Open Stack

Hi Paras,

I also have speed at 0Mbps on my br-tun for everything (GRE, patch-int) in one environment, and 100Mbps on another.

Can you pls provide me with some info about those speed marks? Like, if it`s on 0 - that means that no tunnels are working properly?

Thanks
NM

-- 
Nick Maslov
Sent with Airmail

On November 11, 2013 at 6:41:40 PM, Paras pradhan (pradhanparas at gmail.com) wrote:

Yes br-tun is there.  here is the o/p of ovs-ofctl.

-
root at havana:~# ovs-ofctl show br-tun
OFPT_FEATURES_REPLY (xid=0x2): dpid:00004e6501755448
n_tables:254, n_buffers:256
capabilities: FLOW_STATS TABLE_STATS PORT_STATS QUEUE_STATS ARP_MATCH_IP
actions: OUTPUT SET_VLAN_VID SET_VLAN_PCP STRIP_VLAN SET_DL_SRC SET_DL_DST SET_NW_SRC SET_NW_DST SET_NW_TOS SET_TP_SRC SET_TP_DST ENQUEUE
 1(patch-int): addr:42:9d:7c:13:4b:58
     config:     0
     state:      0
     speed: 0 Mbps now, 0 Mbps max
 LOCAL(br-tun): addr:4e:65:01:75:54:48
     config:     0
     state:      0
     speed: 0 Mbps now, 0 Mbps max
OFPT_GET_CONFIG_REPLY (xid=0x4): frags=normal miss_send_len=0
--

Thanks
Paras.


On Sat, Nov 9, 2013 at 3:44 AM, Igor Cardoso <igordcard at gmail.com> wrote:
I guess br-tun is created in a single node as well, someone correct me if I'm wrong please.
What is the output of ovs-ofctl show br-tun ?


On 9 November 2013 02:35, Paras pradhan <pradhanparas at gmail.com> wrote:
Yes gre . all (controller plus compute) in a single node though.

On Nov 8, 2013 5:17 PM, "Igor Cardoso" <igordcard at gmail.com> wrote:
Are you using GRE?


On 8 November 2013 22:48, Paras pradhan <pradhanparas at gmail.com> wrote:
use_namespace is set to true, iproute is up to date from 12.04 LTS.

This is what I see when tcpdump
-

root at havana:/home/localadmin# tcpdump -i qvo6157a9aa-76
tcpdump: WARNING: qvo6157a9aa-76: no IPv4 address assigned
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on qvo6157a9aa-76, link-type EN10MB (Ethernet), capture size 65535 bytes
16:46:39.006180 IP6 fe80::40fa:38ff:fee3:a3fa > ff02::16: HBH ICMP6, multicast listener report v2, 1 group record(s), length 28
16:46:39.784902 IP6 :: > ff02::16: HBH ICMP6, multicast listener report v2, 1 group record(s), length 28
16:46:39.824273 IP 0.0.0.0.bootpc > 255.255.255.255.bootps: BOOTP/DHCP, Request from fa:16:3e:9a:38:1e (oui Unknown), length 280
16:46:39.933463 IP6 :: > ff02::1:ff9a:381e: ICMP6, neighbor solicitation, who has fe80::f816:3eff:fe9a:381e, length 24
16:46:40.325389 IP6 :: > ff02::16: HBH ICMP6, multicast listener report v2, 1 group record(s), length 28
16:46:40.934808 IP6 fe80::f816:3eff:fe9a:381e > ip6-allrouters: ICMP6, router solicitation, length 16
16:46:42.833239 IP 0.0.0.0.bootpc > 255.255.255.255.bootps: BOOTP/DHCP, Request from fa:16:3e:9a:38:1e (oui Unknown), length 280
16:46:44.936871 IP6 fe80::f816:3eff:fe9a:381e > ip6-allrouters: ICMP6, router solicitation, length 16
16:46:45.840668 IP 0.0.0.0.bootpc > 255.255.255.255.bootps: BOOTP/DHCP, Request from fa:16:3e:9a:38:1e (oui Unknown), length 280
16:46:48.945478 IP6 fe80::f816:3eff:fe9a:381e > ip6-allrouters: ICMP6, router solicitation, length 16
16:47:00.542029 IP 0.0.0.0 > all-systems.mcast.net: igmp query v2
16:47:00.542036 IP6 fe80::d894:29ff:fe08:5ec > ip6-allnodes: HBH ICMP6, multicast listener querymax resp delay: 10000 addr: ::, length 24
--

Thanks!
Paras.



On Fri, Nov 8, 2013 at 4:26 PM, Rami Vaknin <rvaknin at redhat.com> wrote:
On 11/09/2013 12:18 AM, Paras pradhan wrote:
dnsmaq is running, ip netns doesn't return anything.

Empty output from "ip netns" can be either not using namespaces (run grep "use_namespaces /etc/neutron/l3_agent.ini" to find out) or using wrong iproute package.
Could you please also check what does tcpdump report? could you see the discover on the dhcp nic (when using ovs it's a tapxxxxxxxx-xx nic)? could you see an offer sent from that nic?



-Paras.


On Fri, Nov 8, 2013 at 4:07 PM, Rami Vaknin <rvaknin at redhat.com> wrote:
On 11/08/2013 11:35 PM, Paras pradhan wrote:
Hi,

I gpt an instance UP but it doesnot get an IP address. neutron agent-list lists DHCP agent happy. While booting I see cirros stuck at sending discover... multiple times. dnsmasq process is running. how do i debug?

I would start with:
* ps -efl | grep dnsmasq | grep -v grep - to verify that the dnsmasq process is running, you can also check the conf file appeas in this output
* ip netns - to get the list of namespaces
* ip netns exec <the_right_dhcp_namespace_name> tcpdump -i any ... - to look for the dhcp offer or arp issues
* ping each other - instance to dhcp address, and vice versa (the second one should be done from within the namespace)


Thanks


_______________________________________________
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



 --  

Thanks,

Rami Vaknin, QE @ Red Hat, TLV, IL.




--  

Thanks,

Rami Vaknin, QE @ Red Hat, TLV, IL.


_______________________________________________
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




--
http://igordcard.blogspot.com



--
http://igordcard.blogspot.com

_______________________________________________  
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/20131114/50fc7d23/attachment.html>