"up.time" metric from /v1/servers/self endpoint

Not applicable

While troubleshooting an issue, we noticed that the "up.time" metric on our MPs and Routers was showing a value of < 1 day except the process has been up for weeks. The "started.at" metric aligns with the process start time.

What does the "up.time" metric mean?

Example:

$ date
Wed Oct 18 16:28:46 PDT 2017

$ ps -o lstart $(pgrep -f message)
                 STARTED
Tue Oct  3 21:57:12 2017

$ curl -s http://localhost:8082/v1/servers/self | grep -A 1 -e 'started.at' -e '"up.time"' -e type
      "name" : "started.at",
      "value" : "1507093036724"
--
      "name" : "up.time",
      "value" : "18 hours 30 minutes 7 seconds "
--
  "type" : [ "message-processor" ],

We are using OPDK 4.16.01.07

cc @Matias

Thanks,

Eric

2 1 234
1 REPLY 1

Not applicable

@Eric Dahl ,

The below is the up.time from one of the routers on my 1701 planet.
The up.time = (present (system) time - started.at) in my case which is the up time of the router from the last restart.

I might be wrong but as you are on 1601 and you are getting a wrong up.time value makes me think it was a bug in 1601.

{
      "name" : "started.at",
      "value" : "1508187575869"
    }, {
      "name" : "startup.interval",
      "value" : "12240"
    }, {
      "name" : "rpc.port",
      "value" : "4527"
    }, {
      "name" : "Profile",
      "value" : "Router"
    }, {
      "name" : "startup.time",
      "value" : "12 seconds "
    }, {
      "name" : "up.time",
      "value" : "2 days 3 hours 9 minutes 29 seconds"