[ovs-discuss] OpenVSwitch - Error packet OFPFMFC_BAD_COMMAND of code ofp_flow_mod_failed_code coming as malformed.

[Ben Pfaff]

Sorry, it's too hard to get the actual OpenFlow messages out of pcap
files.  If you have the OpenFlow messages, rather than the ".pcap"s,
I'll take a look.

On Wed, May 23, 2012 at 05:35:28AM +0000, Prabina Pattnaik wrote:
> Please find attached following files with this mail, it is coming malformed in wireshark 1.4.11 (with openflow plugin 1.0):-
> 1. malformed OFPFMFC_BAD_COMMAND error packet
> 2. malformed OFPFMFC_ALL_TABLES_FULL error packet
> 
> Observation-
> Error packets which are above 64 bytes of code "ofp_flow_mod_failed_code" are coming as malformed. Other error packets of different code are coming ok.
> After applying our patch for limit the table flows in OVS 1.2.2, OFPFMFC_ALL_TABLES_FULL error packet was also malformed, so enclosing that also for reference.
>  
> Regards,
> Prabina
> 
> 
> -----Original Message-----
> From: Ben Pfaff [mailto:blp at nicira.com] 
> Sent: Tuesday, May 22, 2012 11:41 PM
> To: Prabina Pattnaik
> Cc: bugs at openvswitch.org
> Subject: Re: [ovs-discuss] OpenVSwitch - Error packet OFPFMFC_BAD_COMMAND of code ofp_flow_mod_failed_code coming as malformed.
> 
> Can you show me this malformed error packet, and tell me in what way
> it is malformed?
> 
> On Tue, May 22, 2012 at 09:50:20AM +0000, Prabina Pattnaik wrote:
> > Hi Ben,
> > 
> > We agree that implementation of 'at least 64 bytes' is as per spec (to allow data above 64 bytes depends on vendor specific switch implementation).
> > But in current implementation, when the code (mentioned below) limits the size of data to 64 bytes (if data field of the error packet is more than 64 bytes), a malformed OFPFMFC_BAD_COMMAND error packets gets generated (visible on wireshark with openflow plugin).
> > 
> > As per our analysis, removal of code to limit the data size to 64, resolves the issue of malformed packet.
> > 
> > File: lib/ofputil.c 
> > Routine ofputil_encode_error_msg
> > 
> > If (oh){
> >     Xid = oh->xid;
> >     data = oh;
> >     len = ntohs(oh->length);
> >    - if (len > 64){
> >    -   len =64;
> >    - }
> >   }
> > 
> >  
> > Regards,
> > Prabina
> > 
> > 
> > -----Original Message-----
> > From: Ben Pfaff [mailto:blp at nicira.com] 
> > Sent: Monday, May 21, 2012 10:24 PM
> > To: Prabina Pattnaik
> > Cc: bugs at openvswitch.org
> > Subject: Re: [ovs-discuss] OpenVSwitch - Error packet OFPFMFC_BAD_COMMAND of code ofp_flow_mod_failed_code coming as malformed.
> > 
> > On Mon, May 21, 2012 at 11:15:28AM +0000, Prabina Pattnaik wrote:
> > > As per openflow 1.0 spec "data" contains at "least" first 64 bytes
> > > of failed request. This is working correct also in current OVS 1.2.2
> > > code.
> > >
> > > But it is taking an extra check in code (lib/ofp-util.c) on "data"
> > > length that:-
> > >
> > > if data is greater than 64 bytes then it would be truncated to 64
> > > bytes which is making this packet as malformed. OVS code is working
> > > on "at most" 64 bytes length instead of "at least" 64 bytes.
> > > 
> > > To avoid the truncation of 'data field ' to 64 bytes, the condition
> > > to check the length of data field in file (lib/ofp-util.c), routine
> > > ofputil_encode_error_msg can be removed.
> > 
> > I disagree with your interpretation.  "At least" 64 bytes means that
> > OVS is allowed to truncate the inner message after 64 bytes.  That is
> > what it is doing.
> > 
> > 
> > 
> > DISCLAIMER:
> > 
> > -----------------------------------------------------------------------------------------------------------------------
> > 
> > The contents of this e-mail and any attachment(s) are confidential and
> > intended
> > 
> > for the named recipient(s) only. 
> > 
> > It shall not attach any liability on the originator or NECHCL or its
> > 
> > affiliates. Any views or opinions presented in 
> > 
> > this email are solely those of the author and may not necessarily reflect the
> > 
> > opinions of NECHCL or its affiliates. 
> > 
> > Any form of reproduction, dissemination, copying, disclosure, modification,
> > 
> > distribution and / or publication of 
> > 
> > this message without the prior written consent of the author of this e-mail is
> > 
> > strictly prohibited. If you have 
> > 
> > received this email in error please delete it and notify the sender
> > 
> > immediately. .
> > 
> > -----------------------------------------------------------------------------------------------------------------------
> 
> 
> 
> DISCLAIMER:
> 
> -----------------------------------------------------------------------------------------------------------------------
> 
> The contents of this e-mail and any attachment(s) are confidential and
> intended
> 
> for the named recipient(s) only. 
> 
> It shall not attach any liability on the originator or NECHCL or its
> 
> affiliates. Any views or opinions presented in 
> 
> this email are solely those of the author and may not necessarily reflect the
> 
> opinions of NECHCL or its affiliates. 
> 
> Any form of reproduction, dissemination, copying, disclosure, modification,
> 
> distribution and / or publication of 
> 
> this message without the prior written consent of the author of this e-mail is
> 
> strictly prohibited. If you have 
> 
> received this email in error please delete it and notify the sender
> 
> immediately. .
> 
> -----------------------------------------------------------------------------------------------------------------------