You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@flink.apache.org by Prashant Nayak <pr...@gmail.com> on 2020/09/11 15:18:16 UTC

Flink UI not displaying records received/sent metrics

We are running a Flink 1.11.0 job cluster on Kubernetes.  We're not seeing
any metrics in the Flink Web UI (for the default metrics like Bytes
Received, Records Received, etc.), instead we see a spinner.  See image
below.

However, we have a prometheus metrics exporter configured and see job/task
metrics in Prometheus.

[image: Flink Web UI Spinner.png]

Looking into the network tab, for requests that retrieve those metrics, we
see

GET
/jobs/296402a32a8bbb1917279b5c2b2f40f1/vertices/cbc357ccb763df2852fee8c4fc7d55f2/watermarks

The response is an empty array

if I do this via CURL

GET
/jobs/296402a32a8bbb1917279b5c2b2f40f1/vertices/cbc357ccb763df2852fee8c4fc7d55f2
I get back
{
  "id": "cbc357ccb763df2852fee8c4fc7d55f2",
  "name": "Source: Custom Source -> Timestamps/Watermarks -> Filter",
  "now": 1599776105604,
  "parallelism": 1,
  "subtasks": Array[1][
    {
      "attempt": 0,
      "duration": 197613130,
      "end-time": -1,
      "host": "100.96.18.38",
      "metrics": {
        "read-bytes": 0,
        "read-bytes-complete": false,
        "read-records": 0,
        "read-records-complete": false,
        "write-bytes": 0,
        "write-bytes-complete": false,
        "write-records": 0,
        "write-records-complete": false
      },
      "start-time": 1599578492474,
      "start_time": 1599578492474,
      "status": "RUNNING",
      "subtask": 0,
      "taskmanager-id": "2eb0550f0d3bca170f76dd86f84843a0"
    }
  ]
}

Are we missing anything in our setup?  Any insight would be greatly
appreciated.
Thanks

Re: Flink UI not displaying records received/sent metrics

Posted by Robert Metzger <rm...@apache.org>.
Hi Prashant,

My initial suspicion is that this is a problem in the UI or with the
network connection from the browser to the Flink REST endpoints.

Since you can access the metrics with "curl", Flink seems to do
everything all right.

The first URL you posted is for the watermarks (it ends with
"/watermarks"). Can you make sure that the URL you check in the browser is
the same as you've checked with "curl"?

How are you connecting into your Kubernetes cluster from your machine?
(kubectl proxy?)

Best,
Robert





On Fri, Sep 11, 2020 at 5:19 PM Prashant Nayak <pr...@gmail.com>
wrote:

> We are running a Flink 1.11.0 job cluster on Kubernetes.  We're not seeing
> any metrics in the Flink Web UI (for the default metrics like Bytes
> Received, Records Received, etc.), instead we see a spinner.  See image
> below.
>
> However, we have a prometheus metrics exporter configured and see job/task
> metrics in Prometheus.
>
> [image: Flink Web UI Spinner.png]
>
> Looking into the network tab, for requests that retrieve those metrics, we
> see
>
> GET
> /jobs/296402a32a8bbb1917279b5c2b2f40f1/vertices/cbc357ccb763df2852fee8c4fc7d55f2/watermarks
>
> The response is an empty array
>
> if I do this via CURL
>
> GET
> /jobs/296402a32a8bbb1917279b5c2b2f40f1/vertices/cbc357ccb763df2852fee8c4fc7d55f2
> I get back
> {
>   "id": "cbc357ccb763df2852fee8c4fc7d55f2",
>   "name": "Source: Custom Source -> Timestamps/Watermarks -> Filter",
>   "now": 1599776105604,
>   "parallelism": 1,
>   "subtasks": Array[1][
>     {
>       "attempt": 0,
>       "duration": 197613130,
>       "end-time": -1,
>       "host": "100.96.18.38",
>       "metrics": {
>         "read-bytes": 0,
>         "read-bytes-complete": false,
>         "read-records": 0,
>         "read-records-complete": false,
>         "write-bytes": 0,
>         "write-bytes-complete": false,
>         "write-records": 0,
>         "write-records-complete": false
>       },
>       "start-time": 1599578492474,
>       "start_time": 1599578492474,
>       "status": "RUNNING",
>       "subtask": 0,
>       "taskmanager-id": "2eb0550f0d3bca170f76dd86f84843a0"
>     }
>   ]
> }
>
> Are we missing anything in our setup?  Any insight would be greatly
> appreciated.
> Thanks
>