From mboxrd@z Thu Jan 1 00:00:00 1970 From: Cong Wang Subject: Re: [Patch net] bridge: do not expire mdb entry when bridge still uses it Date: Fri, 08 Mar 2013 10:44:06 +0800 Message-ID: <1362710646.9773.25.camel@cr0> References: <1362708423-23932-1-git-send-email-amwang@redhat.com> <20130308023151.GA27155@gondor.apana.org.au> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: Stephen Hemminger , netdev@vger.kernel.org, bridge@lists.linux-foundation.org, "David S. Miller" , Adam Baker To: Herbert Xu Return-path: In-Reply-To: <20130308023151.GA27155@gondor.apana.org.au> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: bridge-bounces@lists.linux-foundation.org Errors-To: bridge-bounces@lists.linux-foundation.org List-Id: netdev.vger.kernel.org On Fri, 2013-03-08 at 10:31 +0800, Herbert Xu wrote: > On Fri, Mar 08, 2013 at 10:07:03AM +0800, Cong Wang wrote: > > From: Cong Wang > > > > This is a long-standing bug and reported several times: > > https://bugzilla.redhat.com/show_bug.cgi?id=880035 > > http://marc.info/?l=linux-netdev&m=136164389416341&w=2 > > > > This bug can be observed in virt environment, when a KVM guest > > communicates with the host via multicast. After some time (should > > be 260 sec, I didn't measure), the multicast traffic suddenly > > terminates. > > > > This is due to the mdb entry for bridge itself expires automatically, > > it should not expire as long as the bridge still generates multicast > > traffic. It should expire when the bridge leaves the multicast group, > > OR when there is no multicast traffic on this bridge. > > > > I fix this by adding another bool which is set when there is > > multicast traffic goes to the bridge, cleared in the expire timer and > > when IGMP leave is received. I ran omping for 15 minutes, everything > > looks good now. > > I gather from the bugzilla entry that this happens where there is > no querier in the network. > > So my question is why does this only affect the bridge port and > not other ports on the bridge? > This is a good question. It is due to inside br_handle_frame_finish() we use 'skb2' to decide if we deliver packets to bridge itself, and in this case 'skb2' is non-NULL only when: if (mdst || BR_INPUT_SKB_CB_MROUTERS_ONLY(skb)) { if ((mdst && mdst->mglist) || br_multicast_is_router(br)) skb2 = skb; // <======== HERE br_multicast_forward(mdst, skb, skb2); For other ports, br_multicast_forward() will always forward 'skb' to them. Thanks. From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Message-ID: <1362710646.9773.25.camel@cr0> From: Cong Wang Date: Fri, 08 Mar 2013 10:44:06 +0800 In-Reply-To: <20130308023151.GA27155@gondor.apana.org.au> References: <1362708423-23932-1-git-send-email-amwang@redhat.com> <20130308023151.GA27155@gondor.apana.org.au> Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Mime-Version: 1.0 Subject: Re: [Bridge] [Patch net] bridge: do not expire mdb entry when bridge still uses it List-Id: Linux Ethernet Bridging List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Herbert Xu Cc: Stephen Hemminger , netdev@vger.kernel.org, bridge@lists.linux-foundation.org, "David S. Miller" , Adam Baker On Fri, 2013-03-08 at 10:31 +0800, Herbert Xu wrote: > On Fri, Mar 08, 2013 at 10:07:03AM +0800, Cong Wang wrote: > > From: Cong Wang > > > > This is a long-standing bug and reported several times: > > https://bugzilla.redhat.com/show_bug.cgi?id=880035 > > http://marc.info/?l=linux-netdev&m=136164389416341&w=2 > > > > This bug can be observed in virt environment, when a KVM guest > > communicates with the host via multicast. After some time (should > > be 260 sec, I didn't measure), the multicast traffic suddenly > > terminates. > > > > This is due to the mdb entry for bridge itself expires automatically, > > it should not expire as long as the bridge still generates multicast > > traffic. It should expire when the bridge leaves the multicast group, > > OR when there is no multicast traffic on this bridge. > > > > I fix this by adding another bool which is set when there is > > multicast traffic goes to the bridge, cleared in the expire timer and > > when IGMP leave is received. I ran omping for 15 minutes, everything > > looks good now. > > I gather from the bugzilla entry that this happens where there is > no querier in the network. > > So my question is why does this only affect the bridge port and > not other ports on the bridge? > This is a good question. It is due to inside br_handle_frame_finish() we use 'skb2' to decide if we deliver packets to bridge itself, and in this case 'skb2' is non-NULL only when: if (mdst || BR_INPUT_SKB_CB_MROUTERS_ONLY(skb)) { if ((mdst && mdst->mglist) || br_multicast_is_router(br)) skb2 = skb; // <======== HERE br_multicast_forward(mdst, skb, skb2); For other ports, br_multicast_forward() will always forward 'skb' to them. Thanks.