|
|
The 9.2.00 Cisco WAN Manager (CWM) for Solaris software release supports new Cisco MGX 8850 products and the existing Cisco WAN switching products (including the Cisco IGX 8400 series and the Cisco BPX 8600 series) running the Switch software version 9.2 based release and the existing Cisco MGX 8220 product running the software version 5.0 based release.
The CWM 9.2.00 release is fully Y2K compliant. The compliant matrix is listed in next section. There is a product name change in this release from Strata View Plus (SV+) to Cisco WAN Manager (CWM). All references to SV+ are changed to CWM including the start-up script, the desktop and icon names. Currently, the CWM 9.2.00 does not support any CWM upgrades.
The CWM 9.2.00 is bundled with Cisco View WAN switching device packages (WANCV) release 2.0.00 for the device management.
This release note contains the following sections:
Three types of machines are supported for WAN Manager 9.2.00 as standard platforms. They are low-end, mid-range and high-end platforms. The configuration of each platform type is listed in Table 1, "Sun Platform Requirements".
| Platform Type | Machine Type | Number of CPU | Size of RAM | Graphic Card Type | Hard Disk Drive | Swap Space | Desktops Supported (n) | Connections Supported (m) |
|---|---|---|---|---|---|---|---|---|
Low End | Sun Ultra Enterprise 1 Model 170 or Sparc 20 Model 151 or Ultra 10 with SCSI controller | 1 | 256 MB | 24 bits | One 2 GB + one 4 GB drives | 512 MB | n<5 | m<5,000 |
Mid Range option 1 | Sun Ultra Enterprise 2 Model 2300 or Ultra 30 model 300 | 1 (300 MHz processor) | 512 MB | 24 bits | Two 4 GB drives or 1 9 GB drive | 1 GB | 5<=n<10 | 5,000<= m<25,000 |
Mid Range option 2 | Sun Ultra Enterprise 2 Model 2300 or Ultra 60 | 2 (300 MHZ processors) | 512 MB | 24 bits | Two 4 GB drives OR 1 9 GB drive | 1 GB | 5<=n<10 | 25,000<=m<50,000 |
Mid Range option 3 | Sun Ultra Enterprise 2 Model 2300 or Ultra 60 | 2 (300 MHZ processors) | 1 GB
| 24 bits
| Two 4 GB drives OR one 2 GB + one 9 GB drives | 2 GB | 5<=n<10 | 50,000<=m<100,000 |
High End | Sun Enterprise 4000 or Enterprise 450 | at least 4 | 1 GB | 24 bits | One 4 GB drive + one 9 GB drives or disk array | 2 GB | n>=10 | m>=100,000 |
The selection of a proper CWM platform depends on a number of factors, such as number of CWM desktops, number of managed connections and number of statistics collected and stored. Table 2 lists recommended CWM platforms based on the size of network.
The following are additional notes for CWM platform requirements:
Three types of machines are supported for CWM 9.2.00 stand-alone SCM as standard platforms. They are low-end, mid-range and high-end platforms. The configuration of each platform type is listed in Table 2, "Platform Requirements for Stand-Alone SCM". Other equivalent platforms running Solaris 2.6 are also supported.
| Platform Type | Machine Type | Number of CPU | Size of RAM | Hard Disk Drive | Size of Swap Space |
|---|---|---|---|---|---|
Low End | Sun Ultra Enterprise 1 Model 170 or Sparc 20 | 1 | 128 MB | One 2 GB drive | 256 MB |
Mid Range | Sun Ultra Enterprise 2 Model 2000 or Sun Ultra-30 | 1 | 256 MB | One 2 GB drive | 512 MB |
High End | Sun Ultra 60 | 2 | 512 MB | One 2 GB drive | 1 GB |
The CWM 9.2.00 release works with HP Openview Release 5.01 for Solaris 2.6. See "Third Party Product Limitations Notes" for HP OpenView patches required by CWM.
The default color of the background map can make it difficult to see the color of the nodes and the trunks on the map.
For HP OpenView installation requirements and procedures, please refer to "HP OpenView Network Node Manager Products, Installation Guide" (part # J1136-90000 from HP).
The CWM 9.2.00 release uses Informix 7.23 for database.
The Report Application of CWM 9.2.00 uses WingZ 2.5 or later. For proper working of the Report Application the DataLink package of wingZ must also be installed along with WingZ package.
This section provides information about hardware supported by CWM 9.2.00.
Table 3, "Supported Legacy Hardware", lists the legacy hardware supported in CWM 9.2.00.
| Platform Series | Front Card | Back Card | Firmware version | Software version |
|---|---|---|---|---|
IGX 8400 | NPM | SCM | CVJ,CWP | 9.2.00 |
IGX 8400 | NTM | BC-T1/E1/SR/Y1 |
| 9.2.00 |
IGX 8400 | BTM | AIT-T3/E3/E2/HSS1/E1 | AJF,BCF,BDK,BDL | 9.2.00 |
IGX 8400 | FRM | FRI-T1/E1/V.35/X.21 |
| 9.2.00 |
IGX 8400 | FTM | FPC-V35/T1/E1/X.21 |
| 9.2.00 |
IGX 8400 | HDM | V.35/RS-449/RS-232C/RS-232D | CEF | 9.2.00 |
IGX 8400 | LDM | RS-2324/RS-2328/RS-232D |
| 9.2.00 |
IGX 8400 | CVM | BC-T1/E1/J1/Y1/SR | AFF,AHF,BHC | 9.2.00 |
IGX 8400 | UVM | UVI-2T1EC/2E1EC |
| 9.2.00 |
IGX 8400 | UFM-C | UFI-8T1/8E1-DB15,UFI-8E1-BNC |
| 9.2.00 |
IGX 8400 | UFM-U | UFI-V35/X21/HSI |
| 9.2.00 |
IGX 8400 | ARM | ARI | BA | 9.2.00 |
IGX 8400 | UXM | UAI-4OC3MMF/40C3SMF/2OC3SMF/6T3/6E3/3T3/3E3/8T1-IMA/8E1-IMADB15/8E1-IMABNC | AAB,ENW,P0512 | 9.2.00 |
BPX 8600 | BCC | BCC-bc/3-bc | B0406,BKF,BKJ,BMH,BYJ | 9.2.00 |
BPX 8600 | ASM | LM-ASM | AAA,ABA,ACA,BC03 | 9.2.00 |
BPX 8600 | BNI-T3 | LM-3T3 | CCJ,CEL,CHD,CHF | 9.2.00 |
BPX 8600 | BNI-E3 | LM-3E3 | CFD,CFF,CLF | 9.2.00 |
BPX 8600 | BNI-155 | LM-OC3-SMF/SMFLR/MMF |
| 9.2.00 |
BPX 8600 | ASI-1 | LM-2T3/2E3 | BDK,BFJ,BXJ | 9.2.00 |
BPX 8600 | ASI-155 | LM-2OC3-SMF/SMFLR/MMF | CB30,HFA | 9.2.00 |
BPX 8600 | BXM-T3/E3 | BPX-T3/E3-BC |
| 9.2.00 |
BPX 8600 | BXM-15 5 | MMF/SMF/SMFLR-155 | DC1,DL1 | 9.2.00 |
BPX 8600 | BXM-622 | MMF/SMF/SMFLR-622 |
| 9.2.00 |
MGX 8220 | FRSM-4P | DB-15-4T1/4E1, BNC-4E1 |
|
|
MGX 8220 | CESM-4P | DB-15-4T1/4E1, BNC-4E1 |
|
|
MGX 8220 | AUSM-4P | DB-15-4T1/4E1, BNC-4E1 |
|
|
MGX 8220 | FRSM-8P | RJ48-8T1/8E1, SMB-8E1 | 5.0.00 |
|
MGX 8220 | AUSM-8P | RJ48-8E1/8T1 | 5.0.00 |
|
MGX 8220 | CESM-8P | RJ48-8T1/8E1, SMB-8E1 | 5.0.00 |
|
MGX 8220 | IMATM-B | RJ48-T3T1/E3E1/E3T1/T3E1 | 5.0.00 |
|
MGX 8220 | FRSM-HS1 | HSI-4X21/3HSSI | 5.0.00 |
|
MGX 8220 | BNM-T3/E3 | T3E3-B/D | 5.0.00 |
|
MGX 8220 | BNM-155 | 155-SMF |
|
|
MGX 8220 | SRM | DB-15-T1/E1, LM-3T3 |
|
|
MGX 8220 | ASC | ASC-LM | 5.0.00 |
|
Table 4, "Supported New Hardware", lists the new hardware supported in CWM 9.2.00.
| PLATFORM | Front Card | Back Card | Firmware Version | Software Version |
|---|---|---|---|---|
MGX 8220 | FRSM-VHS | HS2 | 5.0.00 |
|
MGX 8220 | AUSM model B | 8T1/E1 | 5.0.00 |
|
MGX 8850 | PXM1 | T3/E3/OC3/OC12 | 1.0.00 |
|
MGX 8850 | FRSM-VHS | HS2/T3/E3/CT3 | 10.0.00 |
|
MGX 8850 | FRSM-8P | RJ48-8T1/8E1, SMB-8E1 | 5.0.00 |
|
MGX 8850 | CESM-8P | RJ48-8T1/8E1, SMB-8E1 | 5.0.00 |
|
MGX 8850 | RPM | 10mbps/100mbps | rpm-js-m2.120-2.5.T |
|
MGX 8820 | SRM | LM-3T3 |
|
|
All event messages are now displayed in HPOV's Event Browser window. There is a limit of 2000 pending messages on HPOV's Event Browser. If there are more than 2000 messages to be processed and displayed by Event Browser it may exit.
The following patches are required for HPOV: PSOV_02091. These patches are supplied on the CWM 9.2.00 CD. If HP OpenView is installed on the system these patches are installed during CWM installation.
CWM 9.2.00 will operate with BPX/IGX releases 9.2 and 9.1. CWM 9.2.00 will operate with MGX 8220 release 4.1 and 5.0 for all supported features. CWM 9.2.00 can also operate in mixed networks of 9.1 and 9.2 switch software.
| Software | Release | Patch | Y2K compliance Certified By | Remarks |
|---|---|---|---|---|
Cisco WAN Manager | 9.2.00 | N/A | Cisco Systems 29 Test Case suite |
|
HP OpenView | 5.01 | PSOV-02091 PSOV_02208 | HP |
|
WINGZ | 2.5 | N/A | IISC |
|
Informix | 7.2.3.UC4 | N/A | Informix |
|
Solaris OS | 2.6 | Solaris 2.6 Recommended Patch Cluster Dated Jan/14/1999 And Solaris 2.6 Y2000 Patch Cluster Dated Jan/04/1999
| Sun Micro Systems | These patch clusters can be downloaded (as of Jan/20/1999) from http://sunsolve.sun.com/pub-cgi/usr/pubpatchpage.pl |
CiscoView MSSBU device packages (WANCV) are upgraded to release 2.0 to support IGX/BPX Release 9.2.00, MGX 8220 Release 5.0 enhancements, and MGX 8800 family products.
The network will allow the connection of up to 12 CWM stations. There is no change in WAN Manager Release 9.2.00 software for this feature; the switch software and MGX 8220 (4.04 onwards) have increased their limits to support this.
Make sure at least 900 MB of disk space in /usr/users/svplus area can be allocated for CWM 9.2.00 installation. In the case of upgrade, disk space used by existing CWM software plus free disk space should exceed 900 MB.
If you plan to use auto install and intend to partition the disk during auto install you have two choices for installing Wingz:
1. Install Wingz in /opt/Wingz before installing CWM and then execute IntegrateWingz.
2. Install Wingz in /usr/users/Wingz after installing CWM and then execute IntegrateWingz.
If you do not plan to partition the disk during auto install or if you are using manual install you have two choices for installing Wingz:
1. Install Wingz either in /usr/users/Wingz or /opt/Wingz before installing CWM.
2. Install Wingz either in /usr/users/Wingz or /opt/Wingz after installing CWM and run IntegrateWingz.
Install Wingz software from the Wingz tape or CD. Note that WingZ version 2.5 should be installed for CWM release 9.2.00. After installing, contact Wingz for a runtime license and blend it with CWM 9.2.00. Failing to do so will result in following error message when clicking on the CWM Report icon: "There is a problem with the registration information for this machine. Contact your system administrator".
If Informix is installed on a disk array (typically the device name is /dev/vx/rdsk/rootdg/vol03) then the ownership of the device and the group ownership of the device should be set to informix. The mode of the device should be 770. Otherwise, after reboot the machine, the following error message will appear in the console, "Shared memory is not initialized".
If you are installing HP/OV 5.01 over an existing version make sure that the /etc/snmpd and /etc/snmpd.conf files are removed or re-named before carrying out the upgrade. If these files are not removed or renamed, it is likely that you will see the following message when trying to re-start OV:
object manager name: netmon behavior: OVs_WELL_BEHAVED state: NOT_RUNNING PID: - last message: Can't talk to snmpd; see /usr/OV/log/netmon.trace for info exit status: exit(1)
Information left in these files causes problems for the new version of OV.
These files are not removed by the ovremove program.
The upgrade of CWM from 9.1.x to 9.2.00 will be supported in future release.
CWM 9.2.00 to Cisco WAN switch (outgoing): ports used in the switch
Cisco WAN switch to CWM 9.2.00 (incoming): ports used in CWM 9.2.00
After performing the creation of a new database by executing the create_db command, if WAN Manager is restarted before it is synchronized with the network, it will result in a database inconsistency. In this case, perform create_db before restarting CWM to avoid database inconsistency.
The logical_connection table count for a node in CWM database may not match the master lcon counts in that node. This will happen if the database contains FastPAD connections, or if the database has slave lcons with no connections.
If Gateway node is upgraded or rebuilt, before CWM has synchup with the network, CWM needs to be cold-started (created CWM database and restart CWM). This is required to ensure CWM database consistency.
CWM database and network may differ by value 1 for parameters PCR(0+1), SCR(0+1), MCR, ICR of some ATM-FR connections. This happens because of the difference (rounding effect) in conversion algorithms used by CWM and switch to convert values in cells/sec units to 100 bit/sec units.
| cell/second | 100 bit/second |
|---|---|
50 | 192 |
5760 | 22119 |
3259 | 12515 |
2834 | 10883 |
| Cell/second | CWM 100 bit/second | Switch 100 bit/second |
|---|---|---|
5760 | 22119 | 22118 |
3259 | 12515 | 12514 |
2834 | 10883 | 10882 |
Any node with IP address 192.0.0.0 or 0.0.0.0 will not be displayed or stored in CWM database. In order to configure the correct network IP address use cnfnwip command on the switch.
Each IPX/BPX/IGX/MGX 8220/MGX 8800 node to be managed by WAN Manager must be configured with an IP Relay IP Address (cnfnwip). The gateway node must be configured with both LAN and IP Relay IP Addresses.
The IPX/BPX routing and IPX/AF feeder modes must be configured with read and write SNMP community names. The SNMP community names "public" and "private" are configured in CWM for GET and SET respectively as defaults.
The directly attached node (Gateway node) must be configured with IP Relay IP address before starting WAN Manager.
Node names, IP Addresses and NWIP addresses of the nodes in a network and across networks in multiple networks scenarios, should be uniquely configured.
Robust message updates must be enabled.
The gateway node name has to be the same as the node name specified in the config.sv file.
The maximal number of nodes can be added through Network Configurator is 2000. Currently, there is no error message to inform the user of this situation. This is a limitation of the CWM 9.2.00 back end processes.
In the case of managing standalone MGX 8850 nodes and a tiered AutoRoute network with automatic discovery enabled, the addition of standalone nodes using the Network Configurator can't take place while the CWM is synchronizing with the network.
A trap will be generated when the network synchronization is done. The user can activate the standalone network configurator by running the tool: /usr/users/svplus/bin/configurator.
For adding a connection the connectionIndex must be 0 (i.e., there is no need to do a GET on connAvailIndex). The actual index is internally generated by the connection service agent itself.
CWM 9.2.00 does not make use of the 'Service Rate' parameter applicable to FRSM channels, while adding FRSM channels. Instead, the CIR is used. If a customer adds FRSM PVCs through the CLI making use of the 'Service Rate', CWM 9.2.00 cannot correctly display ALL parameters of the PVC; also the PVC cannot be correctly modified using CWM.
For DAS connections, deleting an end point from a segment and later adding it to another segment at a very fast rate may result in a database inconsistency.
Test Connection and Test Delay do not work for connections routed through a Cisco virtual trunk cloud when the connection originates from an MGX 8220 or BPX midpoint.
Connection Manager GUI for preferred route displays ALL available trunks without considering the traffic class. As a result, it is possible to set a preferred route for a CBR connection traversing trunks that do not support CBR traffic.
In the Connection Manager GUI, you must enter both the end points (i.e. node, shelf, slot, port) in the Add Connection window before entering any connection or midpoint parameters.
For an FRSM - ASI PVC connection, the error message "Remote SCR value out of range" generated by the remote BPX, indicates that the CIR value of the Frame Relay side is below the acceptable limit (cell rate) of the trunk between the MGX 8220 shelf and its routing node. To correct the problem, one can increase the value of CIR. E.g., if the trunk between the MGX 8220 shelf and the routing node is a BXM trunk, the CIR value should be at least 19.2 kbps (= 50 cell/sec).
ABR.FS connections cannot be added with default values for ATM-FR SIW connection. The 'ICR TO' value has to be changed.
Certain types of connections, such as VBR.1 or CBR, with policing parameters can't be modified by CWM if the default was changed using the switch command line interface.
While adding three segment frame relay connections where one routing node is IGX and another is BPX, connections must be mastered on feeders connected to BPX. Attempting to master connections on feeders attached to IGX hub will generate an "Internal error" error message, and the connection will not be added.
Three segment voice and data connections can only be added when both routing nodes are IGXs.This is due to a switch software limitation.
While adding connections originating and terminating on same feeder nodes, CWM will treat these connections as three segment connections and will add three segments (two on the same feeder node, one on the routing node) to realize this connection.
Switch software supports only ATFR and ATFST type of connections for UXM on a feeder node. CWM does not support ATFR or ATFST for ATM to ATM connection (on Connection manager GUI select Configure->New ATM-TM connection->VBR.3) the connection type on switch CLI will show as ATFR for all 3 segments. If you select ABR.FS connection type for ATM to ATM connection (on connection manager GUI select Configure->New ATM-ATM connection->ABR.FS) the connection type on CLI will be ATFST for all 3 segments. For all other limitations of connection type supported for UXM on feeder node please refer to the switch software release notes for release 9.2.00.
ATM-FR connections with FR endpoint on IGX/IPX running 8.4.x switch software can not be set with channel type SIW (Service InterWorking) transparent or SIW translate. This feature is not supported by IGX/IPX running 8.4.X SWSW. The connections thus set using 9.2.00 Connection Manager would fail.
The default values provided by the Connection Manager GUI for the Frame Relay end point parameters Ingress/Egress ECN Threshold and DE Threshold are not the same as those provided by the connection service agent.
The Connection Manager GUI uses percentage notation for both the ECN Threshold and DE Threshold. However, the service agent MIB uses the absolute value for those MIB variables. If the defaults were used for GUI and service agent, following values would be setup for the PVC:
| Parameter | GUI Default Values | Service Agent |
|---|---|---|
Ingress ECN Threshold | 10% of Ingress VC Q Depth -> 6554 | 6553 |
Ingress DE Threshold | 50% of Ingress VC Q Depth -> 32768 | 32767 |
Egress ECN Threshold | 10% of Egress VC Q Depth -> 6554 | 6553 |
Egress DE Threshold | 50% of EGress VC Q Depth -> 32768 | 32767 |
The switch generates error message Failure validating remote endpoint description when the PVC was modified or deleted using CWM. The problem occurs when CWM is managing a mix release network and the node mastering the pvc has an older release than the remote end, and the remote end point is in a newer card which was introduced after the older release. This is switch software limitation. The pvc can be deleted using the switch CLI.
Connection manager doesn't check for the SVC VPI/VCI partitioning range, if VPI value is ZERO and if SVC Hi VPI and Low VPI are set to ZERO on the switch. Because of this, VPI 0 can be used for provisioning. However, if user reserves only 0 for SVC range, Connection manager will not block the PVC provisioning.
In CWM 9.2.00, the FRSM-VHS card supports 4 separate QoS queues (versus 2 for legacy FRSM). The selection of QoS queue (or channel service type) on the FRSM-VHS channel has restriction on service type of the ATM routing segment. For High Priority class of connection (High Priority Queue, Q1), CWM 9.2.00 will pick CBR service type for the routing segment. For VBR-rt (real-time VBR, Q2), VBR service type is selected. For VBR-nrt (non-real time VBR, Q3), ATFR service type is selected. For ABR-FS (Available Bit Rate, Q4) or UBR (Unspecified Bit Rate, Q4), ATFS is selected.
When you start a stand-alone SCM, the following message is not an error message: Cannot communicate with CWM, either CWM is not active or another GUI is running presently.
You should select a routing node for the SCM time sync node.
For a group of selected nodes from which stats files are to be collected by a stand-alone SCM, the collection period should be set to the same value for all nodes.
If stats files are being saved, the SCM does not purge these files even if the disk is full.
The SCM supports 3100 nodes maximum in stand-alone mode.
Whenever a user performs a create_db operation, the statshistory.stats file must be manually deleted.
The following rules are used when you configure peak interval through CWM. Violating these rules will cause TFTP error 22 displayed in the SCM.
The switch needs some processing time to parse the stats.enable file. Depending on the number of stats and the number of objects this time can be very large. Currently SCM has a finite value for time-out (maximum 420 seconds) to wait for the switch response. Hence there might be cases where SCM will time-out waiting for switch to finish parsing the stats.enable file.If there are excessive time-outs then users should:
1. Cut down on number of stats enabled and be able to get the status of stats-enabling from SCM or
2. Keep the number of stats but lose the status of stats-enabling from the SCM
Users can use CLI commands such as dspstatparms, dspchstatcnf, dsptrkstatcnf, dspportstatcnf and dsplnstatcnf to make sure their desired stats are enabled. Note that there can be occasional time-outs if a node is busy for that period of time, but this case at the best is intermittent and should go away once loading on the node stabilizes Following is the formula to calculate the time to enable statistics on the switch:
time = (stats / cnt) * delay
where
stats = (conns * stats_per_con) +
(CLNs * stats_per_cln) +
(trks * stats_per_trk) +
(ports * stats_per_port)
cnt = number of stats configured in one group before giving up CPU, a configurable parameter on the switch.
delay = a delay time between each group of stats get enabled, this is also a configurable parameter on switch.
This is captured in CR CSCdj39737.
While CWM is running, if the X server crashes for any reason, then CWM should be stopped and restarted.
While CWM is running, if remote display is killed without properly shutting down the CWM Desktop, then reopening it remotely may not succeed.
CWM must be started from CDE environment.
The Wingz "Add Comment" option is not supported for virtual trunk and ATM connections. Wingz sometimes does not scroll up the list box properly.
Sometimes it may take 2-3 minutes to display new nodes after addition of virtual trunks. This is due to the time required to setup link0 and link1 subscriptions.
Always use the CWM ImageDownload program to do a SW/FW download. Since CWM does not make any assumptions on the filename, it requires you to tell the SW the filename and the path to download, avoiding any inconsistencies resulting from the file naming assumptions made by the getfwrev and loadfwrev commands. If the CWM ImageDownload program does not work, then there is some problem with the file transfer or the image.
CWM 9.2.00 supports only ATM connection type and vbr.3 service type for endpoints terminated on the RPM card.
For a RPM VCCON, the ONLY Vpi value allowed is 0. To add a VPCON, an '*' should be specified as the Vci value.
If a Vpi, Vci value is used to establish a RPM channel on one subinterface, the SAME Vpi, Vci pair cannot be used with another subinterface. This is due to a RPM limitation. If such action is attempted, CWM 9.2.00 will emit the error 'Local/Remote end point already exists'.
CWM 9.2.00 does not support provisioning of RPM channels for subinterface values over 32767.
For a PVC involving RPM, CWM 9.2.00 supports ATM Service Type VBR only (VBR.3 for atm-atm and atm-fr for atm to fr pvcs). The UBR Service Type is NOT supported for RPM by CWM 9.2.00.
CWM 9.2.00 provisions the RPM end of a 'End to End' PVC by using the IOS command 'atm pvc..' followed by the 'addcon' command. A RPM end point added using the 'atm pvc' command CANNOT be modified/deleted using the (newer) 'pvc add' command and vice-versa.
To establish a 'channel' on RPM, two IOS commands are required:
(a) 'atm pvc' command: To add the RPM 'leg' of the channel.
(b) 'addcon' command: To add the PXM 'leg' of the channel.
If the RPM leg gets deleted (new configuration restored on RPM), CWM 9.2.00 cannot detect the condition. However, if necessary, CWM 9.2.00 can be used to delete and readd the entire channel by reprovisioning the 'End to End' PVC.
If the PXM leg gets deleted(e.g 'clrallcnf' on MGX 8850), CWM 9.2.00 cannot provision the channel until the RPM leg is manually deleted. In order to do that, the user has to login RPM card and performs following IOS commands:
(a) 'no atm pvc' command: To delete the RPM 'leg' of the channel.
(b) 'no addcon' command: To delete the PXM 'leg' of the channel.
The peak & average values for a feeder PVC involving RPM is limited by the bandwidth of the MGX 8850 feeder trunk. As per 'strata' MIB, the value of average (expressed in cells/sec below) is limited for each trunk type as:
ASI/BNI
T3 : 0-96000
E3 : 0-80000
OC3 : 0-353208
BXM/UXM
T3 : 6-96000
E3 : 6-80000
OC3 : 6-353208
OC12: 6-1412832
ALM-B(BTM)
T3/E3 : 10-24000
The value of peak (express here in cells/s) is limited as for each trunk type as:
T3 : average-96000
E3 : average-80000
OC3 : average-353208
OC12: average-1412832
BXM with HEC line formating
T3_HEC: average-104268
While one CWM 9.2.00 station is used to provision connections on a specific RPM card, attempts to provision on the same card in another CWM 9.2.00 station can result in in failure in provisioning of RPM connections.
Provisioning of RPM can take up to 25 seconds on average. Rarely, CMGUI may time out before iosmgr returns. The returned IOS messages are updated in Result Pane on CMGUI eventually.
The followings are the typical causes of the error message "atm pvc missing":
(a) A pvc already exist on a point-to-point subinterface
(b) A pvc with same vpi/vci already exist on different subinterface.
The followings are the typical cause of the error "add connection to PXM failed":
(a) The resource partition is not configured
If error message encountered is one of the followings:
- "atm pvc missing"
- "add connection to PXM failed"
Do:
Step 1: telnet to MGX 8850 and login to RPM card (in enable mode)
Step 2: type command "show log" and hit <CR> to go to the end.
Note: log may be wrapped on RPM.
For other error messages returned by iosmgr:
Do:
Step 1: vi ~svplus/log/iosmgr.log
Step 2: go to the end of the file and check the command and arguments used for expect script. Check the tracefile as shown at the end of the command.
Step 3: open the tracefile and go to the end of the file. Examine the log of expect session.
The existing RPM connections will not be removed from the CWM database when the RPM card is pulled out of MGX 8850 shelf.
The SNMP master agent is configured to use 8161 as a default port id for agent communication. The configuration file SNMPAgent.conf for the master agent is located in /usr/users/svplus/config. If user is running another Network Manager which uses port number 8161, this UDP port can be configured to some other port number in this file as follows:
The Service MIB (Connection Service MIB and Port Service MIB) uses community strings for authentication. The Connection Service Agent and Port Service Agent processes use /usr/users/svplus/config/SNMPProxy.conf file as a configuration file. The SNMP community name strings can be configured in this file as follows:
The SNMP community names "public" and "private" are configured in this file for GET and SET respectively as defaults.
http://www.cisco.com/univ-src/ccden/data/doc/cview.htm
| Caution A "timeout" or "too big" error response for an SNMP get-request may produce undesired results with CiscoView 4.1 or the initial release of CiscoView 4.2. |
A typical SMU installation example (CiscoView on Solaris, using C shell):
# su root
# cd /var/tmp
# mkdir nmpath; cd nmpath
# ftp 192.31.7.130
# uncompress cview-01-sol-hpov-cv42.tar.Z
# tar -xvf cview-01-sol-hpov-cv42.tar
# source $NMSROOT/etc/install.cshrc
# ./install-smu
The various timers and retry counter should be configured from the "options->properties" menu. For a typical installation, Polling and Monitoring should be about 60 seconds and Timeout should be at least 15 seconds. Faster Polling and Monitoring can be configured if conditions permit.
Each IPX/BPX/IGX/MGX 8220/MGX 8800 node to be managed by WAN Manager must be configured with an IP Relay IP Address (cnfnwip). The gateway node must be configured with both LAN and IP Relay IP Addresses.
The IPX/BPX routing and IPX/AF feeder modes must be configured with read and write SNMP community names. The SNMP community names "public" and "private" are configured in CWM for GET and SET respectively as defaults.
The directly attached node (Gateway node) must be configured with IP Relay IP address before starting WAN Manager.
Node names, IP Addresses and NWIP addresses of the nodes in a network and across networks in multiple networks scenarios, should be uniquely configured.
WANCV 2.0 will operate with BPX/IGX releases 9.2 and 9.1, Releases of MGX 8220 later than 4.0.15, and all available releases of MGX 8850. The user should use CWM Equipment Manager to manage MGX 8220 release 4.0.15 and earlier releases.
Although the use of CiscoView Release 4.2 with all available Software Module Updates is strongly recommended, the WANCV device packages are compatible with CiscoView Release 4.1.
See the CWM 9.2.00 Version Software Release Notes (ENG-33276) for the detailed listing of hardware and firmware versions supported by this release.
The WANCV 2.0 release works with HP Openview Release 5.01 for Solaris 2.6.
For HP OpenView installation requirements and procedures, please refer to "HP OpenView Network Node Manager Products, Installation Guide" (part # J1136-90000 from HP).
The following patches are required for HPOV: PSOV_01688 and PSOV_01863. These patches are supplied on the CWM 9.2.00/WANCV 2.0 CD. If HP OpenView is installed, these patches are installed during CWM installation.
This section provides information about known anomalies and anomalies corrected with the 9.2.00 release of CWM. This section also provides information about anomalies known and corrected in the 2.0.00 release of WANCV.
This section describes known anomalies in the 9.2.00 version of WAN Manager.
| Bug ID | Description |
|---|---|
CSCdm19486 | Symptom: After the resource partition is changed on a FRSM-VHS card, It takes about 5 minutes for the CWM to update the information in the database. Workaround: After the resource partition is changed on a FRSM-VHS card, performs a node resync immediately. |
CSCdm18381 | Symptom: CMGUI can't see the ports on a primary BXM card after the Y-redundancy switch over. Workaround: None |
CSCdm17268 | Symptom: CMGUI will not clear alarm for PVCs added during the port failure Workaround: None |
CSCdm08477 | Symptom: Desktop will not start up when a non-CWM user starts it on sview200 machine in a customer's lab. It fails with a message: Init logger failed. Workaround: None |
CSCdk64308 | Symptom: l_port in rpm_connection and rpm_port table should be integer (4 bytes). Current CWM release does not support port more than 2 bytes long. This is a current limitation when adding a RPM subinterface/port through CWM. Workaround: Since RPM currently can only configure maximum 700 subinterface/port. The user is advised to specify 2 bytes range of the subinterface/port numbers. |
CSCdm04466 | Symptom: Occasionally, the user was logged out from the window manager prematurely when he/she exists CWM DeskTop. This bug is due to Sun Motif library problem. This bug was reported to Sun and a patch of Motif library is expected. It is no harm to the CWM database or any backend process. Workaround: Login CWM station and bring up the desktop again. |
CSCdm17822 | Symptom: Sometimes SvOvTopology core dumps when the user exits the HPOV. Workaround: None. |
CSCdm15223 | Symptom: In a repeated process of adding ports and connections then deleting connections and ports, it was found the port is missing from the CWM database after about 25 cycles. The port information then re-appears in the CWM database after a while. This problem is due to the stress loading of CWM trap processing. This is not a typical customer operation process. Workaround: Put some delays in the script. |
CSCdm17884 | Symptom: CWM image down load application cannot distinguish between backup and regular image for PXM. Workaround: Download PXM regular image and backup image separately. |
CSCdm14365 | Symptom: The "Tx Bytes Tagged DE" and "Tx Frames Tagged DE" statistics for Frame Relay Connections are not being enabled on FRSM 8-port and VHS cards in the MGX 8850 shelves. It can be selected in the SCM, but it doesn't show up on the CLI when the user enters a "dspstatparms" command. Workaround: None |
CSCdm12602 | Symptom: Sometimes when a "clrallcnf" command was performed with around 900 connections on a standalone MGX 8850 node. After the MGX 8850 is rebooted and the CWM synched again with the node, the user_connection table still has entries. Workaround: None |
CSCdm12611 | Symptom: CMGUI cannot delete RPM connections added with new command format. This is a known limitation. The CWM RPM provisioning only supports documented IOS commands. Workaround: Use supported IOS command to provision RPM. |
CSCdm11959 | Symptom: Sometimes CMGRD timeouts while adding feeder connections using ConnProxy. Workaround: Add the connections again. |
CSCdm11406 | Symptom: Occasionally, the error message "error 41" appears when the user stops the CWM core processes. Workaround: Try it again. |
CSCdm05903 | Symptom: Even though two standalone MGX 8850 nodes are being managed by CWM, both belong to same network_id according to node table. There is no way to distinguish if two MGX 8850 nodes being managed are STANDALONEs are ROUTING nodes in the same network. Workaround: None |
CSCdm02353 | Symptom: The internetworking default parameters are wrong. This is due the CWM 9.2.00 not support HI_PRIORITY feature for FRSM-VHS card. It supports only ATM-FR VBR - non real-time, ABR.FS and VHS side ABR supported. Therefore, the HI_PRIORITY is disabled. CWM 9.2.1 will fully support this feature. Workaround: None |
CSCdm02443 | Symptom: Sometimes the AIS and OAM status do not match on two CWM stations CMGUI window for some connections. This problem was not reproducible in the development lab. Workaround: None |
CSCdk68304 | Symptom: Sometimes the CMGUI displays the status of active connections as incomplete. Workaround: None |
Bug ID | Description |
|---|---|
CSCdj81530 | Symptom: Connection Management Proxy and GUI reject VPI 256-4096 for connection end point created for ASI NNI port. Condition: This bug applies when the port is an ASI NNI. Further problem description: ASI NNI port VPI range should be 0-4095 while UNI port range is 0-255. Note that this apply only to ASI NNI, AUSM NNI VPI range is 0-255. |
CSCdk30572 | Symptom: When a VPC is created through ConnProxy, it is stored in the CWM DB with vci=65537. The problem it is going to create is as follows: When user tries to do SNMP get on the endpoints of the VPC connection, or user tries to do SNMP next or SNMP walk on endpoint, it will fail at the VPC entry. Also, any type of SNMP set on the VPC connection which involves SNMP sets on endpoint entry will result in failure. Test Delay failed because it involves doing set on endpoint mib objects. The reason for the failure is that connproxy does not allow vci=65537 for any PVC endpoint entry. The allowable range is 0-65535. The fix that has been put such that connproxy allows vci value to be 65537 for VPC in the DB. This does not mean that user can create a PVC with vci=65537. This has been fixed in CWM 8.4.12, 9.1.04 & 9.2 |
CSCdk35335 | Symptom: For a PVC, get on VSVD always returns disable. When user sets VSVD, it is set properly on the node & the DB. The problem is due to mapping done between DB value and what is displayed to the user. When DB field was set to enable, it was mapped to disable. Now the mapping has been corrected.file modified: atmConnection.ec release-91 (brch-15) release-92 (brch-4) release-90 (brch-5) release-84 (brch-1.1.2.1.2.10.2.2.2.26) |
CSCdk44435 | Symptom: CWM Connection Manager does not allow the "Signalling pass through" option be changeable during pvc addition on CESM structured port. Conditions: The problem only occurs when the CESM structured port speed is greater than 64Kbps (more than 1 T1/E1 timeslot). Further Problem Description: The "Signalling pass through" option is applicable for any CESM structured port no matter the structured port included only one or more than one timeslots. |
CSCdk45518 | Symptom: After you delete a FRP port, it takes sometimes before you can add back the port, even though SNMP get on the port confirms that the port is deleted. The wait is usually more than 30 seconds. Conditions: It happens only to Frame Relay ports on IPX FRP, IGX FRM, IGX UFM, IGX UFM/U. Workaround: No workaround. However you will eventually be able to add back the port after waited for enough time (say more than 30 seconds). |
CSCdk52899 | Symptom: Some of the trap reason enums in CWMNetwork.mib were misspelled. Conditions: Typos in the file SV+Network.mib. Further Problem Description: Typos in SV+Network.mib |
CSCdk56970 | Symptom: line information is missing in trapIdString varbind of Trap 20011 Conditions: When a port fails (port NNI fail), trap 20011 is sent to client registered with RtmProxy. Varbind of this trap has only slot and port information. |
CSCdk60860 | Symptom: BIP8 ES and SES stats for BXM trunk card are missing. |
CSCdk61676 | Symptom: Trap reason in Trap 20004 returns 3056 bcc-card failed when a BNI Slot error is cleared. Conditions: When a BNI slot error is cleared (user can use scp command: "scp 1 50 3 e 1 ff" and "rcp 1 <slot>" to simulate), a trap sent out by RtmProxy Agent with trap number 20004 contains the incorrect trap reason code. Further Problem Description: SV+Network.mib contains an incorrect enum 3056 which has been updated to card active in 9.1 and 9.2. |
CSCdk67145 | Symptom: SCM will skip sending request to tftp GET the stats file for some nodes in a random pattern, hence missing stats in the database. |
CSCdk75148 | Symptom: When PortProxy add/delete port continuously, CWM becomes out of sync with MGX 8850 and MGX 8220 switches Conditions: When PortProxy continuously add/update/delete port on MGX 8850 and MGX 8220 switches. Further Problem Description: A bug in rtm code such that a timer (to send Keep Alive) got reset every time a trap is received. This caused no Keep Alive sent to MGX 8850 and MGX 8220 switches when continuous traps were received for more than 30 minutes and MGX 8850 and MGX 8220 switches de registered the CWM from trap manager. CWM became non-synchronized. |
CSCdk84925 | Symptom: CWM Connection Manager GUI does not show the preferred route properly. Conditions: The problem occurs when the preferred route contains routing node having "-" as part of the node name. |
CSCdk87382 | Symptom: When a user set the permission to "All" for ConnMgr, ConnMgr application did not allow the user to delete a connection since the "Delete" button did not get active. Conditions: Standard CWM 9.2 installation. Should have added the user to the security manager subsystem. Further Problem Description: The problem occurred because Delete was not made active for "All" permission. It is made active now. |
CSCdk92957 | Symptom: The emc process was core dumping. Conditions: The memory leak in emc was caused by the TFTP and trap processing. |
CSCdm02251 | Symptom: Trap number 25300 and 25301 were defined in CWM 9.1. New varbinds were added to these traps in CWM 9.2 without introducing new trap numbers. Some traps (1000-1004, 20009-20010) do not have good descriptions for users to apprehend. Conditions: Always. Further Problem Description: The problems have been corrected in SV+Network.mib. New trap numbers (25302, 25303) are defined to "obsolete" CWM 9.1 traps 25300 and 25301. Descriptions have been added for traps 1000-1004 and 20010. |
CSCdm07349 | Symptom: RtmProxy core dumped Conditions: When in RTM mode (doing SNMPSET nextrapsequence number and SNMPGET nexttrapPdu from RtmProxy) Further Problem Description: SNMPGET of nexttrapPDU is no longer support but code is not removed from RtmProxy causing a memory copy to a unreferenced address. Code has been changed that GET on this OID should return NO SUCH NAME |
CSCdm08471 | Symptom: The emc process was core dumping during MGX 8850 TFTP file parsing. Conditions: The emc was core dumping only with MGX 8850 having garbage ilmi entries in the TFTP file when MGX 8850 was not configured for line, port etc. |
CSCdm08489 | Symptom: Get "node-error: Can't get segment info from Data-base." error message when trying to add or delete connections using Cisco Wan Manager Connection Manager GUI or Connection |
CSCdm09863 | Symptom: Excess error message in the Statistics Collection Manager main window about timed out to a time sync host |
CSCdm11291 | Symptom: AUSM port record in CWM 92 database has no logical port while the document contains it. The CWM database document was fixed to remove the logical_port field. |
This section describes known anomalies in the 2.00 version of WANCV
Bug ID | Description |
|---|---|
CSCdk86469 | Symptom: The BPX front card images don't get the status LED colors updated for line or trunk status changes. Workaround: View the line or trunk status on the rear view of the BPX. |
This section describes anomalies corrected in the 2.00 version of WANCV
Bug ID | Description |
|---|---|
CSCdk05434 | Symptom: Some MIB variables were defined as INTEGER instead of Counter in axis.my. This caused the CiscoView monitoring dialogs to display them as numbers instead of the normal counter display. A Large number of variables was redefined as Counter instead of INTEGER in axis.my on May 18. Workaround: Use the CLI commands to view counter objects. This defect was fixed in AXIS.pkg version 1.03. |
CSCdk07747 | Symptom: The status color updates for the BNM connectors was fixed in AXIS.pkg, version 1.03. |
CSCdk18768 | Symptom: This bug is fixed as a side effect of fix for CSCdk51704. SNMP sessions were unreliable, causing the updating of the monitoring dialogs to fail intermittently. Workaround: This defect was fixed in AXIS.pkg version 2.00. |
CSCdk21445 | Symptom: This defect was fixed along with other performance and index problems in AXIS.pkg version 1.03. |
CSCdk20837 | Symptom: This defect was fixed along with other performance and index problems in AXIS.pkg version 1.03. |
CSCdk21358 | Symptom: The MIB variable slotCardStatus does not show proper value. Conditions: The IPX/IGX/BPX SNMP agent reports the above MIB variable as active when there is no backcard installed. |
CSCdm14151 | Symptom: The label for frLportPortSpeed was enhanced to clearly show the units of 100's of bits per second. This feature enhancement is available in IGX.pkg version 2.00. |
CSCdk57700 | Symptom: When using Ciscoview to configure a port on an SRM3T3 card on an AXIS shelf there is no option to configure the line type to dsx3M23 which is available at the CLI and in the mib. dsx3LineType OBJECT-TYPE SYNTAX INTEGER { dsx3CbitParity(1), g832-g804(2), dsx3M23(3) } ACCESS read-write STATUS mandatory DESCRIPTION This variable indicates the variety of DS3 C-bit or E3 application implementing this interface. The type of interface affects the interpretation of the usage and error statistics. The rate of DS3 is 44.736 Mbps and E3 is 34.368 Mbps. The dsx3ClearChannel value means that the C-bits are not used except for sending/receiving AIS. The values, in sequence, describe: TITLE: SPECIFICATION: dsx3CbitParity ANSI T1.107a-1989. Transport of SDH elements on PDH networks ITU-T G.832 ATM cell mapping into PDH ITU-T G.804" ::= { dsx3ConfigEntry 2 } |
CSCdk91145 | Symptom: Sometimes CiscoView BPX device package does not light red LED for card failure on its screen Conditions: BPX swsw 9.1.06 CiscoView 4.2(1.0) BPX4.2(2.0V) package included in SV+9.1.05patch4 for solaris Workaround: The fix is available in versions later than 2.00W |
CSCdm06256 | Symptom: UXM card with BNC connectors on IGX displays incorrectly in CiscoView 4.2 rear view Conditions: CiscoView 4.2 /usr/users/svplus/wancv/IGX.pkg Version 4.2(1.10) |
The following documents, along with this release note, comprise the Cisco WAN Manager 9.2.00 documentation set:
Cisco Connection Online (CCO) is Cisco Systems' primary, real-time support channel. Maintenance customers and partners can self-register on CCO to obtain additional information and services.
Available 24 hours a day, 7 days a week, CCO provides a wealth of standard and value-added services to Cisco's customers and business partners. CCO services include product information, product documentation, software updates, release notes, technical tips, the Bug Navigator, configuration notes, brochures, descriptions of service offerings, and download access to public and authorized files.
CCO serves a wide variety of users through two interfaces that are updated and enhanced simultaneously: a character-based version and a multimedia version that resides on the World Wide Web (WWW). The character-based CCO supports Zmodem, Kermit, Xmodem, FTP, and Internet e-mail, and it is excellent for quick access to information over lower bandwidths. The WWW version of CCO provides richly formatted documents with photographs, figures, graphics, and video, as well as hyperlinks to related information.
You can access CCO in the following ways:
For a copy of CCO's Frequently Asked Questions (FAQ), contact cco-help@cisco.com. For additional information, contact cco-team@cisco.com.
Cisco documentation and additional literature are available in a CD-ROM package, which ships with your product. The Documentation CD-ROM, a member of the Cisco Connection Family, is updated monthly. Therefore, it might be more current than printed documentation. To order additional copies of the Documentation CD-ROM, contact your local sales representative or call customer service. The CD-ROM package is available as a single package or as an annual subscription. You can also access Cisco documentation on the World Wide Web at http://www.cisco.com, http://www-china.cisco.com, or http://www-europe.cisco.com.
If you are reading Cisco product documentation on the World Wide Web, you can submit comments electronically. Click Feedback in the toolbar and select Documentation. After you complete the form, click Submit to send it to Cisco. We appreciate your comments.
Posted: Wed Mar 31 14:43:20 PST 1999
Copyright 1989-1999©Cisco Systems Inc.
