The following are descriptions of the open and resolved bugs in Cisco WAE Release 6.3.3. The bug ID links you to the Cisco Bug Search tool.
Table 1 Open Bugs
Cisco WAE core dump files are created in the /opt/cariden directory.
snmp_poll command cannot poll large networks.
Table 2 Resolved Bugs
WAE Design may hang indefinitely if it is reopened after borrowing a license.
The TE metric data collected on all Bundle-Ethernet interfaces from Cisco NCS 6000 (IBR) routers running IOS XR version 5.2.1 is incorrect.
The ad hoc report in WAE Live is incorrectly counting interfaces twice.
get_configs command does not collect Hua Wei router configuration information.
Using the Bug Search Tool
Use the Bug Search tool to search for a specific bug or to search for all bugs in a release.
Step 1 Go to http://tools.cisco.com/bugsearch.
Step 2 At the Log In screen, enter your registered Cisco.com user name and password; then, click Log In. The Bug Search page opens.
Note If you do not have a Cisco.com user name and password, you can register for them at http://tools.cisco.com/RPF/register/register.do.
Step 3 To search for a specific bug, enter the bug ID in the Search For field and press Return.
Step 4 To search for bugs in the current release:
a. In the Search For field, enter a problem, feature, or a product name (for example, Cisco WAN Automation Engine) and press Return. (Leave the other fields empty.)
b. When the search results are displayed, use the filter tools to find the types of bugs you are looking for. You can search for bugs by modified date, status, severity, and so forth.
To export the results to a spreadsheet, click the Export Results to Excel link.
This section describes the limitations and restrictions for Cisco WAE.
In some Linux installations with Xfce desktop installed, the documentation does not open from the WAE Design GUI Help menu. The workaround is to do one of the following:
- Open the help files from a terminal in the
- Install a default browser.
- Install the following packages:
yum groupinstall "X Window System"
yum groupinstall "Desktop"
yum groupinstall "General Purpose Desktop"
WAE Collector and WAE Network Interface (NI)
- Due to vendor MIB limitations, Collector cannot represent QoS traffic on interfaces that have more than one VLAN configured. If a network contains such interfaces, their queue traffic statistics are omitted from the collection. The total traffic on these interfaces is still measured. As a result, per class-of-service demands estimated through Demand Deduction are less accurate. Estimates of traffic totals over all classes of services, however, are not affected.
- Due to lack of MIB support, SR tunnel type is not collected for Cisco IOS XR routers through SNMP.
- Collection of interface egress shaping rate for Alcatel-Lucent devices does not support LAG interfaces.
- Shared Risk Link Groups (SRLGs) are not supported in Alcatel-Lucent Service Aware Manager (SAM) collection.
- The interval for continuous LSP discovery in WAE NI cannot be less than 60 seconds.
- LSP's ActualPathHop cannot be resolved when using continuous collection. As a workaround, use interval-based collection.
- If upgrading the Collector server from 5.6x to 6.1x or 6.2, the
$CARIDEN_ROOT/etc/collector/server/db-persistence/DiscoveryEngineImplementation.db file must be removed prior to starting the web server. Since installation automatically starts the web server, the recommendation is to remove this prior to installation.
- OSPFv3 and IPv6 IS-IS databases cannot be collected. The workaround is to use a manual snapshot.
- SNMPv3 is not an available option when configuring default credentials.
– Does not support association of a GRE tunnel with the physical interface it uses to reach the tunnel destination since the IP-Tunnel MIB lacks this information.
– Does not update LAG port status if LAGs are discovered running both
snmp_find_interfaces. The workaround is to run only
- Juniper routers: Signaled standby LSP path option is not available from the standard MPLS-TE MIB for Juniper routers. Only the active path option name is collected.
- Cisco IOS XR routers
– IGP topology collected through
– IS-IS link-state database with TE extensions contains incorrect interface “admin-weights” (TE metric) on Intel-based routers.
– IPv6 IS-IS link-state database does not contain IPv6 interface addresses or parallel interfaces. This information is only available when Cisco IOS XR supports IS-IS IPv6 TE extensions. The
snmp_find_interfaces tool collects this information.
– MAC Accounting is not supported.
snmp_find_rsvp does not set the Standby value in the <LSPPaths> table for signaled backup paths or collect named affinities configured with affinity-maps.
find_bgp does not build BGP pseudo-nodes among internal ASNs.
find_bgp does not collect BGP peers under PE-CE VRFs.
– Does not accurately detect the bandwidth of some Juniper ‘ge’ interfaces that have a capacity of 10 Gbps.
– Collects POS bundles, but has limitations due to unavailability of the port OperStatus property.
- TE Extended Admin Groups (EAGs), also known as extended affinities, are not supported.
- Port circuits are not built for LAG/bundle members whose nodes are not within the same IGP instance as the AS.
- There is no support for building port circuits for LAG members that are not within the same IGP (inter-AS circuits)
- It is not possible to distinguish between physically connected and unconnected LAG ports that are down for LAG port matching.
snmp_find_ospf_db cannot be used when routers have a large number of links that cannot fit into a single PDU.
find_bgpls does not support multi-area OSPF or multi-level IS-IS, non-TE-enabled interfaces, and pseudo-nodes. The workaround is to use SNMP- or login-based discovery.
get_inventory does not collect Juniper multi-chassis router hardware inventory.
- Segment routing
– SR protected adjacency SIDs are not supported.
– Concurrent RSVP-TE and SR-TE paths are not supported on the same LSP.
SAM-OSS Integration with Snapshots
sam_getplan does not populate the <NodeTraffic> table. This table is derived and populated when
sam_getplan and SNMP tools are used together.
sam_getplan does not populate the NetIntActivePath column in the <LSPs> table.
sam_getplan and SNMP tools are used together in the snapshot process for multi-vendor network collection, then Alcatel-Lucent traffic measurements cannot be aligned with those collected from other router platforms.
Cisco Open SDN Controller (OSC)
During detailed PCEP tunnel creation or when modifying PCEP tunnels, affinity values are misinterpreted if multiple affinities are specified. This limits you to specifying one affinity for IncludeAffinity, IncludeAnyAffinity, and ExcludeAffinity, and each of these values must be a number within [0,31].
- LSP affinities are deployed, while interfaces affinities require separate provisioning.
- LSPs that exist in the network by another controller cannot be updated.
- Deployment of each RSVP-TE named-path or SR-TE segment-list is limited to a single LSP.
- Cisco IOS XR: WAE client specifies the XR LSP signaled-name, while NSO service and device use tunnel-id. The workaround is to deploy all Cisco IOS XR LSPs using the tunnel-id and to make sure that existing LSPs are not redeployed.
- NEDs (NSO console)
– For Cisco IOS XR, there is no option to give the IP address of the LSP directly; you can only specify a loopback address. There is no option to give tunnel affinity values directly; you can only specify an affinity-map name.
– For Junos, there is no inter-domain keyword, which is used only when an inter-area LSP is created.
Installation and Startup
- The WAE NI server and the WAE Core server cannot reside on the same device or on the same VM. Note that the Cisco WAE Server Installation Guide assumes that they are on the same device. If needed, contact your support representative for further installation details.
- If the OS is using an old CA certificate to verify the integrity of the EPEL repository, you might see this error from the OS vendor:
Error: Cannot retrieve metalink for repository: epel. Please verify its path and try again.
– One workaround is to perform an offline installation. For instructions, refer to the “Offline Installation” chapter in the Cisco WAE Server Installation Guide.
– Another workaround is to change https to http.
Note This is not a secure solution. For information on how to resolve OS security issues, contact your OS vendor.
1. In the
/etc/yum.repos.d/epel.repo file, change the first instance of https to http.
sudo vim /etc/yum.repos.d/epel.repo
http in the following line:
2. Execute yum to clean up makecache.
sudo yum clean all && yum makecache
3. Rerun the installer. For detailed installation instructions, see the Cisco WAE Server Installation Guide.
sudo bash wae-k9-<version>.bin
$CARIDEN_HOME directory is not automatically added to
$CARIDEN_HOME/bin is). If not in
$CARIDEN_HOME/bin, to start the WAE Design GUI from the command line, you must specify its full path.
embedded_web_server tool is deprecated. The recommendation is to use the
wae-web-server service, which is constantly monitored to be brought up automatically.
By default, this web service starts upon installation completion. Therefore, if you stop the web server using the
embedded_web_server tool (
embedded_web_server -action stop), the web server does not stop. The workaround is the following:
service wae-svcs-mon stop
embedded_web_server -action stop
WAE Statistics UI
The WAE Statistics page does not open in all browsers. The workaround is the following:
1. Click the WAE Statistics link. The URL format is https:// <server_IP> :8443. Example:
2. Copy the URL of this page to another browser window.
3. In the new browser, change the URL port from 8443 to 8843. Example:
4. Follow the browser messages to accept the connection and add it as an exception.
Web User Management
Both the System UI and the WAE Design Archive UI have local user management capabilities. If both are used to configure users, WAE uses the most recently updated information. The recommendation is to use only the System UI to manage local users.
License Check Failures on Newer Linux Distributions
Some newer Linux distributions have started using a new way (via
biosdevname) of naming hardware devices, including the network interfaces. This causes some software that depends on the traditional naming (for example,
eth1) to fail on license checks, including MATE.
The workaround is to append
biosdevname=0 to the kernel line of the grub configuration file and reboot. (Syntax varies among distributions.)
After reboot, you should be able to use ifconfig to verify that the NICs are named
eth1,...) instead of the
biosdevname names (such as
Certain tools (such as
parse_configs) may require more memory to start than what is available. The symptom is an error message similar to the following:
Error occurred during initialization of VM.
Could not reserve enough space for object heap.
Error: Could not create the Java Virtual Machine.
Error: A fatal exception has occurred. Program will exit.
The workaround is to set the maximum memory to a low enough value in the
CARIDEN_JAVA_OPTIONS variable before calling the tool. An example setting is as follows: