Network is slow on Compute Node of Two Node IceHouse Neutron OVS&VLAN CentOS 6.5 (Fedora 20) Cluster

asked 2014-05-20 07:26:01 -0600

dbaxps gravatar image

updated 2014-05-22 08:18:37 -0600

        Both Nodes are KVMs with 2 VCPUS and 4 GB RAM running at F20 KVM Server
        (on non-default Libvirt subnet
        When I deploy two Cirros instances :-
        First instance on Controller ( which also runs nova-compute)
        Second instance on Compute.
        Ping,traceroute,nslookup ( ICMP based utilities) work with Internet with the same speed on both nodes
        When I try `curl` :-

        1. It takes seconds on instance deployed on Controller
        2. It takes 2-3 min on instance deployed on Compute 

     For Cirros VM1  running on Controller node  there is corresponding qvb-xxxxxx & qvo-xxxxx  veth pair
     For Cirros VM2  running on Compute node  there is corresponding  qvb-yyyyy & qvo-yyyyy veth pair
     So, run  on Controller `tcpdump -vv -i qvo-xxxxxx | tee log.controller`
     and on Compute `tcpdump -vv -i qvo-yyyyy | tee log.compute`
Same setup has been done for Two Node IceHouse Neutron OVS&VLAN Fedora 20 Cluster with same issue and same workaround
     Attached report is,actually, log.controller versus log.compute

TCPDUMP Contr-VS-Compute
Answer file not yet updated for non-default Libvirt Subnet

edit retag flag offensive close merge delete

1 answer

Sort by ยป oldest newest most voted

answered 2014-05-21 00:08:53 -0600

dbaxps gravatar image
Running on Compute Node :-
# /sbin/ethtool --offload eth1  tx off
Fixes a problem for me. I was able launch Ubuntu 14.04 VM on Compute, then run within VM
# apt-get update
# apt-get -y install links
and surf the Net with links
edit flag offensive delete link more

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower


Asked: 2014-05-20 07:26:01 -0600

Seen: 3,276 times

Last updated: May 22 '14