Re: [ovs-discuss] Can anyone familiar with the following code please clarify for me?

2012-05-21 Thread Ben Pfaff
On Tue, May 22, 2012 at 02:11:35PM +0800, YIMIN CHEN wrote: > I am working on ovs code based on openvswitch-1.1.0pre2, it is kinda > of old, I know. I ran across this code that I am having hard time > understanding. If anyone is familiar with this part of code, could you > please help me? > > >

[ovs-discuss] Can anyone familiar with the following code please clarify for me?

2012-05-21 Thread YIMIN CHEN
Hi, I am working on ovs code based on openvswitch-1.1.0pre2, it is kinda of old, I know. I ran across this code that I am having hard time understanding. If anyone is familiar with this part of code, could you please help me? /* Compares the specified field in 'a' and 'b'. Returns true if the f

Re: [ovs-discuss] ovs-vswitchd number of datapaths

2012-05-21 Thread Vjekoslav Brajkovic
Hi Ben, I've manged to figure out what the problem is. For some reason ovs-appctl did not log all information -- after setting the verbose flag and the log-file path, I noticed that ovs-vswitchd had been running out of file descriptors. Increasing the maximum number of open FDs fixed the problem.

Re: [ovs-discuss] Open VSwitch - Idle timeout is not resetting when packet matching the flow entry

2012-05-21 Thread Justin Pettit
On May 21, 2012, at 10:33 AM, Ben Pfaff wrote: > On Mon, May 21, 2012 at 10:37:32AM +, Varun Gupta wrote: >> We found a issue related to idle timeout of flow entry while working on OVS >> 1.2.2. >> >> * Bug Description >> Idle timeout is not resetting when packet matching the flow entry. >

Re: [ovs-discuss] Open VSwitch - Idle timeout is not resetting when packet matching the flow entry

2012-05-21 Thread Ben Pfaff
On Mon, May 21, 2012 at 10:37:32AM +, Varun Gupta wrote: > We found a issue related to idle timeout of flow entry while working on OVS > 1.2.2. > > * Bug Description > Idle timeout is not resetting when packet matching the flow entry. I think this is the same as your previous report, is the

[ovs-discuss] Open VSwitch - Idle timeout is not resetting when packet matching the flow entry

2012-05-21 Thread Varun Gupta
Hi, We found a issue related to idle timeout of flow entry while working on OVS 1.2.2. * Bug Description Idle timeout is not resetting when packet matching the flow entry. * What you did that make the problem appear? 1.Made the following bridge: ovs-vsctl show e999897a-932f

Re: [ovs-discuss] Open VSwitch - Idle timeout is not resetting when packet matching the flow entry

2012-05-21 Thread Ben Pfaff
On Mon, May 21, 2012 at 10:42:35AM +, Varun Gupta wrote: > We found a issue related to idle timeout of flow entry while working on OVS > 1.2.2. > > * Bug Description > Idle timeout is not resetting when packet matching the flow entry. I can't reproduce this with current master.

Re: [ovs-discuss] OpenVSwitch - Able to add more than maximum number of flow entries in Flow table

2012-05-21 Thread Ben Pfaff
On Mon, May 21, 2012 at 11:20:00AM +, Prabina Pattnaik wrote: > This variable value is hard coded with 100 with OVS-1.2.2 and we > are able to add more than 100 entries in a same flow table. If you want a hard limit on flow entries, you can implement it in OVS 1.4.x and later via the F

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

2012-05-21 Thread Ben Pfaff
On Mon, May 21, 2012 at 11:15:28AM +, 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:

Re: [ovs-discuss] Open VSwitch - No 'OFPBRC_BAD_VERSION' error packet received when there is version mismatch between feature request and feature reply

2012-05-21 Thread Ben Pfaff
On Mon, May 21, 2012 at 10:52:55AM +, Varun Gupta wrote: > Switch is not notifying the controller for OFPBRC_BAD_VERSION, /* > ofp_header.version not supported. */ when there is version mismatch > between feature request and feature reply. The switch was resetting > the connection and only a wa

Re: [ovs-discuss] OpenVSwitch - "OFPBRC_BAD_VERSION" error packet not received when there is version mismatch between stats request and stats reply

2012-05-21 Thread Ben Pfaff
On Mon, May 21, 2012 at 10:50:19AM +, Varun Gupta wrote: > Switch is not notifying the controller for OFPBRC_BAD_VERSION, /* > ofp_header.version not supported. */ when there is version mismatch > between stats request and stats reply. The switch was resetting the > connection and only a warnin

[ovs-discuss] OpenVSwitch - Able to add more than maximum number of flow entries in Flow table

2012-05-21 Thread Prabina Pattnaik
Hi, I have found following issue while working on OVS 1.2.2 Bug Description As per openflow 1.0 spec, the Table Statistics structure define the variable. uint32_t max_entries; /* Max number of entries supported. */ This variable value is hard coded with 100 with OVS-1.2.2 and we are able

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

2012-05-21 Thread Prabina Pattnaik
Hi, I was working on OVS 1.2.2 and found malformed error packet. Bug Description Error packet generated by switch for FLOW_MOD_FAILED is coming as malformed on traffic analyzer (wireshark with openflow plugin) * What you did that make the problem appear? 1. I made following bridge.

[ovs-discuss] Open VSwitch - No 'OFPBRC_BAD_VERSION' error packet received when there is version mismatch between feature request and feature reply

2012-05-21 Thread Varun Gupta
Hi, We are evaluating OVS 1.2.2 and would like to report an unexpected behavior of OVS. * Bug Description Switch is not notifying the controller for OFPBRC_BAD_VERSION, /* ofp_header.version not supported. */ when there is version mismatch between feature request and feature reply. The swit

[ovs-discuss] OpenVSwitch - "OFPBRC_BAD_VERSION" error packet not received when there is version mismatch between stats request and stats reply

2012-05-21 Thread Varun Gupta
Hi, We are evaluating OVS 1.2.2 and would like to report an unexpected behavior of OVS. * Bug Description Switch is not notifying the controller for OFPBRC_BAD_VERSION, /* ofp_header.version not supported. */ when there is version mismatch between stats request and stats reply. The switch w

[ovs-discuss] Open VSwitch - Idle timeout is not resetting when packet matching the flow entry

2012-05-21 Thread Varun Gupta
Hi, We found a issue related to idle timeout of flow entry while working on OVS 1.2.2. * Bug Description Idle timeout is not resetting when packet matching the flow entry. * What you did that make the problem appear? 1.Made the following bridge: ovs-vsctl show e999897a-932f

Re: [ovs-discuss] arp spoofing

2012-05-21 Thread Sergio Kviato
On May 21, 2012, at 5:11 AM, faicker mo wrote: > > On 2012-5-20, at 上午12:27, Sergio Kviato wrote: > >> >> >> Sent from my iPhone >> >> On May 19, 2012, at 19:02, faicker mo wrote: >> >>> >>> On 2012-5-19, at 下午11:11, Ben Pfaff wrote: >>> On Sat, May 19, 2012 at 09:30:40PM +0800, fai

Re: [ovs-discuss] 回复： Question about openvswitch's performance

2012-05-21 Thread Justin Pettit
I don't know precisely off the top of my head. The number varies depending on the number of actions, but a rough approximation (with quite a bit of wiggle room) is 1000 bytes per rule. --Justin On May 20, 2012, at 11:53 PM, 星言 墨虚 wrote: > Sorry, I forget reply to all sender. > > You means