Cisco Switching/Routing :: 3560 - How To Implement CoS Tagging By TCP Port
May 1, 2013
Our enviornment includes 3560 switches and 2800 routers. We have a few remote offices using an application on TCP port 1677 that use far to much bandwidth. Our WAN provider can throttle and police this for us, if I can TAG this traffic, for example all Traffic from Florida using the Groupwise app on TCP uses TCP port 1677 and I want it tagged with CoS 3.
I have a 3560-48 switch running Cisco IOS Software, C3560 Software (C3560-ADVIPSERVICESK9-M), Version 12.2(44)SE3 and i need to implement basic QOS commands to the fast Ethernet interfaces as well as the gig interfaces and Also I need to create port channels on the switch and need what the port channel syntax are as well for that particular IOS version?
I have only read only access and i can't see what the QOS and Port channels syntax should be for that IOS version.
I have attempted to implement DHCP snooping and have been having some strange issues. I have 5 3560s taht I use for my edge and when I attempt to implement on all five, the VLAN that houses my voice data appears to no longer be able to recieve DHCP lease renewals so after the 24 expiration all of my phones lose their configs. Once I roll back the changes the voice VLAN comes back. The other VLANs seem to function correctly as theya re able to renew their DHCP addresses.
The 3560s tie into each other using GIG Ports 1 & 2 and the top and bottom switches tie into our core switch, a 4507. The config that I use is below, failry simple and straightforward.
4 of the 5 switches feed our general office vlans for voice and data however the 5th switch is there for expansion and not in use. As such I have left the config changes in place on it and have tied myself and a colleague into it and have been operating fine for over a week now. So the config that I use seems sound in theory and should work on the other 4 switches with no issue.
I have a 3560 8 port switch. Int gi0/9 is trunked to another switch downstream. When I try to configure int gi0/10 to trunk to a switch upstream the interface on the switch goes down and I have to either reboot the switch or plug directly into the switch and telnet into it to turn off trunking on the interface. When I configure trunking on the interface on the upstream switch that connects to this interface the same happens on that switch. The upstream switch is a 3750 with 12 sfp ports. Several interfaces are trunking to other switches from this switch. Spanning tree is not configured on the 3750 at all , and is not configured on either gi0/10 or gi0/9 on the 3560. I was consoled into the 3560 during a reboot after the interface went down, a message came up that said something like "Spanning Tree returning gigabit ethernet 10 to constant state" Why would I get this message if spanning tree is not enabled on the gig ports on either end of the trunk? There is no loop to require spanning tree to shut down an interface. I have several other 3560's configured as I would like to configure this switch and they are trunking without issue.
I've just installed 2 of these in my workplace on a PLC network.I'm now looking to set one of the ports up as my diagnostic port and would like to be able to mirror any of the other ports to this port.I believe it is called SPAN on Cisco switches.The only reference I can find to it is configuring via Telnet which I haven't got a clue about.On my old Wiedmuller switches it was just a few clicks away.
I have a 3560 switch with the following ports config [code] I would like to use theses ports on a different vlan to connect 4 pc's to them. Can I just remove them from the vlan, remove the trunk switchport and set up on the vlan i want them on with no trunking?
I want to implement port-based and MAC-based in these two switches: 2960 & 3560 (both of them have this IOS version: 12.2(55)SE1). And I haven't found a way to implement both of them at the same time. This is what I got:
ip dhcp use subscriber-id client-id ip dhcp subscriber-id interface-name ip dhcp excluded-address 192.168.0.0 192.168.0.2 ip dhcp excluded-address 192.168.0.251 192.168.0.255
[code]....
With this configuration I can use port-based, but not MAC based. If I remove the first two lines and change the last line for this one:
address 192.168.0.7 client-id 0112.ae1d.af58.60
Then, the computer with that MAC address got the correct IP, but then the port-based doesn't work. Also, I got this line in the interface what I want to use MAC-based:
I am using 3560.IP rouitng is being turned off on this.Curious to know if I will create etherchannel or port channel.I think etherchannel.Correct me if I am wrong.On connecting switches I have vlan10,20,30 to be allowed.I am sure I need to allow these all vlan in 10,20,30 which are on the trunk port on each side switch.Post that will add channel-port lacp and make it in active mode.Is that correct.This way traffic will be load-balanced/aggregated on minimum 2 ports who are the part of this.
Problem is that at some C65K I have directly connected Unix servers and the don't show MAC address at port, and same has happened at 3560 switched where I have too Unix based equipments connected. When use show mac-address interface XXXX, nothis appears at port and tested them with other equipments that worked fine.
Assume I had Catalyst 3560X/3750X with 24 ports. The partnumber is WS-C3560X-24P-LI would like to how is the numbering defined if the switches have a C3KX-NM-10G installed with 4 SFP-GE-L.
We have 7 3560's in 7 different locations connected to our providor for wan access. Our provider has given us a copper cable at each point and we have connected it directly to our 3560 switch at each location. Each port is configured the same way at each location. Each switch is running eigrp.All of the switch ports on each switch are configured as a trunk and vlan 299 had the ip address for the eigrp connection: [code] This setup is working as each switch see's all of the other switches as an eigrp neighbor. We have also made sure that the switch at our head office has spanning tree priority for vlan 299.
So the problem is, if there is a change in the topology at one of the locations it usually causes one or more of the other connections to go down for some reason. We just cannot pinpoint what is causing this change. There are no log's or anything other than an eigrp hold time expired message.?
I have a pair of 3560's configured with dot1q trunks between them carrying a number of VLANs.
Once deployed there will be a requirement for these physical trunks to be disconnected from time to time. Knowing that this is inevitable I am trying to minimise the period of time for the trunks to recover once the physical connectivity is reinstated.
All of the VLANs on the switches are configured for Spanning Tree Rapid PVST. Current time for the trunks/VLANs to come up is around the 4 second mark.
For many years we've had the following vlan and port security config on our 3560s: [code] This has worked great on 12.2(37)SE1, 12.2(40)SE and 12.2(46)SE. However since 12.2(50)SE, and I've tried all the versions since then, we have a problem with 7900 phones and ATA186s taking upwards of 20 minutes before they can get a valid IP number.The problem on the newer IOSes seems to be related to the inactivity aging.On the older IOS versions the mac address of the voice device appears on the voice vlan straight away.
On the newer IOS versions the mac address of the voice device appears on the DATA vlan and seems to be stuck there until the inactivity aging removes it. It then gets re-learned, sometimes on the voice vlan, and sometimes on the data vlan. If you're unlucky and it gets re-learned on the data vlan you've got to wait until the inactivity time ages the address out again. Repeat until the mac address eventually gets learned on the voice vlan. I don't want to be stuck on 12.2(46)SE forever.
How do I limit broadcast/mulitcast traffic on a switchport to e.g. 5000 pps ? I don't want the port to shut down, just block or drop broadcast traffic that exceeds 5000 pps.
Im trying to access the switch to reset the password to factory defaults (please see switch output Astrix has removed customer identifying information for security purposes.) Each time I reboot the switch and try and access the password recover mode this same output below starts and im still not able to access the recovery area of the switch
cisco WS-C3560G-48PS (PowerPC405) processor (revision F0) with 122880K/8184K bytes of memory. Processor board ID FOC1133Y28Q
In one location we have a HP 2510-48 port switch which we need to connect to a Cisco 3560 switch. The problem I have is that we have issues connecting fibre to the HP switch, so I have decided to attached a CAT 5 to Fibre convertor at this end and another to the cisco switch.
My question you can not connect a HP switch to a Cisco switch, is this true or do I just need to confirgure something on the ports.
I'm intending to purchase a switch for work,and I need to limit the bandwidth of one of the ports to 25 Mbit upload and 25 Mbit download (we have 100/100 Mbit connection and the customer is only paying for 25). I been trying to find information on how this could be "properly" done and what kind of switch I need to buy. As far as I have understood, most L2+ switches support outbound rate limiting, but not inbound, and as I only want the customer to have 25 mbit up and down, I need both.
I been looking at a Cisco Catalyst 3560 switch, and I'm first and foremost wondering if I can limit the inbound AND outbound bandwidth on this switch? Perhaps it can even be done on a simpler, cheaper, switch - as I rather not spend more money then necessary?
Lastly, how to do it, limit the inbound and outbound bandwidth on a single port (perhaps on the above mentioned switch, if possible), to 25 Mbit?
I have 3560's in my current environment, operating in the core/distro/access layers. The switches are in a star configuration, performs only layer 2 switching, and utilizes copper (no plans on moving to fiber).
With a fairly limited budget, I've been contemplating on upgrading the central node to a stacked 3750X to eliminate that single point of failure, and trunk the rest of the 3560's to the stacked switch. I wanted to be sure that the 3750X switches will be right for my environment (90-100 hosts), and if what I explained above is a good solution.
I'm also looking like to upgrade 6-7 of my servers (and SAN) with 10GB network cards. Do the ports on the 3750X have port densities capable of 10GB? If not, what switches provide that capability?
Any way to test in a lab what would happen if a tech mistakingly added "switchport voice vlan XX" to a trunk port? I am try to do some RCA on an issue and this has been identified as a possible cause by one of my techs.
The config is Switch1------Switch2--------Switch3 Each interswitch connection is configured as a dot1q trunk with all vlans allowed. The link between switch2 and 3 is where switchport voice vlan 10 was added. Switch1 is a 3750 and 2/3 are 3560's.
I have 2 3560 switches that are running 12.2(25)SEE2. Port security is enabled on some of the ports. Whenever there is a power failure, when power is restored, 1 port on each switch goes to err-disabled. The mac address that causes this is a valid address for that port. Below is the configuration on one of the ports.
show cdp neighbors command shows two devices connected through the same local port?I have a Cisco 3560 that when I issue the "show cdp neighbors" command I get the following results: [code]
the question is why is this showing that Gig0/3 is being used twice to connect to two different devices. I have verified there is no hub connected to this port and in fact Switch3 is not connected to Switch1 at all yet it still shows up in the CDP table.
I have cleared the CDP table on every device at this location and still get the same results.
We are trying to replace the CSS between our firewall and DMZ with a BigIP. Among it's other functions, it will act as the router between the firewall and the DMZ. To make this work, I need to assign vlan tags values for the vlans I create on the BigIP box and these must match the tags on the cisco switches (3550's) How do I find this information on the switch?
If you have a router with multiple direct vanilla FE (non trunked) interfaces on a switch trying to send QOS tagged packets to a wifi bridge several switches away does the trunking in the switched infrastructure mess with the qos tags if no qos is configured on the switches.
Does it depend on the switch? We have new 2960's running 12.2 and a few older 2950's running 12.1
I'm new to networking and was looking for some assistance. First off im using packet tracer to diagram my senario as I will be receiving my equipment next week to deploy.
Hardware to be used:
1. 2 catalyst 3560 switches 2. all connect to a sonic wall router
I have two companies that work in the same office space. I need to keep these companies seperate on their own vlan. They will however need to share the phone system.(Packet tracer file uploaded to give those who have the time to see what I put together.) [code]
I have a 2960 switch connected to another. The I need to verify that vlan0010 on one switch is forwarding tagged traffic between the other switch it is hooked up to through the Gi0/1 port. How do I verify this? I have a server that's multihomed (Broadcom) on the other side an it is supposed to be on this vlan with one of it's network interfaces. We had a pwer outage and now it cannot communicate on this vlan. However, everything else on the vlan can reach all the other nodes accept this server in the front of my building. All the devices in the same room are linked to the same switch which has one port (fa0/17) on vlan0010 and can ping eachother just fine. The server is hooked to port 24 on my server room switch and Gigabit port one goes to a fiber converter all the way to the back. It then gets converted from fiber to cat5e again and links into the switch (2960) in the backroom.
I want to know if there is way to tag traffic with DCSP tags without having to do all the other requirments of QOS setup. All i want to do is just tag traffic at different DCSP values via source and destination IPs. We do not have a need to be priortizing traffic on out internal switches. We just want to tag the traffic so our MPLS provider can distinguish the different types of traffic.
Our environments is primarily 3750s in all offices.
We have a problem with CDP packets on sent by our Cisco 6509's. Unlike our other Cisco switches (4948G, 5020, etc.), the 6509 tags administrative traffic on the native vlan. As a result the CDP packets are sent with an 802.1Q header with a tag of 1. The other switches send the CDP packets untagged on the native vlan. This causes problems because we have non-Cisco devices in our lab that also receive and send CDP, but they do not process the packets that are tagged by the 6509. They see the packets from the 4948 and 5020 just fine.
How can I disable the administrative native vlan tagging on the 6509? Here is the current setup:
nwkdev-6509-1#show vlan dot1q tag native dot1q native vlan tagging is disabled globally nwkdev-6509-1#show interfaces gigabitEthernet 1/9/1 switchport
I've been experimenting with the 'vlan dot1q tag native' command on a switch and it seems as though tagging the native vlan breaks vty access to my access point.With the 'vlan dot1q tag native' commnand applied, I lose management connectivity to the AP with 'no vlan dot1q tag native' applied, connectivity is restored. Why is this? Is it safe to say that one can access the AP via vty lines using ONLY untagged packets?
I'm Confused from the fact that Vlan tagging is done at access port and trunk port always gets tagged packets (untill its case of native vlan).But I still believe in other fact which says tagging happen only when a frame hit the trunk port which means trunk port gets untagged frame and tagging is not possible at access port.
Would like to know where actually this tagging happens ?
and also which command we can use to encapsulate 802.1q protocol to access port ? The way we do at trunk port is #switchport trunk encapsulation dot1q Is the above command applicable for access mode also?
I am trying to implement an etherchannel on a cisco 2901 (IOS 15.1). i have already created the port-channel but i cannot assign the gig interface to the channel group.
I'm trying to implement the UDLD protocol (Cisco-compatible), but face some problems. There is no enough information for implementing this protocol neither in RFC 5171 [URL] nor on [URL] How can I get a more detailed information regarding protocol state machines and timers?