[ovs-discuss] GRE tunnel to multiple host

Reid Price rprice at nicira.com
Wed Jun 5 21:01:31 UTC 2013


Please don't drop the list.

If star works and mesh does not then it seems very likely that you are just
seeing side effects of looping your network in mesh.  You will have to make
some modifications to your setup if you want a mesh to work, NORMAL action
will send broadcast/unknown packets out every port except the one it came
in on, and you will have [N-2]x the number of packets every hop, until
things crash and burn.

You might look at STP as a solution to this.  You could also add flows to
only send certain traffic.  For example, you could create very dumb
hub-like behavior with 3 hosts like this:

on a given host
  port 0 - local
  port 1 - gre to other
  port 2 - gre to other

 ovs-ofctl add-flow in_port=0,actions=1,2
 ovs-ofctl add-flow in_port=1,actions=0
 ovs-ofctl add-flow in_port=2,actions=0

With a NORMAL flow instead of the above flows, you would see broadcast
traffic from 1 go to 0 and 2, then the host on 2 would send it to someone
else, etc, so your broadcasts would slowly build up.  With 4+ hosts, the
number of broadcasts would be 2x,3x,4x,... after each hop, as in your
current setup, and quickly drown out everything else.

  -Reid


On Wed, Jun 5, 2013 at 3:46 AM, Ahmed Bessifi <ahmed.bessifi at gmail.com>wrote:

> Hi, no answers yet !
> I summarize the problem again. I want to create a full-mesh topology
> whith GRE. Each physical node has gre tunnels to other nodes. When I tested
> this whith 5 nodes my VMs don't respond to ping. However in star topology
> it works fine.
> My question now, does OVS support mesh-topology based on multiple gre
> tunnels or GRE it self couldn't work in mesh topology ?
>
> Best regards,
> Ahmed
>
>
> 2013/6/4 Ahmed Bessifi <ahmed.bessifi at gmail.com>
>
>> Hi,
>>
>> Yes, the same think occurs in other hosts.
>> So you mean that when I created gre tunnels from each host to others this
>> creates loops that bring down
>>  the communication ?
>>
>>
>>
>>
>> 2013/6/4 Reid Price <rprice at nicira.com>
>>
>>> Hi Ahmed,
>>>
>>> Did you do the same thing on several other hosts and leave the default
>>> action for all of them?  The warning messages between the highlighted parts
>>> that you've pasted can occur when you have looped your network.
>>>
>>>   -Reid
>>>
>>> On Jun 4, 2556 BE, at 2:10, Ahmed Bessifi <ahmed.bessifi at gmail.com>
>>> wrote:
>>>
>>> In this page<http://git.openvswitch.org/cgi-bin/gitweb.cgi?p=openvswitch;a=blob_plain;f=WHY-OVS;hb=HEAD>,
>>> it is mentioned that open vSwitch supports thousands of GRE tunnels :
>>>
>>> ===========
>>>
>>> In a similar vein, Open vSwitch supports a GRE implementation that can
>>>   handle thousands of simultaneous GRE tunnels and supports remote
>>>   configuration for tunnel creation, configuration, and tear-down.
>>>   This, for example, can be used to connect private VM networks in
>>>   different data centers.
>>>
>>> ===========
>>>
>>> When I test it in a real case (creating 9 gre tunnels) it seems that all
>>> gre interfaces are created but nothing works ! This is capture of
>>> ovs-vswitchd.log :
>>>
>>> ==> ovs-vswitchd.log <==
>>> 2013-06-04T07:49:20Z|00138|timeval|WARN|context switches: 0 voluntary,
>>> 47 involuntary
>>> 2013-06-04T07:49:20Z|00139|poll_loop|WARN|wakeup due to [POLLIN] on fd
>>> 16 (<->/var/run/openvswitch/db.sock) at ../lib/stream-fd.c:142 (83% CPU
>>> usage)
>>> 2013-06-04T07:49:20Z|00140|poll_loop|WARN|wakeup due to [POLLIN] on fd
>>> 19 (unknown anon_inode:[eventpoll]) at ../lib/dpif-linux.c:1362 (83% CPU
>>> usage)
>>> 2013-06-04T07:49:22Z|00141|timeval|WARN|Unreasonably long 2480ms poll
>>> interval (0ms user, 2049ms system)
>>> 2013-06-04T07:49:22Z|00142|timeval|WARN|context switches: 0 voluntary,
>>> 44 involuntary
>>> 2013-06-04T07:49:22Z|00143|poll_loop|WARN|wakeup due to [POLLIN] on fd
>>> 19 (unknown anon_inode:[eventpoll]) at ../lib/dpif-linux.c:1362 (83% CPU
>>> usage)
>>> 2013-06-04T07:49:24Z|00144|timeval|WARN|Unreasonably long 1782ms poll
>>> interval (0ms user, 1472ms system)
>>> 2013-06-04T07:49:24Z|00145|timeval|WARN|context switches: 0 voluntary,
>>> 36 involuntary
>>> 2013-06-04T07:49:24Z|00146|poll_loop|WARN|wakeup due to [POLLIN] on fd
>>> 16 (<->/var/run/openvswitch/db.sock) at ../lib/stream-fd.c:142 (83% CPU
>>> usage)
>>> 2013-06-04T07:51:25Z|00147|dpif_linux|WARN|system at ovs-system: lost
>>> packet on channel 2 (last polled 301 ms ago)
>>> 2013-06-04T08:51:23Z|00148|bridge|INFO|bridge br0: added interface br0
>>> on port 65534
>>> 2013-06-04T08:51:23Z|00149|bridge|INFO|bridge br0: using datapath ID
>>> 00009efa2748ca4e
>>> 2013-06-04T08:51:23Z|00150|connmgr|INFO|br0: added service controller
>>> "punix:/var/run/openvswitch/br0.mgmt"
>>> 2013-06-04T08:51:23Z|00151|bridge|INFO|bridge br0: added interface gre0
>>> on port 1
>>> 2013-06-04T08:51:23Z|00152|tunnel|WARN|receive tunnel port not found
>>> (172.16.65.25->172.16.65.19, key=0, dp port=2, skb mark=0)
>>> 2013-06-04T08:51:23Z|00153|ofproto_dpif|INFO|received packet on
>>> unassociated port 65535
>>> 2013-06-04T08:51:23Z|00154|tunnel|WARN|receive tunnel port not found
>>> (172.16.65.25->172.16.65.19, key=0, dp port=2, skb mark=0)
>>> 2013-06-04T08:51:23Z|00155|ofproto_dpif|INFO|received packet on
>>> unassociated port 65535
>>> 2013-06-04T08:51:23Z|00156|tunnel|WARN|receive tunnel port not found
>>> (172.16.65.25->172.16.65.19, key=0, dp port=2, skb mark=0)
>>> 2013-06-04T08:51:23Z|00157|ofproto_dpif|INFO|received packet on
>>> unassociated port 65535
>>> 2013-06-04T08:51:23Z|00158|tunnel|WARN|receive tunnel port not found
>>> (172.16.65.25->172.16.65.19, key=0, dp port=2, skb mark=0)
>>> 2013-06-04T08:51:23Z|00159|ofproto_dpif|INFO|received packet on
>>> unassociated port 65535
>>> 2013-06-04T08:51:23Z|00160|tunnel|WARN|receive tunnel port not found
>>> (172.16.65.25->172.16.65.19, key=0, dp port=2, skb mark=0)
>>> 2013-06-04T08:51:23Z|00161|ofproto_dpif|INFO|received packet on
>>> unassociated port 65535
>>> 2013-06-04T08:51:23Z|00162|bridge|INFO|bridge br0: added interface gre1
>>> on port 2
>>> 2013-06-04T08:51:23Z|00163|bridge|INFO|bridge br0: added interface gre2
>>> on port 3
>>> 2013-06-04T08:51:23Z|00164|bridge|INFO|bridge br0: added interface gre3
>>> on port 4
>>> 2013-06-04T08:51:23Z|00165|bridge|INFO|bridge br0: added interface gre4
>>> on port 5
>>> 2013-06-04T08:51:28Z|00166|bridge|INFO|bridge br0: added interface gre5
>>> on port 6
>>> 2013-06-04T08:51:29Z|00167|dpif_linux|WARN|system at ovs-system: lost
>>> packet on channel 2 (last polled 1116 ms ago)
>>> 2013-06-04T08:51:30Z|00168|timeval|WARN|Unreasonably long 2172ms poll
>>> interval (0ms user, 1700ms system)
>>> 2013-06-04T08:51:30Z|00169|timeval|WARN|context switches: 0 voluntary,
>>> 46 involuntary
>>> 2013-06-04T08:51:30Z|00170|coverage|INFO|Dropped 6 log messages in last
>>> 3739 seconds (most recently, 3727 seconds ago) due to excessive rate
>>> 2013-06-04T08:51:30Z|00171|coverage|INFO|Event coverage (epoch
>>> 3141/entire run), hash=3b4de517:
>>> 2013-06-04T08:51:30Z|00172|coverage|INFO|ofproto_dpif_xlate           7
>>> /      6718
>>> 2013-06-04T08:51:30Z|00173|coverage|INFO|flow_extract               101
>>> /      5451
>>> 2013-06-04T08:51:30Z|00174|coverage|INFO|hmap_expand                 35
>>> /     55180
>>> 2013-06-04T08:51:30Z|00175|coverage|INFO|mac_learning_learned         5
>>> /      3736
>>> 2013-06-04T08:51:30Z|00176|coverage|INFO|netdev_get_stats             7
>>> /      1716
>>> 2013-06-04T08:51:30Z|00177|coverage|INFO|poll_zero_timeout            1
>>> /       891
>>> 2013-06-04T08:51:30Z|00178|coverage|INFO|poll_fd_wait                24
>>> /     75088
>>> 2013-06-04T08:51:30Z|00179|coverage|INFO|util_xalloc               1192
>>> /   1344630
>>> 2013-06-04T08:51:30Z|00180|coverage|INFO|netlink_sent               108
>>> /      6860
>>> 2013-06-04T08:51:30Z|00181|coverage|INFO|netlink_received           106
>>> /      7655
>>> 2013-06-04T08:51:30Z|00182|coverage|INFO|netlink_overflow             1
>>> /         4
>>> 2013-06-04T08:51:30Z|00183|coverage|INFO|bridge_reconfigure           0
>>> /        19
>>> 2013-06-04T08:51:30Z|00184|coverage|INFO|ofproto_update_port          0
>>> /        70
>>> 2013-06-04T08:51:30Z|00185|coverage|INFO|ofproto_flush                0
>>> /         2
>>> 2013-06-04T08:51:30Z|00186|coverage|INFO|rev_flow_table               0
>>> /         2
>>> 2013-06-04T08:51:30Z|00187|coverage|INFO|rev_reconfigure              0
>>> /        17
>>> 2013-06-04T08:51:30Z|00188|coverage|INFO|facet_revalidate             0
>>> /        30
>>> 2013-06-04T08:51:30Z|00189|coverage|INFO|dpif_flow_del                0
>>> /         7
>>> 2013-06-04T08:51:30Z|00190|coverage|INFO|dpif_flow_put                0
>>> /        35
>>> 2013-06-04T08:51:30Z|00191|coverage|INFO|dpif_flow_flush              0
>>> /         2
>>> 2013-06-04T08:51:30Z|00192|coverage|INFO|dpif_port_del                0
>>> /         2
>>> 2013-06-04T08:51:30Z|00193|coverage|INFO|dpif_port_add                0
>>> /         4
>>> 2013-06-04T08:51:30Z|00194|coverage|INFO|hmap_pathological            0
>>> /      1757
>>> 2013-06-04T08:51:30Z|00195|coverage|INFO|mac_learning_expired         0
>>> /         5
>>> 2013-06-04T08:51:30Z|00196|coverage|INFO|stream_open                  0
>>> /         1
>>> 2013-06-04T08:51:30Z|00197|coverage|INFO|pstream_open                 0
>>> /         5
>>> 2013-06-04T08:51:30Z|00198|coverage|INFO|netdev_set_hwaddr            0
>>> /         2
>>> 2013-06-04T08:51:30Z|00199|coverage|INFO|netdev_get_hwaddr            0
>>> /         2
>>> 2013-06-04T08:51:30Z|00200|coverage|INFO|netdev_get_ifindex           0
>>> /         2
>>> 2013-06-04T08:51:30Z|00201|coverage|INFO|netdev_set_policing          0
>>> /         4
>>> 2013-06-04T08:51:30Z|00202|coverage|INFO|nln_changed                  0
>>> /        13
>>> 2013-06-04T08:51:30Z|00203|coverage|INFO|55 events never hit
>>> 2013-06-04T08:51:32Z|00204|timeval|WARN|Unreasonably long 1249ms poll
>>> interval (0ms user, 1032ms system)
>>> 2013-06-04T08:51:32Z|00205|timeval|WARN|context switches: 0 voluntary,
>>> 33 involuntary
>>> 2013-06-04T08:51:32Z|00206|coverage|INFO|Event coverage (epoch
>>> 3142/entire run), hash=46038245:
>>> 2013-06-04T08:51:32Z|00207|coverage|INFO|ofproto_dpif_xlate          13
>>> /      6731
>>> 2013-06-04T08:51:32Z|00208|coverage|INFO|flow_extract               101
>>> /      5552
>>> 2013-06-04T08:51:32Z|00209|coverage|INFO|hmap_expand                 51
>>> /     55231
>>> 2013-06-04T08:51:32Z|00210|coverage|INFO|hmap_pathological            7
>>> /      1764
>>> 2013-06-04T08:51:32Z|00211|coverage|INFO|mac_learning_learned         4
>>> /      3740
>>> 2013-06-04T08:51:32Z|00212|coverage|INFO|poll_zero_timeout            1
>>> /       892
>>> 2013-06-04T08:51:32Z|00213|coverage|INFO|poll_fd_wait                24
>>> /     75112
>>> 2013-06-04T08:51:32Z|00214|coverage|INFO|util_xalloc               3402
>>> /   1348126
>>> 2013-06-04T08:51:32Z|00215|coverage|INFO|netlink_sent               106
>>> /      6966
>>> 2013-06-04T08:51:32Z|00216|coverage|INFO|netlink_received           103
>>> /      7758
>>> 2013-06-04T08:51:32Z|00217|coverage|INFO|bridge_reconfigure           0
>>> /        19
>>> 2013-06-04T08:51:32Z|00218|coverage|INFO|ofproto_update_port          0
>>> /        70
>>> 2013-06-04T08:51:32Z|00219|coverage|INFO|ofproto_flush                0
>>> /         2
>>> 2013-06-04T08:51:32Z|00220|coverage|INFO|rev_flow_table               0
>>> /         2
>>> 2013-06-04T08:51:32Z|00221|coverage|INFO|rev_reconfigure              0
>>> /        17
>>> 2013-06-04T08:51:32Z|00222|coverage|INFO|facet_revalidate             0
>>> /        30
>>> 2013-06-04T08:51:32Z|00223|coverage|INFO|dpif_flow_del                0
>>> /         7
>>> 2013-06-04T08:51:32Z|00224|coverage|INFO|dpif_flow_put                0
>>> /        35
>>> 2013-06-04T08:51:32Z|00225|coverage|INFO|dpif_flow_flush              0
>>> /         2
>>> 2013-06-04T08:51:32Z|00226|coverage|INFO|dpif_port_del                0
>>> /         2
>>> 2013-06-04T08:51:32Z|00227|coverage|INFO|dpif_port_add                0
>>> /         4
>>> 2013-06-04T08:51:32Z|00228|coverage|INFO|mac_learning_expired         0
>>> /         5
>>> 2013-06-04T08:51:32Z|00229|coverage|INFO|netdev_get_stats             0
>>> /      1716
>>> 2013-06-04T08:51:32Z|00230|coverage|INFO|stream_open                  0
>>> /         1
>>> 2013-06-04T08:51:32Z|00231|coverage|INFO|pstream_open                 0
>>> /         5
>>> 2013-06-04T08:51:32Z|00232|coverage|INFO|netdev_set_hwaddr            0
>>> /         2
>>> 2013-06-04T08:51:32Z|00233|coverage|INFO|netdev_get_hwaddr            0
>>> /         2
>>> 2013-06-04T08:51:32Z|00234|coverage|INFO|netdev_get_ifindex           0
>>> /         2
>>> 2013-06-04T08:51:32Z|00235|coverage|INFO|netdev_set_policing          0
>>> /         4
>>> 2013-06-04T08:51:32Z|00236|coverage|INFO|nln_changed                  0
>>> /        13
>>> 2013-06-04T08:51:32Z|00237|coverage|INFO|netlink_overflow             0
>>> /         4
>>> 2013-06-04T08:51:32Z|00238|coverage|INFO|55 events never hit
>>> 2013-06-04T08:52:03Z|00239|bridge|INFO|bridge br0: added interface gre6
>>> on port 7
>>> 2013-06-04T08:52:03Z|00240|bridge|INFO|bridge br0: added interface gre7
>>> on port 8
>>> 2013-06-04T08:53:20Z|00241|dpif_linux|WARN|system at ovs-system: lost
>>> packet on channel 2 (last polled 204 ms ago)
>>> 2013-06-04T08:53:29Z|00242|dpif_linux|WARN|system at ovs-system: lost
>>> packet on channel 2 (last polled 244 ms ago)
>>> 2013-06-04T08:54:45Z|00243|bridge|INFO|bridge br0: added interface gre8
>>> on port 9
>>>
>>> 2013-06-04T08:57:06Z|00244|dpif_linux|WARN|system at ovs-system: lost
>>> packet on channel 2 (last polled 86 ms ago)
>>> 2013-06-04T08:57:14Z|00245|dpif_linux|WARN|system at ovs-system: lost
>>> packet on channel 2 (last polled 914 ms ago)
>>> 2013-06-04T08:59:31Z|00246|dpif_linux|WARN|system at ovs-system: lost
>>> packet on channel 2 (last polled 1257 ms ago)
>>> 2013-06-04T08:59:33Z|00247|timeval|WARN|Unreasonably long 2546ms poll
>>> interval (0ms user, 2108ms system)
>>> 2013-06-04T08:59:33Z|00248|timeval|WARN|context switches: 0 voluntary,
>>> 51 involuntary
>>> 2013-06-04T08:59:33Z|00249|coverage|INFO|Event coverage (epoch
>>> 4930/entire run), hash=2979872e:
>>> 2013-06-04T08:59:33Z|00250|coverage|INFO|ofproto_dpif_xlate          13
>>> /     10217
>>> 2013-06-04T08:59:33Z|00251|coverage|INFO|flow_extract               101
>>> /     14425
>>> 2013-06-04T08:59:33Z|00252|coverage|INFO|hmap_expand                 15
>>> /     85655
>>> 2013-06-04T08:59:33Z|00253|coverage|INFO|mac_learning_learned         7
>>> /      5656
>>> 2013-06-04T08:59:33Z|00254|coverage|INFO|poll_zero_timeout            1
>>> /      1381
>>> 2013-06-04T08:59:33Z|00255|coverage|INFO|poll_fd_wait                24
>>> /    118009
>>> 2013-06-04T08:59:33Z|00256|coverage|INFO|util_xalloc                121
>>> /   2058068
>>> 2013-06-04T08:59:33Z|00257|coverage|INFO|netlink_sent               107
>>> /     16690
>>> 2013-06-04T08:59:33Z|00258|coverage|INFO|netlink_received           103
>>> /     17738
>>> 2013-06-04T08:59:33Z|00259|coverage|INFO|netlink_overflow             1
>>> /         9
>>> 2013-06-04T08:59:33Z|00260|coverage|INFO|bridge_reconfigure           0
>>> /        22
>>> 2013-06-04T08:59:33Z|00261|coverage|INFO|ofproto_update_port          0
>>> /        94
>>> 2013-06-04T08:59:33Z|00262|coverage|INFO|ofproto_flush                0
>>> /         2
>>> 2013-06-04T08:59:33Z|00263|coverage|INFO|rev_flow_table               0
>>> /         2
>>> 2013-06-04T08:59:33Z|00264|coverage|INFO|rev_reconfigure              0
>>> /        20
>>> 2013-06-04T08:59:33Z|00265|coverage|INFO|facet_revalidate             0
>>> /        42
>>> 2013-06-04T08:59:33Z|00266|coverage|INFO|dpif_flow_del                0
>>> /         7
>>> 2013-06-04T08:59:33Z|00267|coverage|INFO|dpif_flow_put                0
>>> /        47
>>> 2013-06-04T08:59:33Z|00268|coverage|INFO|dpif_flow_flush              0
>>> /         2
>>> 2013-06-04T08:59:33Z|00269|coverage|INFO|dpif_port_del                0
>>> /         2
>>> 2013-06-04T08:59:33Z|00270|coverage|INFO|dpif_port_add                0
>>> /         4
>>> 2013-06-04T08:59:33Z|00271|coverage|INFO|hmap_pathological            0
>>> /      2682
>>> 2013-06-04T08:59:33Z|00272|coverage|INFO|mac_learning_expired         0
>>> /         6
>>> 2013-06-04T08:59:33Z|00273|coverage|INFO|netdev_get_stats             0
>>> /      2628
>>> 2013-06-04T08:59:33Z|00274|coverage|INFO|stream_open                  0
>>> /         1
>>> 2013-06-04T08:59:33Z|00275|coverage|INFO|pstream_open                 0
>>> /         5
>>> 2013-06-04T08:59:33Z|00276|coverage|INFO|netdev_set_hwaddr            0
>>> /         2
>>> 2013-06-04T08:59:33Z|00277|coverage|INFO|netdev_get_hwaddr            0
>>> /         2
>>> 2013-06-04T08:59:33Z|00278|coverage|INFO|netdev_get_ifindex           0
>>> /         2
>>> 2013-06-04T08:59:33Z|00279|coverage|INFO|netdev_set_policing          0
>>> /         4
>>> 2013-06-04T08:59:33Z|00280|coverage|INFO|nln_changed                  0
>>> /        13
>>> 2013-06-04T08:59:33Z|00281|coverage|INFO|55 events never hit
>>> 2013-06-04T08:59:33Z|00282|poll_loop|WARN|Dropped 19 log messages in
>>> last 4209 seconds (most recently, 4206 seconds ago) due to excessive rate
>>> 2013-06-04T08:59:33Z|00283|poll_loop|WARN|Dropped 12 log messages in
>>> last 4229 seconds (most recently, 4209 seconds ago) due to excessive rate
>>> 2013-06-04T08:59:33Z|00284|poll_loop|WARN|wakeup due to [POLLIN] on fd
>>> 19 (unknown anon_inode:[eventpoll]) at ../lib/dpif-linux.c:1362 (55% CPU
>>> usage)
>>> 2013-06-04T08:59:34Z|00285|timeval|WARN|Unreasonably long 1069ms poll
>>> interval (0ms user, 888ms system)
>>> 2013-06-04T08:59:34Z|00286|timeval|WARN|context switches: 0 voluntary,
>>> 25 involuntary
>>> 2013-06-04T08:59:34Z|00287|coverage|INFO|Event coverage (epoch
>>> 4931/entire run), hash=07bde3a8:
>>> 2013-06-04T08:59:34Z|00288|coverage|INFO|ofproto_dpif_xlate          11
>>> /     10228
>>> 2013-06-04T08:59:34Z|00289|coverage|INFO|flow_extract               101
>>> /     14526
>>> 2013-06-04T08:59:34Z|00290|coverage|INFO|hmap_expand                 15
>>> /     85670
>>> 2013-06-04T08:59:34Z|00291|coverage|INFO|mac_learning_learned         5
>>> /      5661
>>> 2013-06-04T08:59:34Z|00292|coverage|INFO|poll_zero_timeout            1
>>> /      1382
>>> 2013-06-04T08:59:34Z|00293|coverage|INFO|poll_fd_wait                24
>>> /    118033
>>> 2013-06-04T08:59:34Z|00294|coverage|INFO|util_xalloc                125
>>> /   2058288
>>> 2013-06-04T08:59:34Z|00295|coverage|INFO|netlink_sent               107
>>> /     16797
>>> 2013-06-04T08:59:34Z|00296|coverage|INFO|netlink_received           103
>>> /     17841
>>> 2013-06-04T08:59:34Z|00297|coverage|INFO|bridge_reconfigure           0
>>> /        22
>>> 2013-06-04T08:59:34Z|00298|coverage|INFO|ofproto_update_port          0
>>> /        94
>>> 2013-06-04T08:59:34Z|00299|coverage|INFO|ofproto_flush                0
>>> /         2
>>> 2013-06-04T08:59:34Z|00300|coverage|INFO|rev_flow_table               0
>>> /         2
>>> 2013-06-04T08:59:34Z|00301|coverage|INFO|rev_reconfigure              0
>>> /        20
>>> 2013-06-04T08:59:34Z|00302|coverage|INFO|facet_revalidate             0
>>> /        42
>>> 2013-06-04T08:59:34Z|00303|coverage|INFO|dpif_flow_del                0
>>> /         7
>>> 2013-06-04T08:59:34Z|00304|coverage|INFO|dpif_flow_put                0
>>> /        47
>>> 2013-06-04T08:59:34Z|00305|coverage|INFO|dpif_flow_flush              0
>>> /         2
>>> 2013-06-04T08:59:34Z|00306|coverage|INFO|dpif_port_del                0
>>> /         2
>>> 2013-06-04T08:59:34Z|00307|coverage|INFO|dpif_port_add                0
>>> /         4
>>> 2013-06-04T08:59:34Z|00308|coverage|INFO|hmap_pathological            0
>>> /      2682
>>> 2013-06-04T08:59:34Z|00309|coverage|INFO|mac_learning_expired         0
>>> /         6
>>> 2013-06-04T08:59:34Z|00310|coverage|INFO|netdev_get_stats             0
>>> /      2628
>>> 2013-06-04T08:59:34Z|00311|coverage|INFO|stream_open                  0
>>> /         1
>>> 2013-06-04T08:59:34Z|00312|coverage|INFO|pstream_open                 0
>>> /         5
>>> 2013-06-04T08:59:34Z|00313|coverage|INFO|netdev_set_hwaddr            0
>>> /         2
>>> 2013-06-04T08:59:34Z|00314|coverage|INFO|netdev_get_hwaddr            0
>>> /         2
>>> 2013-06-04T08:59:34Z|00315|coverage|INFO|netdev_get_ifindex           0
>>> /         2
>>> 2013-06-04T08:59:34Z|00316|coverage|INFO|netdev_set_policing          0
>>> /         4
>>> 2013-06-04T08:59:34Z|00317|coverage|INFO|nln_changed                  0
>>> /        13
>>> 2013-06-04T08:59:34Z|00318|coverage|INFO|netlink_overflow             0
>>> /         9
>>> 2013-06-04T08:59:34Z|00319|coverage|INFO|55 events never hit
>>> 2013-06-04T08:59:34Z|00320|poll_loop|WARN|wakeup due to [POLLIN] on fd
>>> 19 (unknown anon_inode:[eventpoll]) at ../lib/dpif-linux.c:1362 (55% CPU
>>> usage)
>>> 2013-06-04T08:59:34Z|00321|poll_loop|WARN|wakeup due to [POLLIN] on fd
>>> 19 (unknown anon_inode:[eventpoll]) at ../lib/dpif-linux.c:1362 (55% CPU
>>> usage)
>>> 2013-06-04T08:59:34Z|00322|poll_loop|WARN|wakeup due to [POLLIN] on fd
>>> 19 (unknown anon_inode:[eventpoll]) at ../lib/dpif-linux.c:1362 (55% CPU
>>> usage)
>>> 2013-06-04T08:59:34Z|00323|poll_loop|WARN|wakeup due to [POLLIN] on fd
>>> 19 (unknown anon_inode:[eventpoll]) at ../lib/dpif-linux.c:1362 (55% CPU
>>> usage)
>>> 2013-06-04T08:59:34Z|00324|poll_loop|WARN|wakeup due to [POLLIN] on fd
>>> 19 (unknown anon_inode:[eventpoll]) at ../lib/dpif-linux.c:1362 (55% CPU
>>> usage)
>>> 2013-06-04T08:59:34Z|00325|poll_loop|WARN|wakeup due to [POLLIN] on fd
>>> 19 (unknown anon_inode:[eventpoll]) at ../lib/dpif-linux.c:1362 (55% CPU
>>> usage)
>>> 2013-06-04T08:59:34Z|00326|poll_loop|WARN|wakeup due to [POLLIN] on fd
>>> 19 (unknown anon_inode:[eventpoll]) at ../lib/dpif-linux.c:1362 (55% CPU
>>> usage)
>>> 2013-06-04T08:59:34Z|00327|poll_loop|WARN|wakeup due to [POLLIN] on fd
>>> 19 (unknown anon_inode:[eventpoll]) at ../lib/dpif-linux.c:1362 (55% CPU
>>> usage)
>>> 2013-06-04T08:59:34Z|00328|poll_loop|WARN|wakeup due to [POLLIN] on fd
>>> 19 (unknown anon_inode:[eventpoll]) at ../lib/dpif-linux.c:1362 (55% CPU
>>> usage)
>>>
>>>
>>> 2013/6/4 Ahmed Bessifi <ahmed.bessifi at gmail.com>
>>>
>>>> Hi,
>>>>
>>>> I am working in a cluster and I used OVS (1.10.0) to interconnect 10
>>>> physical nodes. In each node a did the folowing steps :
>>>>   * I create an ovs bridge.
>>>>   * I configure 9 gre tunnel for the rest of node.
>>>>   * I create my 10 containers and attatch them to the ovs bridge.
>>>> The problem, is I cannot ping VMs and even I perceive that hosts become
>>>> slow and respond hardly to ping. However, when I test his with just 2
>>>> physical node it works well.
>>>> I think the problem is the number of gre interfaces that causes this
>>>> troubleshoot.
>>>>
>>>> Is there any way we can create a topology in a better way to
>>>> interconnect bridges ?
>>>>
>>>> Regards,
>>>> Ahmed
>>>>
>>>> --
>>>> *"**We build software for people that build software.."
>>>> --
>>>> Ahmed Bessifi*
>>>> --
>>>> Étudiant ingénieur en informatique à la Faculté des Sciences de Tunis,
>>>> Technicien Supérieur en Télécommunication diplômé de l'ISETCom,
>>>> Membre au Club des Logiciels Libres à la FST.
>>>>
>>>
>>>
>>>
>>> --
>>> *"**We build software for people that build software.."
>>> --
>>> Ahmed Bessifi*
>>> --
>>> Étudiant ingénieur en informatique à la Faculté des Sciences de Tunis,
>>> Technicien Supérieur en Télécommunication diplômé de l'ISETCom,
>>> Membre au Club des Logiciels Libres à la FST.
>>>
>>> _______________________________________________
>>> discuss mailing list
>>> discuss at openvswitch.org
>>> http://openvswitch.org/mailman/listinfo/discuss
>>>
>>>
>>
>>
>> --
>> *"**We build software for people that build software.."
>> --
>> Ahmed Bessifi*
>> --
>> Étudiant ingénieur en informatique à la Faculté des Sciences de Tunis,
>> Technicien Supérieur en Télécommunication diplômé de l'ISETCom,
>> Membre au Club des Logiciels Libres à la FST.
>>
>
>
>
> --
> *"**We build software for people that build software.."
> --
> Ahmed Bessifi*
> --
> Étudiant ingénieur en informatique à la Faculté des Sciences de Tunis,
> Technicien Supérieur en Télécommunication diplômé de l'ISETCom,
> Membre au Club des Logiciels Libres à la FST.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://openvswitch.org/pipermail/ovs-discuss/attachments/20130605/4a2bf9d1/attachment.html>


More information about the discuss mailing list