[ovs-discuss] Problems with MPLS header pushing

thomas.morin at orange.com thomas.morin at orange.com
Fri Jul 24 17:32:34 UTC 2015


Hi Pravin,

Pravin Shelar :
> On Tue, Jul 21, 2015 at 8:07 AM, Daniel Herrmann
> <daniel.herrmann at igd.fraunhofer.de> wrote:
>> I know this may not be a very interesting question, but I do not have
>> any idea why forwarding does not work.
>>
>> Any help would really be appreciated.
> There were few MPLS fixed pushed in branch 2.4, Can you try it on
> master or branch 2.4?

Although 2.4 will be much better for MPLS with the kernel dataplane, I 
think it would still be nice if that bug could be solved in a 2.3.x 
release because, looking at the mailing list, it seems like many people 
may be running into that issue (scherepnin at gmail.com recently, but 
looking at the explanations it seems many others).

This an issue that I've always ran into with 2.3.x (which is why we use 
pre 2.4 for our work).
So, going 2.4 is nicer for sure, but I suspect 2.4 won't make it into 
distros overnight...

For instance (2.3.2 with linux 3.16 kernel module):

ovs-ofctl dump-flows:
lnx01:~# ovs-ofctl dump-flows ovsbr0
NXST_FLOW reply (xid=0x4):
  cookie=0x0, duration=61.390s, table=0, n_packets=13, n_bytes=1274, 
idle_age=1, ip,in_port=1,nw_dst=2.2.2.2 
actions=push_mpls:0x8847,load:0x2a->OXM_OF_MPLS_LABEL[],output:3

2015-07-21T16:17:06.863Z|00185|dpif(handler7)|WARN|system at ovs-system: 
failed to put[create] (Invalid argument) 
skb_priority(0),in_port(2),skb_mark(0/0),eth(src=4a:ee:e6:1d:f9:58/00:00:00:00:00:00,dst=82:4c:1b:8e:a2:62/00:00:00:00:00:00),eth_type(0x0800),ipv4(src=2.2.2.1/0.0.0.0,dst=2.2.2.2/255.255.255.255,proto=1/0,tos=0/0xfc,ttl=64/0xff,frag=no/0xff),icmp(type=8,code=0), 
actions:set(tunnel(tun_id=0x0,src=0.0.0.0,dst=192.168.122.102,tos=0x0,ttl=64,flags(df))),push_mpls(label=42,tc=0,ttl=64,bos=1,eth_type=0x8847),3

in dmesg:
[  364.756239] openvswitch: netlink: Contain more than allowed mask 
fields (mask_attrs=108cc, mask_allowed=180dc).

The bit in masked attributes that is not in mask allowed is bit 11, 
OVS_KEY_ATTR_ICMP.
I'm not familiar enough with the masking code to debug further.

Best,

-Thomas

_________________________________________________________________________________________________________________________

Ce message et ses pieces jointes peuvent contenir des informations confidentielles ou privilegiees et ne doivent donc
pas etre diffuses, exploites ou copies sans autorisation. Si vous avez recu ce message par erreur, veuillez le signaler
a l'expediteur et le detruire ainsi que les pieces jointes. Les messages electroniques etant susceptibles d'alteration,
Orange decline toute responsabilite si ce message a ete altere, deforme ou falsifie. Merci.

This message and its attachments may contain confidential or privileged information that may be protected by law;
they should not be distributed, used or copied without authorisation.
If you have received this email in error, please notify the sender and delete this message and its attachments.
As emails may be altered, Orange is not liable for messages that have been modified, changed or falsified.
Thank you.




More information about the discuss mailing list