Open Stack

Hi Sharma,
   Thank you very much for your reply. You really helped me a lot. After trying ping VM with larger packet, the result shows that when set -s parameter to larger than 1430 VM can not be pinged no matter throw external ips or internal ips. So I checked settings of both physical network interfaces and virtual nic through ethtool -k command, there are many differences between those nics. Firstly setting of eth0 in VM is different from setting of br-tun(all bridges have the same setting). The setting for bridge is as follows where bolded items are different from settings of eth0 in VM.　I think the problem may caused by different setting of tx-gre-segmentation but I am not sure and I do not know how to change the setting to test. Do you or anyone else has any idea about this problem? Thank you very much!!

Features for br-tun: 
rx-checksumming: off [fixed] 
tx-checksumming: on 
       tx-checksum-ipv4: off [fixed] 
       tx-checksum-ip-generic: on 
       tx-checksum-ipv6: off [fixed] 
       tx-checksum-fcoe-crc: off [fixed] 
       tx-checksum-sctp: off [fixed] 
scatter-gather: on 
       tx-scatter-gather: on 
       tx-scatter-gather-fraglist: on 
tcp-segmentation-offload: on 
       tx-tcp-segmentation: on 
       tx-tcp-ecn-segmentation: on 
       tx-tcp6-segmentation: on 
udp-fragmentation-offload: on 
generic-segmentation-offload: on 
generic-receive-offload: on 
large-receive-offload: off [fixed] 
rx-vlan-offload: off [fixed] 
tx-vlan-offload: on 
ntuple-filters: off [fixed] 
receive-hashing: off [fixed] 
highdma: on 
rx-vlan-filter: off [fixed] 
vlan-challenged: off [fixed] 
tx-lockless: on [fixed] 
netns-local: off [fixed] 
tx-gso-robust: off [fixed] 
tx-fcoe-segmentation: off [fixed] 
tx-gre-segmentation: on 
tx-ipip-segmentation: on 
tx-sit-segmentation: on 
tx-udp_tnl-segmentation: on 
fcoe-mtu: off [fixed] 
tx-nocache-copy: off 
loopback: off [fixed] 
rx-fcs: off [fixed] 
rx-all: off [fixed] 
tx-vlan-stag-hw-insert: off [fixed] 
rx-vlan-stag-hw-parse: off [fixed] 
rx-vlan-stag-filter: off [fixed] 
l2-fwd-offload: off [fixed] 
busy-poll: off [fixed]


Regards
hjh

2015-09-28



applyhhj



发件人：Rahul Sharma <rahulsharmaait at gmail.com>
发送时间：2015-09-28 02:41
主题：Re: [Openstack] Data transmission failure between VM and outsidemachines
收件人："Mike Spreitzer"<mspreitz at us.ibm.com>
抄送："applyhhj"<applyhhj at 163.com>,"openstack"<openstack at lists.openstack.org>

Hi hjh,


If you are able to ping to that particular instance (which I hope you would be since the initial tcp handshake gets completed), the other thing which might cause such an issue if the MTU size. We have also seen such issue when the physical host's nic was unable to deal correctly with the fragmented packets. You can test this by first sending ping packets, then increasing their size to 1500 and see if its successful. In our case, we had to disable Large Receive Offload(LRO) on the nic card and then it worked fine. Other could be increasing the MTU size on the nic itself. Do give it a try and see if it helps.


Thanks.


Rahul Sharma
MS in Computer Science, 2016
College of Computer and Information Science, Northeastern University
Mobile:  801-706-7860
Email: rahulsharmaait at gmail.com
Linkedin: www.linkedin.com/in/rahulsharmaait



On Sun, Sep 27, 2015 at 12:29 PM, Mike Spreitzer <mspreitz at us.ibm.com> wrote:

> From: "applyhhj" <applyhhj at 163.com>
> To: "openstack" <openstack at lists.openstack.org>
> Date: 09/27/2015 11:16 AM
> Subject: [Openstack] Data transmission failure between VM and outside machines
> 
> Hi, I have setup the openstack cloud and launched VMs in the cloud. 
> At first everything went very well. But yesterday evening, due to 
> some reasons our lab lost power and all servers were shutdown. This 
> morning I turned on all nodes and try to connect to the VM by ssh 
> but failed. I used netstat to check the status of port 22. It shows 
> that connection between VM and machine in the external network can 
> be established. However the ssh process just stuck at 
> SSH2_MSG_KEXINIT sent. Also I setup a rabbitmq server in the VM and 
> the same situation happened. When connecting rabbitmq server through
> web ui, netstat shows that connections have been established however
> the web browser can not get any data from the server, it shows a 
> blank page. Also I tried to ssh from one VM to another VM through 
> internal network, the 192.168.1.0/24 network, similar things 
> happened. Does anyone know how to fix this problem? Thank you! By 
> the way the br-ex bridge can only be brought up manually, so after 
> boot the network node I brought up the br-ex bridge and restart all 
> relevant network services in network node. Please help me with this 
> problem. Thank you very much!!

When faced with networking mysteries like that my next step is usually to start taking packet traces at various points.

Regards,
Mike

_______________________________________________
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to     : openstack at lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack/attachments/20150928/12efc21b/attachment.html>