You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@mesos.apache.org by "Adam Cecile (JIRA)" <ji...@apache.org> on 2016/11/08 08:11:58 UTC

[jira] [Created] (MESOS-6559) Mesos slave /health reports 200 while failing to register to master

Adam Cecile created MESOS-6559:
----------------------------------

             Summary: Mesos slave /health reports 200 while failing to register to master
                 Key: MESOS-6559
                 URL: https://issues.apache.org/jira/browse/MESOS-6559
             Project: Mesos
          Issue Type: Bug
          Components: slave
    Affects Versions: 1.0.1
            Reporter: Adam Cecile


Hello,

I'm currently having a mesos-slave which is undable to register against the master for some reasons:

{quote}
Nov 08 09:07:46 mario mesos-slave[36438]: W1108 09:07:46.660585 36449 slave.cpp:1141] Already registered with master master@10.99.50.1:5050
Nov 08 09:07:46 mario mesos-slave[36438]: I1108 09:07:46.660655 36449 slave.cpp:1155] Forwarding total oversubscribed resources
Nov 08 09:07:57 mario mesos-slave[36438]: W1108 09:07:57.418692 36447 slave.cpp:1141] Already registered with master master@10.99.50.1:5050
Nov 08 09:07:57 mario mesos-slave[36438]: I1108 09:07:57.418745 36447 slave.cpp:1155] Forwarding total oversubscribed resources
Nov 08 09:07:58 mario mesos-slave[36438]: W1108 09:07:58.978473 36456 slave.cpp:1141] Already registered with master master@10.99.50.1:5050
Nov 08 09:07:58 mario mesos-slave[36438]: I1108 09:07:58.978523 36456 slave.cpp:1155] Forwarding total oversubscribed resources
Nov 08 09:07:59 mario mesos-slave[36438]: W1108 09:07:59.830611 36451 slave.cpp:1141] Already registered with master master@10.99.50.1:5050
Nov 08 09:07:59 mario mesos-slave[36438]: I1108 09:07:59.830658 36451 slave.cpp:1155] Forwarding total oversubscribed resources
Nov 08 09:08:14 mario mesos-slave[36438]: W1108 09:08:14.486706 36456 slave.cpp:1141] Already registered with master master@10.99.50.1:5050
Nov 08 09:08:14 mario mesos-slave[36438]: I1108 09:08:14.486758 36456 slave.cpp:1155] Forwarding total oversubscribed resources
Nov 08 09:08:15 mario mesos-slave[36438]: W1108 09:08:15.914592 36455 slave.cpp:1141] Already registered with master master@10.99.50.1:5050
Nov 08 09:08:15 mario mesos-slave[36438]: I1108 09:08:15.914660 36455 slave.cpp:1155] Forwarding total oversubscribed resources
Nov 08 09:08:17 mario mesos-slave[36438]: W1108 09:08:17.907658 36442 slave.cpp:1141] Already registered with master master@10.99.50.1:5050
Nov 08 09:08:17 mario mesos-slave[36438]: I1108 09:08:17.907707 36442 slave.cpp:1155] Forwarding total oversubscribed resources
{quote}


That's weird but there's something weirder:
{quote}
root:/var/lib/mesos/slaves# curl --verbose http://127.0.0.1:5051/health
* Hostname was NOT found in DNS cache
*   Trying 127.0.0.1...
* Connected to 127.0.0.1 (127.0.0.1) port 5051 (#0)
> GET /health HTTP/1.1
> User-Agent: curl/7.38.0
> Host: 127.0.0.1:5051
> Accept: */*
> 
< HTTP/1.1 200 OK
< Date: Tue, 08 Nov 2016 08:09:54 GMT
< Content-Length: 0
< 
* Connection #0 to host 127.0.0.1 left intact
{quote}

That's a real problem because I cannot detect that this slave is broken. It's not reported in mesos master interface so it really looks like being not joined.

Regards, Adam.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)