Release Notes for the Cisco ME 3800X, ME 3600X and ME 3600X-24CX Switches, Cisco IOS Release 15.3(3)S and Later Releases
Available Languages
Table of Contents
Software Licenses and Features
Finding the Software Version and Feature Set
Installing Software Images and Licenses
Open Caveats in Cisco IOS Release 15.3(3)S7
Open Caveats in Cisco IOS Release 15.3(3)S6
Open Caveats in Cisco IOS Release 15.3(3)S5
Open Caveats in Cisco IOS Release 15.3(3)S4
Open Caveats in Cisco IOS Release 15.3(3)S3
Closed Caveats in Cisco Release 15.3(3)S3
Open Caveats in Cisco Release 15.3(3)S2
Open Caveats in Cisco Release 15.3(3)S1a
Open Caveats in Cisco Release 15.3(3)S
Resolved Caveats in Cisco IOS Release 15.3(3)S9
Resolved Caveats in Cisco IOS Release 15.3(3)S7
Resolved Caveats in Cisco Release 15.3(3)S6
Resolved Caveats in Cisco Release 15.3(3)S5
Resolved Caveats in Cisco Release 15.3(3)S4
Resolved Caveats in Cisco Release 15.3(3)S2
Resolved Caveats in Cisco Release 15.3(3)S1a
Resolved Caveats in Cisco Release 15.3(3)S
Obtaining Documentation and Submitting a Service Request
Release Notes for the Cisco ME 3800X, ME 3600X and ME 3600X-24CX Switches, Cisco IOS Release 15.3(3)S and Later Releases
This release note includes important information about Cisco IOS Release 15.3(3)S and later releases that run on the Cisco ME 3800X, ME 3600X and ME 3600X-24CX switches.
This release note also includes the limitations, restrictions, and caveats that apply to the release.
You can verify the release notes in the following ways:
- If you are installing a new switch, see the Cisco IOS release label on the rear panel of your switch.
- If your switch is on, use the show version privileged EXEC command. See the “Finding the Software Version and Feature Set” section.
- If you are upgrading to a new release or a different image, see the software upgrade filename for the software version. See the “Deciding Which Files to Use” section.
For the complete list of Cisco ME 3800X, ME 3600X and ME 3600X-24CX switch documentation, see the “Related Documentation” section.
You can download the switch software from this site (registered Cisco.com users with a login password):
http://www.cisco.com/cisco/software/navigator.html?a=ahttp://www.cisco.com/cisco/web/download/index.htmli=rpm
Contents
- “Hardware Support” section
- “Software Licenses and Features” section
- “Upgrading the Switch Software” section
- “Installation Notes” section
- “New Software Features” section
- “Important Notes” section
- “Open Caveats” section
- “Resolved Caveats” section
- “Related Documentation” section
- “Obtaining Documentation and Submitting a Service Request” section
Hardware Support
Software Licenses and Features
If you have a service support contract and order a software license, or if you order a switch, you receive the universal software image, available in crypto and noncrypto versions. If you do not have a service support contract, such as a SMARTnet contract, download the image from Cisco.com.
The ME 3600X supports these licenses:
- Metro IP access is the universal image.
- Advanced Metro IP access license.
- 10 Gigabit Ethernet upgrade license—enables 10 Gigabit Ethernet on the SFP+ uplink ports.
For differences in feature support for each license, see Table 2 and Table 3.
The ME3600X-24CX supports the above licences including the following:
The ME 3800X supports these licenses and a scaled license that can be installed with any of these licenses to increase the supported values for that license, for example, more MAC addresses, VLANs, IPv4 routes, and so on.
- Metro Ethernet services is the universal image.
- Metro IP services license.
- Metro Aggregation services license.
- Scaled license for any of the above licenses.
For differences in feature support for each license, see Table 4 and Table 5.
To install a software image, see the “Upgrading the Switch Software” section and the “Working with the Cisco IOS File System, Configuration Files, and Software Images” chapter in the software configuration guide.
To install a software license, see the “Cisco IOS Software Activation Tasks and Commands” chapter in the Cisco IOS Software Activation Configuration Guide:
http://www.cisco.com/en/US/docs/ios/csa/configuration/guide/csa_commands.html
An emergency evaluation license is embedded in the software image and does not require the installation of a license file. Specify which evaluation license to enable by using the license boot level command.
Enabling evaluation license on an ME 3800X:
Note Only MetroAggrServices is supported during evaluation. Accept the EULA.
Enabling evaluation license on an ME 3600X:
Note Only AdvancedMetroIPAccess is supported during evaluation. Accept the EULA.
After entering the license boot level command, you are prompted to accept the End-User Licensing Agreement (EULA). After accepting the EULA, exiting configuration mode, and saving the running configuration to memory, reload the switch to apply the evaluation license.
Note The evaluation period is valid for 60 days. When the 60 day evaluation period ends, the evaluation license will be unusable after the next reload.
Upon installation of a license file, the license will automatically update to the new license type. There is no need to clear the evaluation license.
Table 3 ME 3600X and ME 3600X-24CX License Scaling and Template
Upgrading the Switch Software
- “Finding the Software Version and Feature Set” section
- “Deciding Which Files to Use” section
- “Installing Software Images and Licenses” section
Finding the Software Version and Feature Set
The Cisco IOS image is stored as a bin file in a directory that is named with the Cisco IOS release. The image is stored on the system board flash device (flash:).
Note The flash memory can store a maximum of two IOS images or tar files. If you try to copy or archive upgrade beyond the flash memory capacity, the action aborts.
You can use the show version privileged EXEC command to see the software version that is running on your switch. The second line of the display shows the version.
You can also use the dir filesystem : privileged EXEC command to see the directory names of other software images that you might have stored in flash memory.
Deciding Which Files to Use
The software installation procedures in these release notes describe how to perform the upgrade by using a combined tar file. This file contains the Cisco IOS image file. To upgrade the switch through the command-line interface (CLI), use the tar file and the archive download-sw privileged EXEC command.
Installing Software Images and Licenses
The Cisco ME 3800X, ME 3600X and ME 3600X-24CX switches are shipped with the latest software image installed. Follow the instructions in this section if you need to reinstall or upgrade the software image.
Before installing your switch software, make sure that you have archived copies of the current Cisco IOS release and the Cisco IOS release to which you are upgrading. You should keep these archived images until you have upgraded all devices in the network to the new Cisco IOS image and until you have verified that the new Cisco IOS image works properly in your network.
Cisco routinely removes old Cisco IOS versions from Cisco.com. See End of Sale and End of Life Products at this URL:
http://www.cisco.com/en/US/products/sw/iosswrel/prod_category_end_of_life.html
You can copy the software image file on the flash memory to the appropriate TFTP directory on a host by using the copy flash: tftp: privileged EXEC command. You can also configure the switch as a TFTP server to copy files from one switch to another without using an external TFTP server by using the tftp-server global configuration command. For more information about the tftp-server command, see the “Basic File Transfer Services Commands” section of the Cisco IOS Configuration Fundamentals Command Reference at this URL:
http://www.cisco.com/en/US/docs/ios-xml/ios/fundamentals/command/Cisco_IOS_Configuration_Fundamentals_Command_Reference.html
This procedure is for copying the combined tar file to the switch. You copy the file to the switch from a TFTP server and extract the files. You can download an image file and replace or keep the current image.
To download software, follow these steps:
Step 1 Use Table 6 to identify the file that you want to download.
Step 2 Locate the software image file:
a. If you are a registered customer, go to this URL and log in.
http://www.cisco.com/cisco/software/navigator.html?a=ahttp://www.cisco.com/cisco/web/download/index.htmli=rpm
b. For ME 3800X, navigate to Switches > Service Provider Switches - Ethernet Aggregation .
For ME 3600X, navigate to Switches > Service Provider Switches - Ethernet Access .
c. Navigate to your switch model.
d. Click IOS Software , then select the latest IOS release.
Note When you select a crypto graphic image, you must also accept the terms and conditions of using crypto graphic images.
Step 3 Download the image to a TFTP server and make sure that the server is properly configured.
For more information, refer to Appendix B in the software configuration guide for this release.
Step 4 Log into the switch through the console port or a Telnet session.
Step 5 (Optional) Ensure that you have IP connectivity to the TFTP server by entering this privileged EXEC command:
For more information about assigning an IP address and default gateway to the switch, refer to the software configuration guide for this release.
Step 6 Download the image file from the TFTP server to the switch by entering this privileged EXEC command:
- For // location , specify the IP address of the TFTP server.
- For / directory / image-name .tar , specify the directory (optional) and the image to download. Directory and image names are case sensitive.
- The /overwrite option overwrites the software image in flash memory with the downloaded one.
- The /reload option reloads the system after downloading the image unless the configuration has been changed and not saved.
This example shows how to download an image from a TFTP server at 192.0.2.1 and to overwrite the image on the switch:
You can also download the image file from the TFTP server to the switch and keep the current image by using the /leave-old-sw option instead of the /overwrite option.
Note There can be only two image directories in the flash memory.
The installation process extracts the tar file with all the files and the IOS image, and sets the BOOT directory to the created directory in flash memory. The process takes approximately 5 to 10 minutes, and at some stages might appear to have stopped.
Step 7 The switch is configured to boot automatically, but you can enter the show boot privileged EXEC command to verify the boot path list and see if a manual boot is necessary.
Step 8 Save the configuration and reload the switch.
After the installation, the switch is running the universal image. Follow these steps to install a purchased license with increased capabilities. To purchase a license, contact Cisco.
Step 1 Copy the license file to flash or TFTP.
Step 2 Enter this command to install the license:
Step 3 Enter these commands to boot from the new license:
Step 4 If you have a a scaled license, install the scaled license
Note To revert to a non-scaled license, enter the license clear scaled_license_name privileged EXEC command.
New Software Features
The following sections provide information on new software features:
Cisco IOS Release15.3(3)S
The following new software features are introduced in the Cisco IOS Release 15.3(3)S:
- Auto-IP - The Auto-IP feature addresses the problem of manually reconfiguring nodes during insertion, deletion, and movement of nodes within an auto-IP ring. The auto-IP feature automatically provides IP addresses to the node interfaces inserted into an auto-IP ring.
http://www.cisco.com/en/US/docs/ios-xml/ios/ipaddr_ipv4/configuration/xe-3s/ipv4-xe-3s-book_chapter_011.html
The following commands were introduced or modified: auto-ip-ring, debug auto-ip-ring, show auto-ip-ring.
- Hierarchical Color-Aware Policing - This feature provides two levels of policing where the policer ordering is evaluated from child to parent, and there is preferential treatment of certain traffic at the parent level.
http://www.cisco.com/en/US/docs/switches/metro/me3600x_3800x/software/release/15.3_3_S/configuration/guide/swhier_ca_police.html
http://www.cisco.com/en/US/docs/switches/metro/me3600x_3800x/software/release/15.3_3_S/chassis/configuration/guide/OC3_Ifc_Module.html
- Multiprotocol Label Switching (MPLS) Load Balancing - The Cisco ME3800X and ME3600X Switches support IPv4 and IPv6 load balancing at the LER and LSR. Effective with Cisco IOS Release 15.3(3)S, the following features are supported:
– Layer 2 VPN load balancing at LER and LSR
– Layer 3 VPN load balancing at LER and LSR
– Load balancing over port channel at LER and LSR
http://www.cisco.com/en/US/docs/switches/metro/me3600x_3800x/software/release/15.3_3_S/configuration/guide/swmplsloadbalancing.html
http://www.cisco.com/en/US/docs/switches/metro/me3600x_3800x/software/release/15.3_3_S/chassis/configuration/guide/sw_port_licensing.html
http://www.cisco.com/en/US/docs/switches/metro/me3600x_3800x/software/release/15.3_3_S/configuration/guide/swqos.html
- Service Performance for the Cisco ME3600X-24CX switch - Y.1564 is an Ethernet service activation test methodology, and is the standard for turning up, installing, and troubleshooting Ethernet-based services. Y.1564 is the only standard test methodology that allows a complete validation of Ethernet service level agreements (SLAs) in a single test.
Service performance testing is designed to measure the ability of a device under test (DUT) or a network under test to properly forward traffic in different states.
http://www.cisco.com/en/US/docs/switches/metro/me3600x_3800x/software/release/15.3_3_S/chassis/configuration/guide/swy1564.html
- Y.1731 Performance Monitoring - Each Maintenance End Point (MEP) transmits frames with one-way ETH-DM information to its peer MEP in a point-to-point ME. This transmission facilitates either one-way frame delay or one-way frame delay variation measurements, or both, at the peer MEP. The one-way ETH-DM is supported only on the Cisco ME3600X-24CX-M Switch.
http://www.cisco.com/en/US/docs/switches/metro/me3600x_3800x/software/release/15.3_3_S/configuration/guide/swy1731pm.html
Important Notes
- FRR failover time is delayed by seconds, when a neighboring router static PoCH is shut down.
- The egress policing is now supported on non-priority queue.
- To adjust the channel on a DMDM-XFP-C tunable SFP, proceed as follows:
Note Based on the ITU grid, the channel number can vary.
Please wait till you receive the "Image Extraction Done" success message, or any other message reporting Failure !!!!!If the operation fails, a failure message is displayed citing the cause of failure.
If the operation is successful, no message is displayed.
Note The archive download files are extracted to a temporary RAM disk. The RAM disk is deleted after the new files are populated in the flash. The "Deleting Ramdisk" informational message is displayed during this RAM filesystem format operation. If the RAM FS is not formatted successfully, the error message "Filesystem "*" not formatted” is also displayed along with this informational message.
Open Caveats
The following sections provide information about open caveats:
- Open Caveats in Cisco IOS Release 15.3(3)S7
- Open Caveats in Cisco IOS Release 15.3(3)S6
- Open Caveats in Cisco IOS Release 15.3(3)S5
- Open Caveats in Cisco IOS Release 15.3(3)S4
- Open Caveats in Cisco IOS Release 15.3(3)S3
- Closed Caveats in Cisco Release 15.3(3)S3
- Open Caveats in Cisco Release 15.3(3)S2
- Open Caveats in Cisco Release 15.3(3)S1a
- Open Caveats in Cisco Release 15.3(3)S
Open Caveats in Cisco IOS Release 15.3(3)S7
Incorrect thresholds set for GLC-FE-100BX-U and GLC-FE-100BX-D
xconnect not forwarding traffic with local SPAN on Attachment Circuit
Open Caveats in Cisco IOS Release 15.3(3)S6
"Voltage low‘” messages are seen on the Cisco ME-3800X-24FS.
Open Caveats in Cisco IOS Release 15.3(3)S5
Voltage low error messages were seen on ME-3800X-24FS.
MCAST stops for bridge-domain if the 1st host joined MCAST group leaves.
Open Caveats in Cisco IOS Release 15.3(3)S4
The PTP sideband processor reboots after every fourteenth day.
Source traffic is flooded on all port of the bridge domain in spite of snooping being enabled and Mrouter port learnt.
Open Caveats in Cisco IOS Release 15.3(3)S3
Symptom: The ECMP max-path can configure values from 1 to 16 when using the MetroIPAccess.
Conditions: This issue occurs when you boot with MetroIPAccess.
Workaround: Use values from 1 to 4
Symptoms: The Bridge-domain does not learn the MAC address but still forwards the traffic.
Conditions: This issue occurs when the cross-connect is configured on the switchport. On configuring the switchport for the first time, the configuration is rejected.
Workaround: Reapply the configuration.
Symptom: The snmpEngineBoots integer does not increment on the Cisco ME 3600X and ME3800X after the device is reloaded. Instead, it is reinitialized.
Conditions: This issue occurs only when the device is rebooted.
Workaround: Perform a warm restart of the device to increment the snmpEngineBoots integer.
Symptom: The VLAN loadbalacing does not work in REP segment 1.
Conditions: This issue occurs when there are more than one segment on the REP ring. This issue is observed only for Segment 1.
Workaround: Either shut down the other segments or perform a shut down followed by no shut on the other alternate port on segment 1.
Symptom: When the interface transceiver monitoring is disabled, the interface transceiver reappears after a reload.
Conditions: This issue occurs when the interface transceiver monitoring is disabled with the following commands:
After a reload, the commands reappear in the configuration.
Workaround: An EEM script can be applied to remove the commands from the configuration at bootup. This EEM script looks for the syslog message indicating the device was reloaded
If AAA command authorization is enabled, configure the event manager session command. Else, the EEM script fails.
Symptom: CE-to-CE ping fails when the TE Tunnel is "Up‘” on core.
Closed Caveats in Cisco Release 15.3(3)S3
Symptom: The license boot level tag ‘license boot level’ appears in show run command, and this
Conditions: This issue occurs when ‘license boot level’ license boot level is configured and is removed the next time after a reload.
Workaround: There is no workaround.
Symptom: The v6 multicast traffic is punted to host queue.
Conditions: This issue occurs on a L2 box with no multicast configuration. Workaround: There is no workaround.
Symptom: System crashes on configuring card type t1/e1 without t1/e1 license on the switch. Conditions: This issue occurs on configuring card type t1/e1 without t1/e1 license on box Workaround: Configure card type after having t1/e1 license on box.
Symptom: The Cisco ME3600X does not display time source information when running PTP.
Conditions: This issue occurs when the ME3600X does not display time source information when running PTP protocol. There is no functional issues noticed with PTP time synchronization. The time source field says “Unknown”.
Workaround: There is no workaround.
Symptom: Enabling PBR breaks the EIGRP multicast.
Conditions: This issue occurs when:
1. A Route map is created for any random IP address (not necessarily related to multicast/protocol IP).
2. The PBR is enabled on an interface with EIGRP peering.
Workaround: There is no workaround.
Symptom: Targeted LDP session between Customer Edge devices flap every 3 minutes. IGP
Interface LDP between CE devices run fine.
Conditions: This issue occurs when MPLS LDP packet is carried over L2VPN cloud with Control word OFF.
Workaround: Configure control word ON in L2VPN cloud.
Symptom: A few transceiver related alarms are not reported in the CLI after running the show facility-alarm status command. SNMP traps are not generated either for the same events.
Conditions: This issue occurs when the Cisco ME 3800X switch is running Cisco IOS Release 15.3(3)S1 and later.
Workaround: There is no workaround.
Symptom: ICMP destination unreachable message is not generated by the device.
Conditions: This issue occurs when a packet is dropped because there is no route in the routing table
Workaround: There is no workaround.
Symptom: The CISCO-EVC-MIB for ME3600X switch is not available with MetroIPAccess license.
Conditions: This issue is observed only with MetroIPAccess license.
Workaround: Upgrade the license to AdvancedMetroIP license.
Symptom: The Y1731 Performance Monitoring (PM) does not collect the statistics when the EVC tag X popped on ingress traffic before transiting MPLS PW and pushed EVC tag Y at egress traffic.
Conditions: The issue occurs when you send the CFM untagged packets over a pseuduowire.
Workaround: IP SLA works fine when the cos value matches with the cfm domain level.
Symptom: STP BPDU loss is observed when you want to tunnel or forward BPDUs across a EVC xconnect.
Conditions: The issue occurs when you want to tunnel or forward BPDUs across a EVC xconnect. This issue is specific to l2pt forward or tunnel and is not seen with other forms of Xconnect.
Workaround: There is no workaround.
Symptom: The Cisco ME 3600X switch does not respond.
Conditions: This issue occurs when a port-channel interface is configured in a bridge-domain where other interfaces are configured for ‘l2protocol tunnel lldp’ and LLDP packets are received.
Workaround: Remove ‘l2protocol tunnel lldp’ configuration from the interfaces in the bridge-domain.
Symptom: The targeted LDP is not working when TE tunnel is through the Cisco ME3600x switch. Conditions: This issue occurs under the following scenario:
Scenario: ME3600x-1-------ME3600x-2------ME3600x-3
This issue occurs when no mpls ip propagate-ttl forwarded command is configured on the midpoint router (ME3600x-2).
Workaround: Remove no mpls ip propagate-ttl forwarded configuration.
Symptom: Priority tagged traffic fails with type-5 VC.
Conditions: This issue occus when priority tagged (vlan id 0) is sent over vc type-5 with encapsulation default in EVC.
Workaround: Convert the vc type from type-5 to type-4.
BFD Fails to Establish Neighbor when running ISIS
Symptom: After an interface flap or a device reload, a BFD neighbor fails to establish. Protocols registered to BFD like ISIS also fails to establish.
Conditions: This issue only impacts the Cisco ME3600-cx platforms running ISIS with BFD. The issue is seen with the command show platform ho-fpga tx-buffer-table detail 258.
Workaround: Flapping the interface (shut/no shut), or removing and reapplying BFD can resolve the issue.
Symptom: CPU Hogs and Tracebacks seen on the Cisco ME 3600 switch.
Conditions: This issue occurs when multiple ports are part of the MSTP. and the ports are shut/no shut.
Workaround: There is no workaround.
Symptom: High CPU utilization is seen on the Cisco ME 3600 switch.
Conditions: This issue is seen when traffic flows through a service instance, which has MAC security configured.
Workaround: There is no workaround.
Symptom: Neither the SNMPwalk of EntityPhysical MIB nor the Show Inventory command reports the existence of the ME-FANTRAY-XL part.
Conditions: This issue occurs under normal conditions.
Workaround: There is no workaround.
Symptom: The control protocols flap intermittently when random traffic drops when SPAN is configured. Drops are also seen on interfaces that may or may not be a part of SPAN destination port.
Conditions: This issue occurs when SPAN is configured.
Workaround: Disable the SPAN configuration, especially the egress SPAN.
Symptom: Ping fails for around 35 seconds.
Conditions: This issue occurs due to a change in topology, and when Alternate port becomes Root port in RSTP mode.
Workaround: In the problem state, use the clear arp command to resume the traffic.
Symptom: Output packet drop is seen for the pause frames when, receive flow control is enabled on
Conditions: This issue occurs under the following conditions:
– when you get a pause frame from a connected device
– when you disable the flowcontrol on the Cisco ME3600x switch interface
Open Caveats in Cisco Release 15.3(3)S2
Symptoms: The entity MIB is not responding for the T1/E1 controller ports on the Cisco ME3600X-24CX switch.
Conditions: This issue occurs when you walk entitymib OID for the entPhysicalEntry where, none of the T1/E1 port details are returned.
Workaround: There is no workaround.
Symptom: Serial interface remains down on CEM interface.
Conditions: This issue occurs with the following configuration:
3. Configure 3 DS3 SAToP circuits
4. Do CEM interface shut (int cem 1/0/0 shut) at PE1
5. Do PW wire shut (int cem 1/0/0 cem 0 shut) at PE1
6. Do no shut CEM interface (int cem 1/0/0 no shut) at PE1
Serial interface of second PW remains down on 7600_CE1 and in up down on 7600_CE2
Workaround: Perform no shut pseudowire.
Symptom: Invalid EFP configuration error is observed upon trying to configure the double-tag EFP.
Conditions: This issue occurs when the interface having the media-type X configuration along with switchport trunk configurations is set as a default.
Workaround: Reconfigure the ‘media-type’ command on the interface that is set as a default.
Symptom: Memory leak is observed at NQ_PD_POL_ACTION.
Conditions: This incremental leak is observed on dynamic modification of marking with multiple set statements in the class.
Workaround: There is no workaround.
Symptom: CDP/LLDP neighbors are not forwarded with Layer 2 Protocol Tunneling (L2PT) and H-VPLS.
Conditions: This issue occurs with H-VPLS and SVI MPLS at core.
Workaround: Use the tunnelling option ‘l2pt tunnel’ instead of ‘l2pt forward’.
Symptom: Output packet drop is seen for the pause frames when, receive flow control is enabled on the 1gigabit interfaces.
Conditions: This issue occurs under the following conditions:
– when you get a pause frame from a connected device
– when you disable the flowcontrol on the Cisco ME3600x switch interface
Workaround: There is no workaround.
Symptom: Ping fails for approximately 35 seconds.
Conditions: This issue occurs when there is a change in the topology, and when the Alternate port becomes the Root port in RSTP mode.
Workaround: In the problem state, use the clear arp command to resume the traffic.
Symptom: Random packet drops are observed on the router when SPAN configuration is enabled. Drops are seen on the interfaces that may or may not be a part of SPAN destination port. This causes control protocols to flap intermittently.
Conditions: This issue is observed only when the SPAN configuration is enabled.
Workaround: Disable the SPAN configuration, especially the egress SPAN.
Symptom: The SNMP walk of EntityPhysical MIB or the Show Inventory command does not report the existence of the ME-FANTRAY-XL part.
Conditions: This issue occurs when ME-FANTRAY-XL is present in the chassis.
Workaround: There is no workaround.
Symptom: Traffic forwarding issue is observed with static MAC address on port-channel. Following are the configuration steps:
2. Configure static MAC address
After the reboot, traffic generated using the static MAC address are not egressing out of the port-channel.
Conditions: This issue occurs after the reboot egress forwarding stops.
Workaround: Reconfigure the static MAC address.
Symptom: Packets entering an VRF interface and the exiting the MPLS core interface are switched using the global table instead of the VRF table.
Conditions: This issue occurs when packets are punted due to egress mpls mtu fail command.
Workaround: Avoid the punt condition.
Symptom: Remark statement in IPv6 ACL stops all the IPv6 traffic when used in outbound condition.
Conditions: This issue occurs when IP routing is enabled, and the outbound IPv6 ACL has a remark statement as shown in the following example:
Workaround: Remove the remark from IPv6 ACL.
Symptom: The targeted LDP is not working when TE tunnel is through the Cisco ME3600x switch.
Conditions: This issue occurs under the following scenario:
Scenario: ME3600x-1-------ME3600x-2------ME3600x-3
This issue occurs when no mpls ip propagate-ttl forwarded command is configured on the midpoint router (ME3600x-2).
Workaround: Remove no mpls ip propagate-ttl forwarded configuration.
Symptom: The license boot level tag ‘license boot level’ appears in show run command, and this tag is not removable.
Conditions: This issue occurs when ‘license boot level’ license boot level is configured and is removed the next time after a reload.
Workaround: There is no workaround.
Symptom: High CPU utilization is observed due to non-QoS event processes after a reload.
Conditions: This issue occurs due to high scale of cross connects and QoS configuration.
Workaround: There is no workaround.
Symptom: Traceroute fails between CEs in the ethernet CFM domain for the UP and Down MEPs. LTM packets are not forwarded in the Cisco ME3600x and ME3800x switches.
Conditions: This issue occurs when the core devices are configured with static MIP.
Workaround: Do not configure static MIP.
Symptom: High CPU utilization is seen due to NL3U background process.
Conditions: This issue occurs when there are many entries in failed adjacency list.
Workaround: Perform clear ip route command.
Symptom: The STP BPDU loss is observed.
Conditions: This issue occurs when BPDUs are tunnelled/forwarded across an EVC cross connect. This issue is not seen with other forms of cross connects, but is specific to Layer 2 Protocol Tunneling (L2PT) forward/tunnel.
Workaround: There is no workaround.
Symptom: Entitymib is not responding for OC3/OC12/T3/E3 controller ports in the Cisco ME3600x-24CX-M switch.
Conditions: This issue occurs when the ME 3600X-24CX-M switch is running XE 15.3(3)S image.
Workaround: There is no workaround.
Symptom: The CISCO-EVC-MIB for ME3600X switch is not available with MetroIPAccess license.
Conditions: This issue is observed only with MetroIPAccess license.
Workaround: Upgrade the license to AdvancedMetroIP license.
Symptom: The IPv6 classification fails without IPv6 routing enabled for IPv6 switched packets.
Conditions: This issue occurs when IPv6 traffic is transported on an layer 2 network with no IPv6 routing.
Workaround: Enable ipv6 unicast-routing command for the classification to work fine.
Open Caveats in Cisco Release 15.3(3)S1a
Symptom: The MAC address used by a port-channel is invalid (all zeros).
Conditions: This issue is only seen if a session is created over a port-channel and then the switch is reloaded.
Workaround: Unconfiguring and reconfigure the scheduled session instead of restarting it.
Symptom: The policy is not rejected even after an invalid configuration.
Conditions: This issue occurs when the policy-map, configured along with a table-map in a logical class, is accepted.
Workaround: There is no workaround.
Symptom: Pseudowire traffic in the imposition direction fails.
Conditions: This issue occurs when a pseudowire traffic in the imposition direction fails.
Workaround: Reprovision the pseudowires.
Symptom: Invalid EFP configuration error is observed upon trying to configure the double-tag EFP.
Conditions: This issue occurs when the interface having the media-type X configuration along with switchport trunk configurations is set as a default.
Workaround: Reconfigure the ‘media-type’ command on the interface that is set as a default.
Symptom: Traceroute ethernet CFM fails on the Cisco ME3600x switch when the source is the second hop or beyond.
Conditions: This issue occurs under normal conditions.
Workaround: There is no workaround.
Symptom: Device Manager (DM) deletion of priority level command gives irrelevant error message. This is ‘QOS: Egress hierarchical policing supported only if child policer is on PQ’.
Conditions: This issue occurs when you configure priority level command on a PHB level class and attach it to an interface. This issue also occurs when you remove the priority command dynamically.
Workaround: Firstly, detach the interface. Secondly, reconfigure with priority removed. Lastly, reattach with the interface.
Symptom: Unable to ping the directly connected SVI interface from the PC when the MAC ACL is applied on the interface.
Conditions: The issue is triggered by the following conditions:
– performing a shut/no shut of interface connected to the PC
– removing the ACL, clearing the ARP, and reapplying the MAC ACL on the interface
Workaround: The ping starts working upon:
– clearing the ARP entry for the PC
– performing a shut/no shut on the VLAN interface
– removing the MAC ACL on the interface
Symptom: Ping fails for around 35 seconds.
Conditions: This issue occurs due to a change in topology, and when Alternate port becomes Root port in RSTP mode.
Workaround: In the problem state, use the clear arp command to resume the traffic.
Symptom: ARP resolution fails on routed pseudowire.
Conditions: This issue occurs only with a routed pseudowire, when an IP address is configured on a VLAN interface with the <i>xconnect</i> command. For example:
Open Caveats in Cisco Release 15.3(3)S
Symptom: Broadcast traffic flows over the Standby spoke virtual circuit (VC) and then gets punted.
Conditions: The traffic comes over the Core VCs (as they are ‘up’) and then flows over the Standby spoke VC from Hub to Spoke. In the spoke, the traffic received from the spoke VC gets punted to the CPU.
Workaround: There is no workaround.
Symptom: The Precision Timing Protocol (PTP) session does not start on removing the registered loopback interface for the master-slave setup.
Conditions: This issue occurs when the loopback interface (that is used in PTP configuration) is deleted and added back after the PTP is configured.
Workaround: First add the loopback interface and then reconfigure the PTP session.
Symptom: Internet Group Management Protocol (IGMP) snooping does not work with H-VPLS.
Conditions: This issue occurs when the interworking type for H-VPLS is VLAN. This results in IGMP snooping not to occur on H-VPLS.
Workaround: Change the interworking type to ethernet.
Symptom: Intermittent REP flaps are observed.
Conditions: This issue occurs when fast hellos (denoting the lowest timer that can be enabled on an REP) are enabled on a scaled node.
Workaround: Remove fast hellos.
Symptom: Dynamic deletion of the class-map filter does not take effect in a policy-map.
Conditions: This issue occurs when deleting the class-map filter operation fails. Dynamic deletion still classifies and applies the policer actions under the same class.
Workaround: Detach and reattach the policy-map.
Symptom: Measurements are not done when the active flex link port is down and the standby takes over as the active port on the Cisco ME3600X or Cisco ME3800X switch.
Conditions: This issue occurs on the Cisco ME3600X or Cisco ME3800X switch running Cisco IOS Release 15.2(4)S1, 15.3(1)S, or 15.3(2)S with:
– Flex link is configured on uplink ports
– Y1731 SLM is active on the flex link
Workaround: There is no workaround.
Symptom: Address Resolution Protocol (ARP) resolution fails when Dynamic ARP Inspection (DAI) is enabled.
Conditions: This issue occurs after enabling the DAI on another VLAN.
Workaround: Enable DAI on the relevant VLAN.
Symptom: The Cisco ME 3600X or ME 3800X switch takes 3.5 seconds for traffic convergence on the SVI.
Conditions: This issue is observed when the flex link is configured in non-VLB mode and the preemption mode is configured as forced.
Workaround: There is no workaround.
Symptom: When the Cisco ME 3600X or ME3800X switch is configured as the PTP Ordinary Slave or PTP Boundary Slave, the slave stays in the Acquiring State and never achieves the Phase Aligned State.
Conditions: This issue occurs under normal conditions.
Workaround: There is no workaround.
Symptom: Chunk leaks are observed with MPLS IP and IP FRR configured with some scale.
Conditions: This issue is observed when MPLS LDP flaps and clears.
Workaround: There is no workaround.
Symptom: The CFI bit of the inner-VLAN tag is altered.
Conditions: This issue occurs when the q-in-q EVC has rewritten the POP 1.
Workaround: There is no workaround.
Symptom: ALL IGP protocols generate LSA/LSP for 127.0.0.1/16 and 127.1.0.0/16 network.
Conditions: This issue occurs when the passive interface default is enabled under IGP protocol (OSPF, ISIS, EIGRP, or RIP).
Workaround: Remove the passive interface default from IGP.
Symptom: IPv6 6VPE traffic fails to flow.
Conditions: This issue occurs when you add the IPv4 address to an IPv6 VPN-enabled interface.
Workaround: Do not add an IPv4 address to an interface that does not have IPv4 VPN address family defined.
Symptom: Duplicate ARP request packet is seen over SVI Ethernet over MPLS (EoMPLS).
Conditions: This issue occurs when the ARP request (broadcast) packets are sent over the SVI EoMPL, and the other end receives duplicate packets.
Workaround: There is no workaround.
Symptom: 6VPE/VPNv4 traffic fails to egress Ethernet virtual circuit (EVC) with an explicit etype traffic.
Conditions: This issue occurs when the dual stack enabled bridge domain (BD) is attached to two different EVCs with an explicit etype traffic.
Workaround: There is no workaround.
Symptom: Traceback messages are displayed while adding and removing the MPLS LDP password.
Conditions: This issue is specific to IP FRR.
Workaround: There is no workaround.
Symptom: QoS-related CPU hog error messages are displayed.
Conditions: This issue occurs when you use the clear xconnect command.
Workaround: There is no workaround.
Symptom: the IS-IS neighbor ship goes down between the Cisco ME3600X nodes after a reload or a power cycle.
Conditions: This issue occurs after the router is reloaded or is power cycled.
Workaround: There are two workarounds to fix this issue. One is to use the clear isis * command, and the other is to configure “mpls ldp igp sync holddown 60000”.
Symptom: High CPU utilization is observed due to an NL3U background process, and this happens once all resources are exhausted.
Conditions: This issue occurs due to many failed adjacencies.
Symptom: Traffic loss is observed after switching to Protect path.
Conditions: This issue occurs with MPLS TP. The core traffic loss in the protect path is observed while bringing down the Active path where the interface is shut at mid point.
Workaround: Perform a no shut and a shut action on the mid-point interface.
Symptom: The policy is not rejected even after an invalid configuration.
Conditions: This issue occurs when the policy-map, configured along with a table-map in a logical class, is accepted.
Workaround: There is no workaround.
Symptom: The IPv6 ping fails even when the route is installed in the IPv6 route table.
Conditions: This issue occurs when an interface is configured with IPv6 address A along side another address B and removing the address A from the interface.
Workaround: There is no workaround.
Symptom: EVC filtering fails, as the wrong traffic egresses EVC.
Conditions: This issue occurs due to the dynamic modification of encapsulation with etype IPv6/IPv4 configured under service instance.
Workaround: Perform a shut and a no shut of BD or SVI interface.
Symptom: Traffic drops for a few prefixes.
Conditions: This issue occurs when layer 3 VPN with ECMP paths is configured, and when more than 10K prefixes are populated in the routing table.
Workaround: Remove the ECMP paths.
Symptom: Pseudowire traffic in the imposition direction fails.
Conditions: This issue occurs when a pseudowire traffic in the imposition direction fails.
Workaround: Re-provision the pseudowires.
Symptom: The Cisco ME3600X crashes if VRF configuration is removed from routed PW VLAN interface.
Conditions: This issue occurs when the routed PW VLAN interface is configured with the secondary IP address.
Workaround: There is no workaround as there is no support for secondary IP address on routed PW VLAN interface.
Symptom: Ping on IPv6 fails with VRF.
Conditions: This issue occurs while disabling and enabling the IPv6 under VLAN SVI.
Workaround: There is no workaround.
“Receive Power Threshold(++,+,-,--)” is not displayed by the show interface transceiver command.
Conditions: This issue occurs on the Cisco ME-3800X-24FS-M device running 15.3(2)S1.
Workaround: There is no workaround.
Symptom: After the TCAM limits are met, the freed TCAM is not allocated further.
Conditions: This issue occurs when the TCAM limit is beyond the allowed limit and then operates within TCAM limits.
Workaround: Following are the workarounds:
Symptom: High CPU utilization is observed when multicast IPv6 ICMP traffic enters L2VPN.
Conditions: This issue occurs when multicast IPV6 traffic is sent to an L2VPN.
Resolved Caveats
The following section provides information about resolved caveats:
- Resolved Caveats in Cisco IOS Release 15.3(3)S9
- Resolved Caveats in Cisco IOS Release 15.3(3)S7
- Resolved Caveats in Cisco Release 15.3(3)S6
- Resolved Caveats in Cisco Release 15.3(3)S5
- Resolved Caveats in Cisco Release 15.3(3)S4
- Resolved Caveats in Cisco Release 15.3(3)S2
- Resolved Caveats in Cisco Release 15.3(3)S1a
- Resolved Caveats in Cisco Release 15.3(3)S
Resolved Caveats in Cisco IOS Release 15.3(3)S9
This is a special release in Cisco IOS software that addresses the Cisco Product Security Incident Response Team (PSIRT) caveats.
Resolved Caveats in Cisco IOS Release 15.3(3)S7
DHCP relay does not remove Option82 in Offer forwarded to client
Resolved Caveats in Cisco Release 15.3(3)S5
Crash in MetroEthernet switch due to memory corruption.
OID to get the power supply temperature value on ME switches.
ME-3600-Class w/ CIR police showing incorrect cbQosQueueingCfgBandwidth.
ME3600 port going into half duplex.CSCur46230
ARP incomplete on routed Port-channel26.
ME3600x C-Tag incorrectly written to 0 for STP.
ARP incomplete on routed Port-channel26
Resolved Caveats in Cisco Release 15.3(3)S4
vlanTrunkPortDynamicState OID is not available when the interface is administratively up
snmpEngineBoots does not increase after reboot
VLAN load balancing in REP segment is broken.
OSPF packets are not forwarded.
Even though interface transceiver monitoring is disabled, it reappears after a reload.
CE to CE ping fails when the TE tunnel is administratively up.
REP interface may flap frequently and randomly.
HSRPv2 groups do not converge when configured on multiple SVI bridge domains.
Removing of service policy on interface with traffic cause device to freeze. CPUHOG messages are seen in the logs which may be followed by a crash due to watchdog timeout.
Flex link backup is administratively down when the VLB configuration is removed.
The VLAN interface state is administratively up even when the physical port state is down.
IPv6 and IPv4 VRF pings fail due to incorrect VRF label programmed in the hardware.
WANPHY flag j1 transmit command availability displays inconsistent behavior with running-config and start config commands
cbQosQueueingCfgBandwidth returns value of zero.
REP does not converge and MALLOC traceback messages are observed.
Extra bytes are added when the CFM frame is transmitted.
QinQ configuration does not process ARP requests.
MAC flaps happen in VPLS setup when standby AC power supply is configured.
System adds VLAN configuration automatically for each VLAN created on the device.
Resolved Caveats in Cisco Release 15.3(3)S2
Symptom: The policer on the Cisco ME3600x and ME3800x switches do not work after changing the rewrite type dynamically.
Conditions: This issue occurs when the policer has the table-map, and the rewrite type is changed dynamically.
Workaround: Attach and detach the policy.
Symptom: The MAC address used by a port-channel is invalid (all zeros).
Conditions: This issue is only seen if a session is created over a port-channel, and then the switch is reloaded.
Workaround: Unconfigure and reconfigure the scheduled session instead of restarting it.
Symptom: A high CPU utilization was observed due to BFD process.
Conditions: This issue occurs when the BFD flaps with the port-channel enabled.
Workaround: There is no workaround.
Symptom: QoS-related CPU hog error messages are displayed.
Conditions: This issue occurs when you use the clear xconnect command.
Workaround: There is no workaround.
Symptom: Pseudowire traffic in the imposition direction fails.
Conditions: This issue occurs when a pseudowire traffic in the imposition direction fails.
Workaround: Reprovision the pseudowires.
Symptom: The PTP sessions are not responding, and are in the ‘holdover’ state.
Conditions: This issue occurs when you configure the master and the slave with ToD format.
Workaround: Configure only the ToD format as cisco, after the PTP session is up with ToD format in NTP.
Symptom: The following messages are seen on bootup of The Cisco ME3600x and ME3800x switch.
*Jul 1 23:25:13:402: Table type:0, total entries:1024 *Jul 1 23:25:13:402: Table type:1, total entries:1024Conditions: This issue occurs on booting the Cisco ME3600x switch without any configurations.
Workaround: There is no workaround.
Symptom: The SNMP context cannot be configured on the Cisco ME3600X switch.
Conditions: This issue occurs when the switch is running Cisco IOS 15.2(4) release.
Workaround: There is no workaround.
Symptom: Remote ping fails in gigabit0 interface after the cross connect is configured.
Conditions: This issue occurs when the cross connect is configured on a management interface.
Workaround: There is no workaround.
Symptom: An error message is seen upon configuring storm-control.
Conditions: This issue occurs when you configure storm-control with EVCs on the interface.
Workaround: There is no workaround. This is a cosmetic issue and does not affect functionality in anyway.
Symptom: ARP resolution fails on routed pseudowire.
Conditions: This issue occurs only with a routed pseudowire, when an IP address is configured on a VLAN interface with the <i>xconnect</i> command. For example:
Symptom: The CEM interface counter is not incrementing.
Conditions: This issue occurs when the traffic is routed to CEM CIR.
Workaround: There is no workaround.
Symptom: Traceroute ethernet CFM fails on the Cisco ME3600x switch when the source is the second hop or beyond.
Conditions: This issue occurs under normal conditions.
Workaround: There is no workaround.
Symptom: Traceback messages are displayed when you configure Ethernet Loopback (ELB) on the service instance that does not have a bridge-domain configured.
Conditions: This issue occurs when you configure an EFP without configuring BD.
Workaround: Configure BD before enabling ELB.
Symptom: The VRF IPv4 ping fails upon removing the IPv6 address.
Conditions: This issue occurs when you remove the IPv6 address under the interface SVI that again is under VRF.
Workaround: There is no workaround.
Symptom: Unable to ping the directly connected SVI interface from the PC when the MAC ACL is applied on the interface.
Conditions: The issue is triggered by the following conditions:
– performing a shut/no shut of interface connected to the PC
– removing the ACL, clearing the ARP, and reapplying the MAC ACL on the interface
Workaround: The ping starts working upon:
– clearing the ARP entry for the PC
– performing a shut/no shut on the VLAN interface
– removing the MAC ACL on the interface
Symptom: The SPAN sessions are not working on dynamic modification of the session parameters.
Conditions: This issue occurs when you have multiple source ports, which changes SPAN direction.
Workaround: Remove SPAN session and reconfigure.
Symptom: Traceback messages are seen on switching of SyncE clock reference in the PTP OC master node.
Conditions: This issue occurs when SyncE is configured along with PTP.
Workaround: There is no workaround.
Symptom: QOS invalid policy-map is accepted on EFP.
Conditions: This issue occurs when EFP is not associated with the bridge-domain, or with the cross connect.
Workaround: There is no workaround.
Symptom: QoS egress policy-map is not attached on VRF interface. The egress classification does not work.
Conditions: This occurs only with a certain sequence of configuration. Following is an example:
1. Configure the switchport trunk interface with allowed VLAN.
2. Set this interface as default.
3. Configure the same interface as egress VRF interface.
4. Attach the egress policy-map.
The ‘QoS: out of tcam resources’ error message is displayed.
Workaround: There is no workaround.
Symptom: MAC flaps are seen on a shut service instance after a reboot.
When the Cisco ME600x or ME3800x switch is reloaded with an ethernet service instance in an administratively shutdown state, it reports MAC-flap messages between the administratively down service instance and the another port, if there is a loop with the service instance active.
Conditions: This issue occurs when the MAC-flaps on the shut EFP is seen only at bootup.
Workaround: Perform no shut and shut of the service instance.
Symptom: Traffic of Ethertype 0x8847 is not passing through cross connect on the Cisco ME-3600X-24CX-M switch.
Conditions: This issue occurs under normal conditions.
Workaround: There is no workaround.
Symptom: The .bin file that is extracted from .tar file does not have the right image name. And, when upgrading from 15.3(3)S to 15.3(3)S1 using the archive download-sw <TAR file> the router drops to rommon.
Conditions: This issue happens when the .bin file is extracted from .tar file (and/or) when extracted using the archive download-sw command.
Workaround: There is no workaround for the wrong image name. Switch dropping to rommon during the upgrade through archive download-sw command can be avoided by using archive download-sw or leave-old-sw <TAR file> command.
Symptom: In a REP setup, when you have a port in the FAILED state due to NO_NEIGHBOR, and in this state if the other port of the REP segment is brought up and moves to OPEN state, or if the other port is already in OPEN state is shutdown, the FAILED port is gets unblocked in the hardware. This results in a layer 2 loop.
Conditions: This issue is seen only with ports in FAILED state due to no neighbor condition.
Workaround: A shut/no-shut on the FAILED port moves it back to the blocking state.
Symptom: Layer 2 flow hits the layer 3 ACL entry.
Conditions: This issue occurs when the SVI interface is configured with no ip unreachables command.
Workaround: Following are the workarounds:
– Create a specific permit entry for the direction of ping (ICMP traffic) in the ACL.
– Remove no ip unreachables command configuration from the SVI.
Symptom: For CEM MPLS packets imposed with EXP=5 by default, the COS=5 is not set on the MPLS transport VLAN.
Conditions: This issue occurs when the SAToP/CESOP CEM pseudowires are initiated from a PE device and egresses out of the PE device into a layer 2 network, the MPLS transport VLANs COS value is expected to be 5 by default but it is not in this case.
Workaround: There is no workaround. Enter the following command in SD CLI after bootup:
Symptom: Egress classification is not working on ‘switchport trunk’ interface with and without allowed VLANs configured.
Conditions: This issue occurs when the egress QoS applied on the interface where ‘switchport trunk’ interface is configured.
Workaround: There is no workaround.
Symptom: MVPN traffic is not received properly by the receivers as the core replication does not work.
Conditions: This issue occurs when a lot of scale is present, or when the data MDT is used with scale for mVPN.
Resolved Caveats in Cisco Release 15.3(3)S1a
Symptom: The Connectionless Network Service (CLNS) maximum transmission unit (MTU) on an interface remains as a default, even when the interface MTU is changed. This causes the IS-IS process to drop the link state advertisements (LSAs) that exceed the CLNS MTU.
Conditions: This issue occurs on any Cisco ME3600 or ME3800 switch running 12.2(52)1EY where IS-IS adjacency has been established, and then the interface MTU is changed.
Workaround: Manually configure the CLNS MTU.
Symptom: Broadcast traffic flows over the standby spoke virtual circuit (VC) and then gets punted.
Conditions: The traffic comes over the core VCs (as they are ‘up’), and then flows over the standby spoke VC from hub to spoke. In the spoke, the traffic received from the spoke VC gets punted to the CPU.
Workaround: There is no workaround.
Symptom: The Cisco ME 3600X switch tags packets on the native VLAN of a dot1Q trunk when the packets are routed. Local generated packets goes untagged, and packets that have only been switched through the Cisco ME 3600X goes untagged. But, if the packet enters the Cisco ME 3600X on another VLAN and is routed to the destination VLAN, it gets tagged when leaving on a dot1Q trunk.
Conditions: This issue occurs under normal conditions.
Workaround: There is no workaround.
Symptom: Address Resolution Protocol (ARP) resolution fails when Dynamic ARP Inspection (DAI) is enabled.
Conditions: This issue occurs after enabling the DAI on another VLAN.
Workaround: Enable DAI on the relevant VLAN.
Symptom: When the Cisco ME 3600X or ME3800X switch is configured as the PTP ordinary slave or PTP boundary slave, the slave stays in the Acquiring state and never achieves the Phase Aligned state.
Conditions: This issue occurs under normal conditions.
Workaround: There is no workaround.
Symptom: Out-of-range error messages are seen along with traceback messages.
Conditions: This issue occurs when 64th Ethernet Flow Point (EFP) is associated with a bridge-domain that has Protocol Independent Multicast (PIM) enabled.
Workaround: Reload the switch without 64th EFP configuration.
Symptom: Error messages are seen on soaking the setup.
Conditions: This issue occurs when RLFA tunnel flaps over portchannel.
Workaround: There is no workaround.
Symptom: Duplicate ARP request packet is seen over SVI Ethernet over MPLS (EoMPLS).
Conditions: This issue occurs when the ARP request (broadcast) packets are sent over the SVI Ethernet over MPLS (EoMPL), and the other end receives duplicate packets.
Workaround: There is no workaround.
Symptom: Memory chunk leaks are observed.
Conditions: This issue occurs when PIM is disabled on an SVI.
Symptom: The Cisco ME3600X switch is not responding.
Conditions: This issue occurs when you copy the configuration from the TFTP server having card types oc3 and t1 configurations.
Workaround: Perform a power cycle of the switch.
Symptom: The switch port configured as backup port does not move to standby normally.
Conditions: This issue occurs when a port is configured as a switchport backup to other port.
Workaround: Shut and unshut the port. The backup port moves to standby normally.
Symptom: EVC filtering fails, as the wrong traffic egresses EVC.
Conditions: This issue occurs due to the dynamic modification of encapsulation with etype IPv6/IPv4 configured under service instance.
Workaround: Perform a shut and a no shut of BD or SVI interface.
Symptom: Unable to disable the MAC-learning for VLAN on Cisco ME3600X/ME3800X switches.
Conditions: This issue occurs when a VLAN is not part of the EVC or BD.
Workaround: There is no workaround.
Symptom: After the TCAM limits are met, the freed TCAM is not allocated further.
Conditions: This issue occurs when the TCAM limit is beyond the allowed limit and then operates within TCAM limits.
Workaround: Following are the workarounds:
Symptom: SyncE is not working.
Conditions: This issue occurs under the following conditions when you:
– Configure network-clock synchronization automatic on router A and router B
– Configure synchronous mode on gigabitEthernet interface connected between router A and router B
– Configure network-clock input-source 1 interface gigabitEthernet 0/7 on router B
– Configure the gigabitEthernet 0/7, and is not taken as internal clock
Workaround: There is no workaround.
Symptom: On the Cisco ME3600X or 3800X switch, RIP packets are not punted to CPU from routing TCAM. This is because the CPU is not attached to the TCAM. This results in all the control plane using multicast RR entry to go down.
Conditions: Under certain rare conditions, the issue can occur when:
– The last EVC on the Bridge-domain is removed
– The IP address on the VLAN corresponding to this BD is removed
– The VLAN/BD is recreated with the new IP address and a new EVC.
Workaround: There is no workaround.
Symptom: The BFD and IS-IS flaps on the systems that are subjected to an ARP storm, either deliberate, or intermittent network topology loops.
Conditions: This issue occurs under exceptional cases, where there is a misbehaving CPE, or a broadcast loop that may be created.
Workaround: There is no workaround.
Symptom: The BERT is not getting synced on different paths other than 1(one).
Conditions: This issue occurs when the BERT is run on paths other than 1.
Workaround: There is no workaround.
Symptom: Packet drop is seen in the protected label-switched path (LSP) after deleting the working LSP.
Conditions: This issue occurs after deleting the working LSP.
Workaround: There is no workaround.
Workaround: There is no workaround.
Symptom: The CRC errors are observed on sending traffic greater than 1G on 10G WAN links. Packet drops are also seen.
Conditions: This issue occurs when 10G links are configured on WAN mode, and traffic rate is more than 1 Gbps.
Workaround: There is no workaround.
Symptom: Memory leaks are observed during bootup.
Conditions: This issue occurs when the Cisco ME 3600X-24CX switch operates as a U-PE.
Workaround: There is no workaround.
Symptom: Multicast traffic drops as is hits the signal-to-noise ratio on the first hop node and the same is not forwarded.
Conditions: This issue occurs after running triggers like interfaces in core, access links shut/no shut, clear ip mroute command, and so on.
Workaround: There is no workaround.
Symptom: When two GigabitEthernet interfaces are provisioned with network-clock for clock recovery (as primary and secondary), the network-clock switches to internal clock rather than selecting the secondary interface upon shutdown (or link down) of the primary interface.
Conditions: This issue occurs when the primary and secondary interfaces belong to a same group of four ports. For example, the interfaces GigabitEthernet 0/1 to GigabitEthernet 0/4 are grouped together, similarly GigabitEthernet 0/5 to GigabitEthernet 0/8, GigabitEthernet 0/9 to GigabitEthernet 0/12, and so on are to be grouped together.
Workaround: Select the primary and secondary GigabitEthernet interface for clock recovery from different group of four ports.
Symptom: Blocking (BLK) or Alternate (Altn) state trunk link starts forwarding traffic causing a loop for EIGRP hellos.
Conditions: This issue occurs while adding or deleting a VLAN from an allowed VLAN list on a trunk link that is in BLK state.
Workaround: Perform a shut/no shut of the trunk links.
Symptom: Wrong DMM values in Y.1731 DMM.
Conditions: This issue occurs when the DMM session is between the Cisco ASR903 router and the Cisco ME3800X switch. The CFM is on the cross connect between these devices. In the core, the MPLS TE tunnel is the path between the Cisco ASR903 router and the Cisco ME3800X switch.
Only on receiving end (Rx), the timestamp values remain zero.Workaround: Remove MPLS TE in core and have path over Layer 3 main interface.
Symptom: A node does not respond when peering network is not stable.
Conditions: This issue occurs when frequent interface flaps on peer node, or reload of peering node. This can happen if node is connected to microwave links.
Workaround: There is no workaround.
Symptom: After removing the cross connection from an EFP, the EFP still points to the same VSI ID.This causes the traffic to flow when it is suppose to drop.
Conditions: This issue occurs after removing the cross connect from an EFP.
Workaround: Remove the entire service instance configuration.
Symptom: High CPU utilization with Small form-factor pluggable transceiver SFF-8472.
Conditions: This issue occurs when GLC-LH-SMD is used.
Workaround: Use an SFP without DOM or disable DOM monitor.
Symptom: The message "SD read failed!" is seen on IOS bootup.
Conditions: This issue occurs when IOS is booted with SD card (2G 16-3932-01 (C-Temp)) plugged in.
Workaround: Use 2GB: 16-3795-01 (I-Temp) SD card.
Symptom: When multiple priority configurations are on the policy-map, an error message is seen that blocks the configuration:
Conditions: This issue occurs when there is a need to set multiple priority on the same policy-map. This is true only for priority levels and not for setting priority.
Workaround: There is no workaround.
Symptom: Chunk memory leak is observed at eaid_hw_t.
Conditions: This issue occurs when Layer 2 VPN core or AC flaps with multicast traffic.
Workaround: There is no workaround.
Symptom: The BGP or LDP sessions are not shown up with VRF enabled interfaces.
Conditions: The issue is caused by CSCuj07571 commit which is not present in any released image.
Workaround: There is no workaround.
Symptom: The .bin file that is extracted from .tar file does not have the right image name. And, when upgrading from 15.3(3)S to 15.3(3)S1 using the archive download-sw <TAR file> the router drops to rommon.
Conditions: This issue happens when the .bin file is extracted from .tar file (and/or) when extracted using the archive download-sw command.
Workaround: There is no workaround for the wrong image name. Switch dropping to rommon during the upgrade through archive download-sw command can be avoided by using archive download-sw or leave-old-sw <TAR file> command.
Resolved Caveats in Cisco Release 15.3(3)S
Symptom: REP reconverges in a 1:1 (one active and one standby) scenario, when the hot standby converges to active, and the port-channel does not come down.
Conditions: This issue occurs when the hot standby converges to active, and the port-channel does not come down.
Workaround: There is no workaround.
Symptom: Traceback messages are seen with unicast traffic flood control.
Conditions: This issue occurs after sending the unicast traffic to the next-hop.
Workaround: There is no workaround.
Symptom: Ethernet service instances go down after defaulting the port channel and applying back the configurations on it.
Conditions: This issue occurs under normal conditions.
Workaround: Perform a shut and no shut on the port channel to bring the EFPs up, or configure the port channel by using CLI commands instead of copy and pasting the configurations.
Symptom: DWDM SFP+ are supported on the Cisco ME3600X switches, but it goes unrecognized when it is inserted into the switch. The switch is not recognizing the DWDM in SNMP ENTITY-MIB.
Conditions: This issue occurs when a DWDM SFP+ is present on the ME3600X switch.
Workaround: There is no workaround.
Symptom: Serial interface on HDLC does not support a customized MTU.
Conditions: This issue occurs when configuring an MTU on the serial interface.
Workaround: There is no workaround.
Symptom: Serial interface is up despite CRC mismatch.
Conditions: This issue occurs when configuring different CRC values on both the sides of the serial interface.
Workaround: There is no workaround.
Symptom: Observing CPU hog at IP RIB Update after multiple flaps of IGP and MPLS TE tunnels.
Conditions: This issue occurs when multiple MPLS enabled interface flaps resulting in IP RIB update crash.
Workaround: There is no workaround.
Symptom: Traceback messages are seen when TP tunnel interface is shut down while the working path is down.
Conditions: This issue occurs under normal conditions.
Workaround: There is no workaround.
Symptom: Traceback messages are seen during the NPM main process when the Cisco ME3600X/ME3800X switch reloads. These tracebacks are seen only with scaled EVC configurations and EVCs are part of the G8032 ring.
Conditions: This issue occurs when the switch reloads.
Workaround: There is no workaround.
Symptom: Internal port is made editable and the POST status is displayed on rebooting Cisco ME3600X/ME3800X switch.
Conditions: This issue is seen in releases 15.3 and later.
Workaround: There is no workaround.
Symptom: The value of the output packet counter is twice as that of the input with EVC local connect.
Conditions: This issue occurs after changing the encapsulation on EVC.
Workaround: There is no workaround.
Symptom: Ping fails over routed PW.
Conditions: This issue occurs with SVI based MPLS uplink.
Workaround: Disable MAC learning.
Symptom: Deleting and adding the user-defined class-map results in assert messages. This happens with scale EVC cross connect configuration when the same policy-map used for EVC cross connect configuration is applied on all the other EVCs.
Conditions: This issue occurs when adding and deleting the user-defined class-map.
Workaround: Remove the policy-map and make a dynamic modification on the policy-map. Or, there has to be a delay in the removal and addition of the class-map.
Symptom: This issue occurs if the Cisco ME3800X and ME7600 are working as PE devices and if they have multiple core connections between them.
If one of the core interface between ME3800X and 7600 and the attachment circuit (AC) is bounced simultaneously on ME3800X, the VPLS status on ME3800X shows 'up' but it shows 'down' on 7600.Conditions: This issue occurs when there are multiple connection between the core devices.
Workaround: Bounce the attachment circuit (AC) on the ME7600.
Symptom: The following DWDM SFPs do not work on a Gigabit Ethernet or 10Gigabit Ethernet port on the Cisco ME3600X switch:
When any one of the DWDM SFPs mentioned above is inserted into a GE or 10GE port on the Cisco ME3600X or ME3800 switch, the switch logs the following error messages:
Conditions: This issue occurs under normal conditions.
Workaround: There is no workaround.
Symptom: Traffic drops for a few VPLS VCS when it is configured with ECMP links.
Conditions: This issue occurs when one of the ECMP path is shut when more than 200 VPLS VCS are configured.
Workaround: There is no workaround.
Symptom: When MAC security is configured on an EFP port, the traffic received on this EFP port and designated to one MAC address learned on the same EFP is bounced back on the EFP.
Conditions: This issue occurs when traffic bounces back during one of the following conditions:
– Split-horizon group is configured on the EFP
– EFP is replaced by a legacy layer 2 trunk.
Workaround: There is no workaround.
Symptom: Error messages are displayed on the default interface that has QoS policy and the double-tag EVC.
Conditions: This issue occurs when the default interface is configured with double -tag EFP and the QoS policy.
Workaround: There is no workaround.
Symptom: Service policy in EVC cannot be reattached after changing pop to push rewrite operations.
Conditions: This issue occurs when a service-policy is configured on EVC. This issue also occurs when rewrite ingress tag is changed, and configured on the service-policy on EVC.
Workaround: Before changing the rewrite ingress tag, remove the service-policy on EVC.
Symptom: Following a crash, a CPU hog is observed.
Conditions: This issue occurs when the peer device flaps or when there is network instability.
Workaround: This issue does not occur on a stable network.
Symptom: The Cisco ME3600X/ME3800X switch does not respond due to crash.
Conditions: This issue occurs when having CFM over cross connect with PC in the core and run Y.1731 DMM on it.
Workaround: There is no workaround.
Symptom: Traceback messages are seen @adjmgr_l3_delete_nh.
Conditions: This issue occurs when configuring VFI cross connect and enabling PIM under SVI. Traceback message is observed on removing IP address under SVI.
Workaround: There is no workaround.
Symptom: File copy from SD flash to SD flash stops working properly.
Conditions: This issue occurs when the destination file size is less than the file size of the source file after coping the first file. This makes the SD flash data unusable.
Workaround: Format the directory sdflash:/.
Symptom: End-to-end layer 3 traffic is affected if the host queue (CPU queue 2) increments continuously at high rates (2000 packets/s and above).
Conditions: This issue occurs when end-to-end layer 3 traffic is affected if the host queue (CPU queue 2) increments continuously at high rates (2000 packets/s and above).
Workaround: There is no workaround.
Symptom: CoS inner value gets changed on marking with CoS in egress on q-in-q service instance without rewrite.
Conditions: This issue occurs when q-in-q service instance works without a rewrite operation.
Workaround: There is no workaround.
Symptom: When an SFP is plugged into a slot, which previously had an SFP+ inserted, the show interface status displays the SFP status as “Not present”.
Conditions: This issue occurs when an SFP is plugged into SFP+ port, which previously had an SFP+ port.
Workaround: Use the show controller ethernet controller Te 0/x command to find the SFP type of SFP plugged into SFP+ slot.
Symptom: The MAC flaps in the network happens on reloading the device.
Conditions: The MAC flaps occur, because the multicast BPDUs are being sent back into the VPLS core after they reach the destination. This behavior causes a MAC to flap on every device that are on the path through which the BPDU traverses.
Workaround: Apply split horizon at the bridge-domain where the MAC flaps occurs.
Symptom: Unable to set up maximum MTU when the port-channel is deleted and created again.
Conditions: This issue occurs when deleting the port-channel and re-creating it.
Workaround: Reload the device.
Symptom: Traceback messages are observed when the ip_macaddrs_update command is used.
Conditions: This issue is observed during ethernet loopback with QoS script run.
Workaround: There is no workaround.
Symptom: The Cisco ME3600X/ME3800X switch crashes as the network link flaps.
Conditions: This issue occurs when the link flaps.
Workaround: There is no workaround.
Symptom: Unexpected huge counter values are observed on the gigabit 0/1 interface and on the port-channel soon after creating any port-channel.
Conditions: This issue occurs when creating a port-channel using the gigabit 0/1.
Workaround: There is no workaround.
Symptom: OSPF on R-PW is bound at Exstart/Exchange state.
Conditions: The issue occurs when you configure a there is CFM on the device; OSPF on R-PW does not get to the full state.
Workaround: Remove "ethernet cfm global" configuration to bring up OSPF on R-PW.
Symptom: The IP SLA does not show any statistics, and the database does not get populated.
Conditions: This issue occurs when the core interface is a switch port trunk.
Workaround: There is no workaround.
Symptom: PTP slave is not able to start a session with the master.
Conditions: PTP session is not started if slave's and master's loopback IP addresses are in the same subnet.
– Master loopback address: <ipaddress>
– Slave loopback address: <ipaddress>
Workaround: Configure slave's and master's loopback IP addresses in different subnets.
Symptom: User-defined classes in the policy-map that are applied on the EVC with rewrite push is not supported. This configuration gets accepted in certain conditions.
Conditions: This issue occurs when the QoS policy is applied first to the EFP, and then the bridge domain configuration is applied.
Workaround: There is no workaround.
Symptom: CPU errors are seen with (*, G/M) entries on ACL.
Conditions: This issue occurs on Cisco ME3600X-24CX-M switches operating in Mode 3 or
Mode 4.Workaround: Operate the Cisco ME3600X-24CX-M switch in Mode 2.
Symptom: Old image present in the board is neither deleted nor displayed while performing archive download to upgrade or downgrade a software version.
Conditions: This issue occurs during “archive download”.
Workaround: There is no workaround.
Symptom: Configuring the same level of priority queues more than once does not get rejected.
Conditions: This issue occurs under normal conditions.
Workaround: There is no workaround.
Symptom: Classification based on the precedence/DSCP egress policy does not work as expected.
Conditions: This issue occurs in layer 2 VPN scenarios with the following configurations:
– DSCP/precedence based policy on egress access EVC of SVI based EoMPLS
– CoS based policy on egress access EVC cross connect
Workaround: There is no workaround.
Symptom: The IP SLA becomes unstable in a single link scenario when CFM is working under EVC BD configuration on CE6 and CE7 uplink port.
Conditions: This issue occurs when the flex link is configured on uplink ports, and when CFM remote MEPs do not come up.
Workaround: Perform a manual switchover of flex link ports until platform forwarding vlist is not empty. This causes CC packets to be sent correctly, and remote MEPs come up.
Symptom: Platform assert failures and displays traceback messages.
Conditions: This issue occurs on attaching invalid shaper to EFP and the interface.
Workaround: There is no workaround.
Symptom: Port based EoMPLS VC goes down after flapping the interface. The following steps recreate the issue:
5. Configure port-based EoMPLS.
6. Shut the access interfaces. Other side access port will be admin down hence the VC is down. This is expected as per protocol.
7. Now shut the other access interface.
8. Do “no shut” on both the access interfaces, now we expect the VC to come up as per protocol but it does not come up.
Conditions: The issue is observed on port-based EoMPLS with remote link failure notification feature. It affects the releases from Cisco IOS Release 15.3 onwards.
Workaround: Under the cross connect configuration use no remote link failure notification command to bring up the VC. Then, reconfigure the remote link failure notification feature.
Conditions: This issue is observed during the upgrade from 3.7.2 to 3.8.1.
Workaround: There is no workaround.
Symptom: Classification does not work properly.
Conditions: This issue is observed only if there are classes based on ACL match and normal DSCP match. Only the ACL class will classify properly, and other classes do not work.
Workaround: There is no workaround.
Symptom: Link OAM does not work without the layer 2 protocol peer configuration on an untagged EVC.
Conditions: This is a regression caused by CSCud63960 commit for the link OAM flow.
Workaround: There is no workaround.
Symptom: A CPU hog is seen at nile_mgr_bdomain_get_efp_count and is followed by a crash.
Conditions: This symptom occurs on booting the router with some tunnel configurations.
Workaround: There is no workaround.
Symptom: Layer 3 traffic fails.
Conditions: This issue occurs when configuring 4n+1th EFP in a bridge-domain where n=1,2, 3... an so on.
Workaround: Clear the MAC-address table.
Symptom: Drop rate is not displayed with ingress policing policy.
Conditions: This issue occurs when an ingress policer with exceed action has drop rate displayed as zero(0).
Workaround: Drop rate can be measured from the conformed and exceeded BPS.
Symptom: ARP does not pass for tunneled VLANs.
Conditions: This issue occurs when an IP ARP inspection is enabled on the switch where dot1q tunnel is configured.
Workaround: Remove ARP inspection.
Symptom: Use the no ip icmp redirect command globally does not result in ICMP redirects to stop.
Conditions: This issue occurs under normal conditions.
Workaround: There is no workaround.
Symptom: On a the Cisco ME 3600X or ME 3800X, OSPF sessions flap or go down. Flaps are seen when OSPF is configured on both the interface VLAN and routed ports.
Conditions: This issue is observed when high rate (18,000 PPS) PIM packets are sent. This occurs with any packets going on the multicast queue 7; for example: IGMP joins and, queries.
Workaround: If the high rate PIM or multicast packets are coming from an interface where the multicast traffic can be controlled, configuring COPP will serve as workaround, otherwise there is no workaround.
Symptom: Traffic drops for a few VPLS BGP Signalling VCs.
Conditions: This issue occurs while using the clear ip ospf proc command.
Workaround: Perform a shut and a no shut action on all the core interfaces.
Symptom: Connections to the Cisco ME3600X or ME3800X through the console, telnet or SSH is very slow with normal CPU utilization. Pings to IP addresses display intermittent packet loss.
Conditions: This issue occurs when there are large amounts of broadcast or flooded unicast traffic.
Workaround: There is no workaround.
Symptom: Cisco ME3600X switch displays wrong optical alarm threshold values for SFP-GE-Z.
Conditions: This issue occurs with IOS 15.3(1)S and SFP-GE-Z.
Workaround: There is no workaround.
Symptom: Pinging fails upon a path switchover from EVC to routed PW.
Conditions: This issue occurs when the MAC address moves from the EVC to a routed PW, and the ping fails.
Workaround: Clear MAC address table.
Symptom: CFM trace route fails.
Conditions: This issue occurs in CFM with VPLS in the core. Configure the Up MEP on BD which has the VFI terminated.
Workaround: There is no workaround. The issue is specific to VPLS in the core and Up MEP on the same BD.
Symptom: Switch crashes upon configuring VLAN first and reconfiguring the policy-map later.
Conditions: This issue occurs when the QoS policy-map applied on the switchport interface is reconfigured or modified after the addition of a VLAN globally.
Workaround: Configure the VLAN and wait until the CPU utilization is normal, then modify the policy-map.
Symptom: Packet loss is observed over pseudowire and high CPU.
Conditions: This issue is observed when IPv6 site-local multicast MAC traffic is sent over SVI EoMPLS. The traffic is looped between the PE of the EoMPLS.
Workaround: There is no workaround.
Symptom: Memory leaks are observed with “nile_add_rep_entry”.
Conditions: This issue occurs when the rep admin vlan command is configured at the global level.
Workaround: There is no workaround.
Symptom: Pings to Switch Virtual Interface (SVI) are successful when the MAC address table is cleared, but fails later.
Conditions: This issue occurs while pinging the SVI interface.
Workaround: There is no workaround.
Symptom: Asserts are displayed when the interface with the configured cross connect flaps or when the cross connect comes up.
Conditions: This issue occurs under normal conditions.
Workaround: There is no workaround.
Symptom: ICMP replies get duplicated on the Cisco ME3600X switch.
Conditions: This issue occurs when creating service instances on port-channel interfaces on ME3600X in a topology is as shown below:
IXIA---7600(Te2/1-2)====(PO1)====(Te0/1-2)ME3600(Gi0/4-5)====(PO2)====(Gi1/0/25-26)SW3750------IXIA
Workaround: There is no workaround.
Symptom: CPU hog error message is seen at nile_mgr_bdomain_get_efp_count and then followed by a crash.
Conditions: This issue is observed on booting the switch with scaled mVPN configurations.
Workaround: There is no workaround.
Symptom: CFM transparency fails over EFP local connect.
Conditions: This issue occurs when the CFM is enabled globally.
Workaround: There is no workaround.
Symptom: The switch crashes upon defaulting the backup switch interface configuration immediately after performing a shut/no shut operation on it.
Conditions: This issue occurs after an administrative shut/no shut operation. Wait for some time (till the port comes up) before returning to the default interface configuration.
Workaround: There is no workaround.
Symptom: The VRF dual stack is not working.
Conditions: This issue occurs when there are 15 VRFs with dual stack and 15 VRFs for IPv4 only.
Workaround: Maintain all VRFs with dual stack.
Symptom: The ingress classification based on the numbered IP ACL does not work correctly.
Conditions: This issue occurs if an ingress QoS policy contains match based on numbered ACL, classification does not work, and the actions do not take effect.
Workaround: Use a named ACL instead of numbered ACL for the match.
Symptom: The Cisco ME 3600X/3800X switches may reload when a BGP session flaps.
Conditions: This issue occurs only if there are more than one BGP neighbors configured on the switch.
Workaround: There is no workaround.
Symptom: The archive download command fails which is being used for software upgradation.
Conditions: This issue is observed only on the Cisco ME3600X-24CX-M switch.
Workaround: There is no workaround.
Symptom: The peer-end interface shows the link status as Up.
Conditions: This issue occurs when the interface on the Cisco ME3600X-24CX-M switch is administratively down and the switch is reloaded.
Workaround: Perform a shut and a no shut on the Cisco ME3600X-24CX-M interface.
Symptom: The classification based on QoS group egress policy does not work correctly.
Conditions: This issue occurs with layer 3 VPN configuration, the core interface packets should be classified based on EXP and marked with QoS-group. On the egress interface packets should be classified based on QoS group on the service instance.
Workaround: There is no workaround
Symptom: Delay Measurement Reply (DMR), Session Level Redundancy (SLR), Linktrace Message (LTM), Linktrace Reply (LTR), Loopback Message (LBM), and Loopback Reply (LBR) are single tagged when the core-facing interface is a switch port.
Conditions: This issue is observed when the UP MEP and core facing interface are configured as switch port trunk.
Workaround: There is no workaround.
Symptom: Errors are displayed on flapping L2VP Pseudowire.
Conditions: This issue occurs when there are L2VPN services on the device.
Workaround: There is no workaround.
Symptom: Memory leak is observed with ELB with QoS configuration.
Conditions: This issue occurs when the Cisco ME3600X-24CX-M switch is loaded with ELB and QoS configuration.
Workaround: There is no workaround.
Symptom: Error occurs on connection manager when interfaces with the local connect are defaulted.
Conditions: This issue occurs when the interface has EFP configured with local connect.
Workaround: There is no workaround.
Symptom: Traffic from the routed VPLS does not trigger ARP.
Conditions: This issue is observed under the following conditions:
– A routed VPLS network that has multiple CE routers is connected to a PE router.
– A local CE router is connected to the PE router through an EVC.
– A ping is sent from a local CE router to the remote CE router.
Workaround: Ping the first interface VLAN of the EVC to resolve the ARP.
Symptom: Console messages display status of internal ports.
Conditions: This issue occurs when reloading the device.
Workaround: There is no workaround.
Symptom: Trust port configuration is invalid on DHCP snooping over EVC. The DHCP discover request cannot be snooped correctly and will be dropped on EVC in an layer 2 flat environment.
Conditions: This issue occurs when configuring a new interface regarding BD of trust port, or when reloading after saving the configuration.
Workaround: There is no workaround.
Symptom: Ping drops on controller interface GigabitEthernet0. Receive stall in the ETSEC driver.
Conditions: This issue occurs under normal conditions.
Workaround: Perform a shut and a no shut on the GigabitEthernet0 interface.
Symptom: VRF ping does not work after deleting and recreating the VRF.
Conditions: This issue occurs when VRF is deleted and added back multiple times.
Workaround: There is no workaround.
Symptom: Traffic is not forwarded from the service instance.
Conditions: This issue occurs after a reload, but can also be seen with other negative triggers, such as clear ip mro *, shut no shut of port, and so on.
Workaround: Perform a shut and a no shut on the link.
Symptom: Node crashes when Layer 3 and Layer 2 services are reloaded. This terminates the peer end nodes.
Conditions: This issue occurs when a reload or flap of MPLS interface is enabled at the peer end.
Workaround: There is no workaround.
Symptom: Traffic is observed after applying the policy-map under 10G interface on the Cisco ME 3600X switch.
Conditions: This issue occurs when the interface is configured as Layer 3 with VRF under the 10G interface, without VRF or Layer 2.
Workaround: Perform a shut and a no shut after adding or removing policy map.
Symptom: snmpwalk on ifDescr and ifAlias shows the hidden interfaces in Cisco ME3600X-24CX-M.
Conditions: This issue occurs when there is snmpwalk on ifDescr.
Workaround: There is no workaround.
Symptom: LDP sessions flaps, and they are seen recovered when the PIM joins are stopped.
Conditions: This issue occurs when high rate(18,000 PPS) PIM packets are sent.
Workaround: There is no workaround.
Symptom: Cisco ME 3600X switch fails to use archive upload-sw command while uploading an image. This failure leads to unsuccessful creation of the temporary directory ramfs:update.
Conditions: This issue occurs under normal conditions.
Workaround: Use copy flash: tftp: instead of archive upload-sw command.
Symptom: Multicast traffic stops getting forwarded when a MROUTER (SVI) VLAN is shut. The incoming interface and outgoing interface are on same VLAN.
Conditions: This issue occurs when IGMP snooping has to be enabled, and SVI has to be acting as a MROUTER. When this MROUTER is shut, traffic should still flow, because layer 2 connectivity is up. But an outage is observed.
Workaround: Enable and disable IGMP snooping.
Symptom: Traffic is not forwarded from the Cisco ME3600X switch, when routed over the EFP that has double VLAN tag.
Conditions: This issue occurs when the Cisco IOS image is upgraded to Cisco IOS Release 15.3(1)S1a.
Workaround: Upgrade to Cisco IOS release 15.3(1)S2.
Symptom: Multicast traffic is not forwarded.
Conditions: This issue is observed in the following scenario:
Considering the above dual-home scenario, this issue is observed when the following conditions are true:
– when one of the dual-home links to the receiver (N2-Switch) is down.
– when EVC is configured between N1 and N2.
Workaround: Following are the workarounds:
– Use trunks between N1 and N2.
Symptom: MAC flaps are observed with Resilient Ethernet Protocol (REP).
Conditions: This issue occurs when there is traffic.
Workaround: There is no workaround.
Symptom: Static MAC address configuration does not drop traffic from the specific MAC.
Conditions: This issue occurs under normal conditions.
Workaround: There is no workaround.
Symptom: The command to get PHY information without interface option is rejected.
Conditions: This issue occurs when the PHY information CLI command is used without interface option.
Workaround: Specify the interface option to get the PHY information.
Symptom: Netflow and BFD configurations on the same interface (Layer 3 port, portchannel, and so on) make BFD unresponsive and BFD stays in the “AdminDown” state.
Conditions: This issue occurs if there is an interface (individual Layer 3 port, port-channel) configured with Netflow and BFD. This issue can occur on the Cisco ME3600X and ME3800X switches.
Workaround: Disable the IP flow ingress command from the interface configured with BFD. Netflow is not supported on the Cisco ME3600X and ME3800X platforms.
Symptom: On a system with high Qos TCAM usage, a trunk port flap will cause the CPU to spike to 100%. The more the trunk port is flapped the longer the duration the CPU spikes to 100%.
Conditions: This issue occurs when a system has a high Qos TCAM usage and when the interface is flapped.
Workaround: There is no workaround.
Symptom: When an ACL with the range command is used for QoS, tracebacks are observed.
Conditions: This issue is observed when the policy-map is applied to the interface. The specified range is 1024 to 65535.
Workaround: Apply the same ACL, but modify the range to 1024 to 65534.
Symptom: Ingress drop counters are not incremented when the show policy-map command is issued.
Conditions: This issue occurs when ELB is enabled on an EFP port.
Workaround: There is no workaround.
Symptom: Traffic fails when the case is switched from Unicast to Flood.
Conditions: This issue occurs when traffic switches from Unicast over EFP to Flood over EFP and PW does not happen.
Workaround: There is no workaround.
Symptom: OSPF/LDP packets are not punted to the CPU despite having IP enabled on the VLAN.
Conditions: VLAN up trigger to multicast is missed. This is a very rare case. This happens in an environment where VLAN up and down is common.
Workaround: Perform a shut and a no shut on the VLAN.
Symptom: For circuit emulation psuedowires (CEM PW), the value of EXP is set to 5 by default. However, the value of COS, which should also be set to 5 is not being set.
Conditions: This issue occurs when all CEM traffic flows into the core from imposition the PW box without any QoS policies.
Workaround: There is no workaround.
Symptom: Frame loss is observed when configuring an existing class-map to a new service instance. The frame loss is observed on the existing service instance.
Conditions: This issue is observed when the class-map has the same value as the SET value (such as SET DSCP and SET QOS group).
Workaround: There is no workaround.
Symptom: After the reload of Cisco ME3600X switch, a part of the configuration disappears. As a result, a portion of the network does not work until those commands are added back to the configuration.
Conditions: This issue occurs after a reload.
Workaround: There is no workaround.
Symptom: Traceback messages are observed on removing EVC due to bad magic number in chunk header.
Conditions: This issue occurs when ScEoMPLS pseudowire is configured on the EVC on Cisco ME3600X.
Workaround: There is no workaround.
Symptom: After the TCAM has been exhausted, and resources removed, a service policy cannot be applied. The TCAM clean up is not functioning properly.
Conditions: This issue is observed on the Cisco ME 3600X switch after exhausting the TCAM.
Workaround: Reload the switch.
Symptom: MET consumption increases on sending traffic through a VPLS configuration. This leads to MET resources being exhausted.
Conditions: This issue occurs while sending traffic through a VPLS configuration.
Workaround: Reload the switch.
Symptom: Ping does not work from Provider Edge (PE) to PE over RPW, and PM-4-BAD_VLAN_ID traceback messages are observed.
Conditions: This issue occurs with SVI based Multiprotocol Label Switching (MPLS) link.
Workaround: Configure the core facing interface with default values, and apply the configurations back.
Symptom: When changing a QoS policy for classification, such as altering a class-map or changing an associated ACL, while the policy is attached to an interface, the classification stops working and remarks traffic in a strange way.
Conditions: This issue is observed when changes are made to the match statement on the class-map or in the access-list.
Workaround: Remove the policy from the interface and reattach it.
Symptom: Internet Control Message Protocol (ICMP) Layer 4-type is not accepted in the QoS ACL configuration.
Conditions: This issue occurs under normal conditions.
Workaround: There is no workaround.
Symptom: Unable to delete pseudowire interface after its creation.
Conditions: This issue occurs when a pseudowire interface defined with only one value is deleted.
Workaround: Define the pseudowire with two numbers. Note the second number as it is not displayed in the running configuration.
Symptom: Traceback messages are observed when the interface is removed and added back quickly.
Conditions: This issue occurs when the interface is removed or configuration is replaced such that a port is removed and added back quickly.
Workaround: There is no workaround.
Symptom: Broadcast traffic sent through span destination port forwards traffic, and other ports of the Cisco ME3800X switch receives it.
Conditions: This issue occurs under normal conditions.
Workaround: There is no workaround.
Symptom: Unable to run Ethernet Loopback (ELB) commands on service instances when the physical port is down.
Conditions: This issue occurs when the service provider does the loopback testing of EVCs on a port that is down.
Workaround: There is no workaround.
Symptom: Multicast traffic drops as it hits the signal-to-noise TCAM entries.
Conditions: This issue occurs when signal-to-noise entry matches the source IP address of the multicast packet.
Workaround: There is no workaround.
Symptom: Egress classification based on CoS for traffic on EVC bridge-domain does not work.
Conditions: This issue occurs when QoS is applied on the main interface and on the same port EVC cross connect is configured.
Workaround: Apply the QoS policy-map on EVC instead on main interface.
Symptom: Traffic fails when packets flow out of access interface with MPLS tag.
Conditions: This issue occurs when the outgoing packet size is large.
Workaround: There is no workaround.
Symptom: MAC flaps seen in VPLS hub-spoke model.
Conditions: This issue occurs when the first VC that comes up is a spoke VC.
Workaround: Reconfigure Layer 2 VFI or flap the VLAN interface.
Symptom: Unable to save license related information in secondary storage device (for example, flash).
Conditions: This issue occurs under normal conditions.
Workaround: There is no workaround.
Symptom: Access-list counter for denied packets do not work on the Cisco ME3600X switch.
Conditions: This issue occurs when the ACE specifies a “deny” action with the logging option enabled.
Workaround: There is no workaround.
Symptom: Static MAC address configured on the same VLAN disappears after reload.
Conditions: This issue occurs when the static MAC address is configured on the same VLAN.
Workaround: There is no workaround.
Symptom: The configuration disappears after a switch reload when the speed is configured with a value equal to 1000 and duplex is configured as “full”.
Conditions: This issue occurs after a switch reload.
Workaround: Perform a reconfiguration.
Symptom: The egress traffic across a few gigabit ports on the Cisco ME3600-24CX-M Switch stops.
Conditions: This issue occurs when Cisco ME3600-24CX-M Switch is configured in the IO Mode 3, and the MTU is greater than 2600 bytes.
Workaround: There is no workaround.
Symptom: IGMP Snooping is not functional up on adding a new interface.
Conditions: This issue occurs when IGMP Snooping is enabled, and TCN is disabled.
Workaround: There is no workaround.
Symptom: Traffic is incorrectly classified under class-default in the TAG2IP case.
Conditions: This issue occurs when a QoS-group is set at the ingress core interface and the egress access interface matching the QoS-group traffic gets incorrectly classified to class-default.
Workaround: There is no workaround.
Symptom: Traffic is not looped back for EFP with a double tag configured for data plane loopback with cos as the filter criteria for loopback.
Conditions: Traffic to be looped back is having q-in-q tags and different outer cos and inner cos specified. Loopback is configured with cos match criteria.
Workaround: Traffic to be looped back should have both outer and inner cos as same value and this cos value should be used in the data plane loopback configuration.
Symptom: Ingress policy-map is rejected on main interface.
Conditions: This issue occurs when egress policy-map is applied on the service instance.
Workaround: Apply the ingress policy-map first and then the egress policy-map on the service instance.
Symptom: Missing packets and jitterbuffer underrun counters are incrementing on a Cisco ME3600CX node where E1 pseudowire (SATOP) is created between a Cisco ME7600 and the Cisco ME3600CX over the MPLS cloud.
Conditions: This issue occurs when trigger for this issue is to add “mpls ldp explicit null” on PE routers.
Workaround: There is no workaround.
Symptom: ‘Confirm’ and ‘Exceed’ action counters in the show policy-map interface are cleared with scale configurations.
Conditions: This issue occurs on clearing the counters.
Workaround: Remove the policy-map and add it back.
Symptom: Traffic burst differences from 10G.
Conditions: This issue occurs when the traffic burst from 10G is connected to ASIC1 is twice compared to asic0.
Workaround: There is no workaround.
Symptom: Additional 4 bytes are added into Maximum Transmission Unit (MTU) calculation. The packets are punted to CPU even if the size of the packet is under the configured MTU.
Condition: This issue occurs on the Cisco ME3600X switch running 15.1(2)EY1a, and during the MPLS label swap situation.
Related Documentation
These documents provide complete information about the switch and are available from these Cisco.com sites:
ME 3800X switch:
http://www.cisco.com/en/US/products/ps10965/tsd_products_support_series_home.htmlME 3600X and ME 3600X-24CX switch:
http://www.cisco.com/en/US/products/ps10956/tsd_products_support_series_home.html
Note Before installing, configuring, or upgrading the switch, see these documents:
- For initial configuration information, see the “Configuring the Switch with the CLI-Based Setup Program” appendix in the hardware installation guide.
- For upgrading information, see the “Downloading Software” section in the release notes.
- Cisco ME 3800X and ME 3600X Switch Software Configuration Guide
- Cisco ME 3800X and ME 3600X Switch Command Reference
- Cisco ME 3800X and ME 3600X System Message Guide
- Cisco ME 3800X and ME 3600X Switch Hardware Installation Guide
- Cisco ME 3800X and ME 3600X Switch Getting Started Guide
- Installation Notes for the Cisco ME 3800X and ME 3600X Switch Power-Supply and Fan Modules
- Regulatory Compliance and Safety Information for the Cisco ME 3800X and ME 3600X Switches
- Cisco ME 3600X-24CX Switch Hardware Installation Guide
- Cisco ME 3600X-24CX Switch Getting Started Guide
- Installation Notes for the Cisco ME 3600X-24CX Switch Power-Supply and Fan Modules
- Regulatory Compliance and Safety Information for the Cisco ME 3600X-24CX Switch
- Cisco Small Form-Factor Pluggable Modules Installation Notes
- Cisco 10-Gigabit XFP Transceiver Modules Install Note
- Cisco CWDM GBIC and CWDM SFP Installation Notes
These compatibility matrix documents are available from this Cisco.com site:
http://www.cisco.com/en/US/products/hw/modules/ps5455/products_device_support_tables_list.html
- Cisco Gigabit Ethernet Transceiver Modules Compatibility Matrix
- Cisco 100-Megabit Ethernet SFP Modules Compatibility Matrix
- Cisco CWDM SFP Transceiver Compatibility Matrix
- Cisco Small Form-Factor Pluggable Modules Compatibility Matrix
- 10-Gigabit Ethernet Transceiver Modules Compatibility Matrix
- Compatibility Matrix for 1000BASE-T Small Form-Factor Pluggable Modules
Obtaining Documentation and Submitting a Service Request
For information on obtaining documentation, using the Cisco Bug Search Tool (BST), submitting a service request, and gathering additional information, see What’s New in Cisco Product Documentation .
To receive new and revised Cisco technical content directly to your desktop, you can subscribe to the What’s New in Cisco Product Documentation RSS feed . The RSS feeds are a free service.
This document is to be used in conjunction with the documents listed in the “Related Documentation” section.
Cisco and the Cisco logo are trademarks or registered trademarks of Cisco and/or its affiliates in the U.S. and other countries. To view a list of Cisco trademarks, go to this URL: www.cisco.com/go/trademarks . Third-party trademarks mentioned are the property of their respective owners. The use of the word partner does not imply a partnership relationship between Cisco and any other company. (1110R)
FeedbackContact Cisco
- Open a Support Case
- (Requires a Cisco Service Contract)