function Route:new...

On Fri, Dec 1, 2017 at 3:23 PM, Luke Gorrie wrote: Do you ever see raptorjit moving to using an 128 bit word? I am not sure how feasible or...

I have worked on this some more and have some additional observations. To prevent the `failed to allocate mcode memory` I have increased the luajit tuneables in commit a268212. I...