Hello Its loop avoidance initiating it isn't a problem- Both R2 -R3 will advertise that prefix originating from R1 to R4 and R4 sees both paths from those rtts to that prefix as equal so it will possibly choose based on the oldest route or lowest bgp rid. R4 will choose R2 as its best path so it will advertise the prefix it receives from R2 towards R3, As such R3 will also have two paths to that prefix and rtr2 will only have the one unless it loses connection towards R1 then it will obtain the prefix via rtr 4 -rtr3 path. ... View more

Hello Below is a slight revised configuration for your NAT and icmp replies.   no access-group INSIDE_access_in in interface INSIDE no route OUTSIDE 0.0.0.0 0.0.0.0 66.27.194.152 1 no route OUTSIDE 0.0.0.0 0.0.0.0 66.27.194.151 1 no access-list INSIDE_access_in extended permit ip any any no access-list 100 extended permit icmp any any no object network obj_any no nat (INSIDE,OUTSIDE) source dynamic OBJ_NETWORK_ALL interface object network OBJ_NETWORK_ALL no subnet 0.0.0.0 0.0.0.0 subnet 10.0.1.0 255.255.255.0 object-group network NAT network-object object OBJ_NETWORK_ALL access-list INSIDE_access_in extended permit icmp any object OBJ_NETWORK_ALL echo-reply access-group INSIDE_access_in in interface OUTSIDE nat (INSIDE,OUTSIDE) after-auto source dynamic NAT interface ... View more

Hello Qos priority is low latency queuing (LLQ) with is a version of priority queuing (PQ) but it doesn't starve lower priority class when congestion is on the link like PQ did Policing is more invasive and will remark or drop packets once its defied police limit has been reached as apposed to shaping which wont drop packets but will buffer them until its can send them. ... View more

Hello   @carl_townshend wrote: Hi All We are having some issues with audio/video such as webex over the internet. We have a 500Mbit circuit coming in which can get very busy during times. We have frequent audio/video drop outs due to the congestion, it is generally in the download direction. I know there is nothing you can do once on the internet with qos, is there any way I can control this at least coming in to me, or would it need to happen further upstream and so I have no control? My plan was to put a qos policy on the inside interface of the internet router facing us, and on the outbound direction i.e coming towards us, this would control the download side. Or will it not work as it needs to be controlled further upstream? cheers If you have a congested wan link then depending how often this is qos would only be temporary solution you may eventually need to increase it , But your right once traffic has hit your wan interface ingress it has already traversed the wan  link so not much really you can do from your end. However you can negotiate with you isp what CIR you have and shape accordingly to that, Is it possible the other end of your wan link has a higher CIR then you do? -  If so they may be overwhelming the connection towards you (egress blocking) so also possibly check that and if so ask them to shape to your CIR. ... View more

Hello What does the eigrp topology D/B state, any inaccessible routes showing, Do you have any summary routes conflicting with the routes your trying to redistribute? Sh ip eigrp topology ... View more

Hello fan.gao@academy.com wrote: so basically is match tag seems not work in route map when I redistribute it from EIGRP to BGP . When you redistribute igp into bgp then the asn tag is lost correct?  Try the following but you may find the origin of the route is incomplete , please verify if that is the case. route-map eigrp-into-bgp set as-path tag router bgp xxx redistribute eigrp xx match external route-map eigrp-into-bgp   ... View more

Hello switch ip route 0.0.0.0 0.0.0.0 vlan 1 192.168.2.1  Also allow vlan 1 on the trunks   rtr no int vlan 40 ip route 192.168.10.0 255.255.255.0 vlan1 192.168.2.7   ip route 192.168.40.0 255.255.255.0 vlan1 192.168.2.7   Router rip ver 2 no auto summary  network 192.168.2.0 redistribute static    int x/x description link to switch  Switchport mode access switchport  access vlan 1 spanning-tree portfast   ... View more

Hello Just like to add you dont really need to set the tags on the redistributed bgp routes, by default the asn they originated from will be the tagged value in the eigrp topology database,  So try matching on the asn tag instead? sh ip eigrp topology | in tag ... View more

Hello Do you have enough space in the cam table for new entries, any port security applied to the port the device is attached to possibly negating learning of the devices L2 address, Another alternative is buggy nx-os software What is the device attached, it is a server? ... View more

Hello On all rip rtrs make sure you have rip verion2  and auto-summerization disabled   all rtrs router rip ver2 no auto-summary BB tap rtr router rip version 2 network 10.0.0.0 network 172.16.0.0 network 192.168.1.0 network 192.168.4.0 no auto-summary ... View more

Hello Arp and mac table have a very different default timeout values ( arp 4hrs) (mac 5 mins), So inactivty of the end host could mean that the mac entries have expired and been flushed from the cam table. An oui lookup on that mac-address states it is an HP device (possible printer) so it is indeed possible that devcie has gone into a standby mode and but due the the high timeout value of the arp table it is still registered in arp but not in the cam table. ... View more

Hello You could try Policy based routing to achieve this:   Possible example: interface x/x nameif Isp2 ip address x.x.x.x y.y.y.y <Isp2 wan facing interface) nat (inside,Isp2) source dynamic any interface object-group services Srv_ports port-object eq 80 port-object eq 443 access-list Srv-acl extended permit tcp host 172.16.1.1 host <isp host) object-group Srv_ports   route-map Pbr_rm match ip address Srv-acl set ip next-hop y.y.y.y <– ISP2next hop)   interface x/x nameif inside policy-route route-map Pbr_rm ... View more

Hello HA suggests that the servers are some way linked together however your explanation is very vague as to the running of these server interms of what they provide and how they are physically connected.   However if you just wish to have the one servers link active while another is disabled then possible a simple flex link could perform the job but again it all depends on how the servers are physically connected (directy to the router or via switch.   Rtr int X/0 description Link to primary server backup interface int X/1 backup delay 10 30 int X/1 description Link to secondary server Switch int X/0 description switchport to primary server switchport backup interface int X/1 preempt delay xx int X/1 description switchport to secondary server ... View more