[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: skydns i/o timeout



This sounds similar to what i encountered at a customer site. We applied the solution specified in these knowledge base articles and it seemed to resolve the issue. The 3690091 article has most of what you need but the other article specifies a few more ip table rules.
https://access.redhat.com/solutions/3690091
https://access.redhat.com/solutions/972673




On Wed, Jan 30, 2019 at 6:25 PM Cameron Braid <cameron braid com au> wrote:
Hi,

I am having some DNS issues in my origin 3.7 cluster

I am seeing occasional dns lookup delays of around 1 to 5 seconds every 5 - 10 minutes

In trying to find the cause I found in the origin-node logs the following about every second

Jan 31 13:10:18 node01-2018.drivenow.com.au origin-node[39574]: I0131 13:10:18.042221   39574 logs.go:41] skydns: failure to forward request "read udp 10.118.56.32:53613->10.118.56.32:53: i/o timeout"


However I can successfully do dns resolution on the node as per 

dig google.com @10.118.56.32 +short +search

216.58.200.110


And resolve internal IPs too

dig docker-registry.default.svc @10.118.56.32 +short +search

172.30.99.40


Am I correct in my testing that origin-node is accessing the dnsmasq server on 10.118.56.32 ?

I get the same issue on all three nodes.

I can successfully run "oc adm diagnostics NetworkCheck" without any errors

Any thoughts on where else to look ?

Cameron

_______________________________________________
users mailing list
users lists openshift redhat com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users


--

Jose Perez Espinoza

Consultant

Red Hat

@RedHat   Red Hat   Red Hat

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]