Try setting:
parsoidConfig.txstatsdHost = 'statsd.eqiad.wmnet'; parsoidConfig.txstatsdPort = 8125;
in localsettings.js and parse a file locally outside the WMF cluster. The workers die with the following error message.
{"name":"parsoid","hostname":"earth","pid":2964,"level":60,"logType":"fatal","process":{"name":"worker","pid":2964},"msg":"uncaught exception getaddrinfo ENOTFOUND","stack":"Error: getaddrinfo ENOTFOUND\n at errnoException (dns.js:37:11)\n at Object.onanswer [as oncomplete] (dns.js:124:16)","longMsg":"uncaught exception\ngetaddrinfo ENOTFOUND","time":"2016-02-24T16:40:45.567Z","v":0}This is broken behavior. We should catch the error and disable stats rather than crashing. This needs fixing for the ops quarterly goal of switching over from eqiad to codfw.