Cisco Service Control Management Suite Subscriber Manager User Guide, Rel 3.1.5 LA

Error message:

The following sequence of output appears (in response to the command p3sm --sm-status ):

>p3sm --sm-statusCould not connect to SM.

Probable cause:

The SM server has not been started.

Solution:

Use the following command to start/restart the SM server:

>p3sm --start

Error message:

The following sequence of output appears (in response to the command p3sm --sm-status ):

>p3sm --sm-statusSM is running. 
SM operational state is Failure 
Command terminated successfully

Probable cause:

The SM server restarted three times in 30 minutes due to an internal error.

Solution:

This error can happen only in a cluster setup. Check the pcube user log and the Veritas Cluster Server log for the reason for the failure that caused the reboots. Act according to the problem in the logs.

Additional operations that can be taken are:

•Use the following command to extract a support file:

>p3sm --extract-support-file -f 
../support.zip

•Send the support file to Cisco's customer support

•Use the following command to start/restart the SM server and get out of the Failure state:

>p3sm --restart

Error message:

The following sequence of output appears when running one of the installation/upgrade scripts:

>./<script-name> 
<script-name>: Cannot run this script - 
/etc/motd file exists 
<script-name>:  Please remove the file or 
move it to another location 
<script-name>: and run the script again

This error applies to the following scripts:

•install-sm.sh

•upgrade-sm.sh

•cluster-upgrade.sh

•install-vcs-agents.sh

Probable cause:

The /etc/motd file exists or has a size greater than 0.

Solution:

Remove the file or move it to another location.

Error message:

The following sequence of output appears (in response to the command ./install-sm.sh ):

>./install-sm.shinstall-sm.sh: Starting SM 
installation sequence 
install-sm.sh: Error - this script must be 
run by root - exiting.

Probable cause:

You started the installation sequence as user and not as superuser .

Solution:

Run the install-sm.sh script as superuser .

Error message:

The following sequence of output appears (in response to the command ./install-sm.sh ):

# ./install-sm.shinstall-sm.sh: Starting SM 
installation sequence 
install-sm.sh: Error - pcube user exists 
and has home /export/home/pcube, not 
/opt/pcube - remove it or use -o - exiting. 
#

Probable cause:

Your machine already has the user pcube .

Solution:

Run the installation using the -o option (overwrite), as follows:

# ./install-sm.sh -o

Error message:

The following sequence of output appears (in response to the command install-tt.sh/export/home/pcube/lib/tt ):

>install-tt.sh 
/export/home/pcube/lib/ttinstall-tt.sh: 
Starting TimesTen P-Cube installation 
sequence 
install-tt.sh: Error - This script must be 
run by root - exiting.

Probable cause:

You started the installation sequence as user and not as superuser .

Solution:

Run the install-tt.sh script as superuser .

Note that the TimesTen directory name given, /lib/tt/ , must be relative to the (pcube) user directory. For example, if the user directory is pcube, install TimesTen in /export/home/pcube/lib/tt/.

Error message:

The following sequence of output appears (in response to the command p3sm --sm-status ):

>p3sm --sm-statusSM is running. 
SM operational state is XXX 
Error - Times-Ten DB is not setup 
correctly: 
[TimesTen][TimesTen 5.0.35 CLIENT]Cannot 
find the requested DSN 
(PCube_SM_Repository_CS) in ODBCINI

Probable cause:

The TimesTen Client DSN is not configured correctly in file ~pcube/.odbc.ini.

Solution:

Ensure that file ~pcube/.odbc.ini contains the following:

[ODBC Data Sources] 
PCube_SM_Repository_CS=TimesTen 5.0 Client 
Driver

Error message:

The following sequence of output appears (in response to the command p3sm --sm-status ):

>p3sm --sm-statusSM is running. 
SM operational state is XXX 
Error - Times-Ten DB is not setup 
correctly: 
[TimesTen][TimesTen 5.0.35 ODBC Driver]Data 
source name not found and no default driver 
specified

Probable cause:

The TimesTen Client DSN is not configured correctly in file ~pcube/.odbc.ini.

Solution:

Ensure that file ~pcube/.odbc.ini contains the following:

[PCube_SM_Repository_CS] 
TTC_SERVER_DSN=PCube_SM_Repository

Ensure that file /var/TimesTen/sys.odbc.ini contains the following:

[ODBC Data Sources] 
PCube_SM_Repository=TimesTen 5.0 Driver 
[PCube_SM_Repository] 
Driver=__TTDIR__/TimesTen/pcubesm22/lib/lib
tten.so 
DataStore=__VARDIR__/pcube_SM_Repository

Error message:

The following sequence of output appears (in response to the command p3sm --sm-status ):

>p3sm --sm-statusSM is running. 
SM operational state is XXX 
Error - Times-Ten DB is not set up 
correctly: 
[TimesTen][TimesTen 5.0.35 CLIENT]Unable to 
connect to data source 
(DSN: pcube_SM_Repository_CS; Network 
Address: X.X.X.X; Port Number: XXX): 
This operation has Timed Out. Try 
increasing your ODBC timeout attribute 
or check to make sure the target TimesTen 
Server is running

Probable cause:

The following causes are possible:

•The address of the Server DS is incorrect.

•The port of the Server DS is incorrect.

•TimesTen is not active.

Solution:

The Service Control solutions for the above causes are:

•(The address of the Server DS is incorrect.) Ensure that file ~pcube/.odbc.ini contains the following:

•TTC_SERVER=127.0.0.1

•(The port of the Server DS is incorrect.) On a default installation, ensure that file ~pcube/.odbc.ini does not contain "TTC_SERVER_PORT".

On a non-default installation, ensure that file ~pcube/.odbc.ini does contain "TTC_SERVER_PORT=Non-default-port".

•(TimesTen is not active.) Run the following command:

•~pcube/lib/tt/TimesTen/pcubesm22/bin/ttStatus

If TimesTen is not working, re-install TimesTen.

If the above solutions do not work, please refer to the TimesTen manual.

Error message:

The following sequence of output appears (in response to the command p3sm --sm-status ):

>p3sm --sm-statusSM is running. 
SM operational state is XXX 
Error - Times-Ten DB is not setup 
correctly: 
[TimesTen][TimesTen 5.0.35 ODBC 
Driver][TimesTen]TT0836: 
Cannot create data store shared-memory 
segment, 
error 1455 -- file "db.c", lineno 6289, 
procedure "sbDbConnect()"

Probable cause:

There is not enough memory for creating TimesTen's in-memory database.

Solution:

Do all of the following:

•Ensure that the Unix machine has at least 1024 MB of memory installed.

•Ensure that the configured memory size parameters ( PermSize and TempSize ) specified in file /var/TimesTen/sys.odbc.ini are less than the total amount of memory installed in the machine.

•For Solaris, ensure that the maximum shared memory (parameter shmsys:shminfo_shmmax ) specified in file /etc/system is less than the total amount of memory installed in the machine.

•For Red Hat, ensure that the maximum shared memory (parameter kernel.shmmax ) specified in file /etc/sysctl.conf is less than the total amount of memory installed in the Linux Machine.

Error message:

The following sequence of output appears (in response to the command p3sm --sm-status ):

>p3sm --sm-status 
SM is running. 
SM operational state is XXX. 
Error - Times-Ten DB is not setup 
correctly: 
[TimesTen][TimesTen 5.0.35 ODBC 
Driver]Overflow in converting data store or 
log file size from megabytes to bytes, 
or in converting log buffer size from 
kilobytes to bytes

Probable cause:

The memory definitions of DSN are incorrect.

Solution:

Ensure that the configured permanent memory size and log file size (parameters PermSize and LogSize ) are less than the total amount of memory and of disk space specified in file /var/TimesTen/sys.odbc.ini.

Error message:

The following sequence of output appears (in response to the command p3sm --sm-status ):

>p3sm --sm-status 
SM is running. 
SM operational state is XXX 
Error - Times-Ten DB is not setup 
correctly: 
[TimesTen][TimesTen 5.0.35 ODBC 
Driver][TimesTen]TT0925: Cannot create data 
store semaphores (Invalid argument) -  
file "db.c", lineno 5124, procedure 
"sbDbCreate()", sqlState: 08001, errorCode: 
925

Probable cause:

TimesTen was unable to create the data store semaphores that are defined in the kernel configuration file (Solaris: /etc/system for Solaris; for Red Hat: /etc/sysctl.conf for Red Hat).

Solution:

Do all of the following:

•Ensure that the machine has at least 1024 MB of memory installed.

•Reboot the machine after the first time that TimesTen is installed.

•Verify the contents of the system (kernel) configuration file:

–For Solaris, ensure that file /etc/system contains the following:

•semsys:seminfo_semmni = 20
semsys:seminfo_semmsl = 100
semsys:seminfo_semmns = 2000
semsys:seminfo_semmnu = 2000

–For Red Hat, ensure that file /etc/sysctl.conf contains the following:

•*---- Begin settings for TimesTen
kernel.sem = "SEMMSL_250 SEMMNS_32000 SEMOPM_100 SEMMNI_100
*---- End of settings for TimesTen

Error message:

The following sequence of output appears (in response to the command p3sm --sm-status ):

>p3sm --sm-status 
SM is running. 
SM operational state is XXX. 
Error - Times-Ten DB is not setup 
correctly: 
[TimesTen][TimesTen 5.0.35 ODBC 
Driver][TimesTen]TT0845: Cannot read data 
store file. 
OS-detected error: Error 0 -- file "db.c", 
lineno 6320, procedure "sbDbConnect()"

Probable cause:

TimesTen was unable to read the data store file, probably due to an error during the installation. This error occurs when installing a TimesTen application on top of an existing TimesTen, without having first uninstalled the old TimesTen database.

Solution:

Do the following:

1. Remove the database, by using the SM p3db CLU with the following commands:

>p3db -destroy-rep-db >p3db 
-destroy-local-db

2. Uninstall TimesTen with the following commands:

>su Password: # 
~pcube/lib/tt/TimesTen4.5/32/bin/setup.
sh -uninstall

3. Re-install TimesTen, either by running the SM install-tt.sh script or by using the installation files supplied by TimesTen.

Error message:

The following sequence of output appears in the SM log (while using the SM APIs):

java.io.IOException: Failure in putting 
subscriber 
45977166__00:50:bf:97:c1:b2 : 
[TimesTen][TimesTen 5.0.35 ODBC 
Driver][TimesTen]TT0802: Data store space 
exhausted -- file "blk.c", lineno 
1571, procedure "sbBlkAlloc"

Probable cause:

The TimesTen database has already reached its maximum capacity, which caused the operation of adding a new subscriber to the database to fail.

Solution:

Usually, doing just one of the following is sufficient:

•Reduce the number of the subscribers handled by the SM (of course, this solution is not always possible).

•Configure the system to support a larger number of subscribers. Note that this solution may require editing one or more of the TimesTen configuration files discussed in Introduction as well as rebooting the machine.

•Move the SM to a more powerful machine; this could be a faster CPU (or more CPUs), a larger disk, more RAM, etc.

For help and guidance in implementing the last two solutions, please contact Cisco Technical Support. See Obtaining Documentation and Submitting a Service Request, page -xvii.

Error message:

The following sequence of output appears (in response to the command p3net --connect ):

>p3net --connect --ne-name=YYYY Error - 
failed to connect to element 'YYYY' 
Operation timed out: connect

Probable cause:

The following causes are possible:

•The IP address is incorrect.

•The element YYYY is down.

Solution:

The Service Control solutions for the above causes are:

•Ensure that the IP address is correct.

•Ensure that the element YYYY is online and is connected via its management port.

Error message:

The following sequence of output appears (in response to the command p3net --show-ne ):

>p3net --show --ne-name=se0 
Network Element Information: 
============================ 
Name:        YYY 
Description: testing element 
Host:        X.X.X.X 
Ip:          X.X.X.X 
Port:        14374 
Status:      Connection down (Failure in 
connecting to agent on host, Connection 
refused: connect) 
Type:        SCE1000 
Domain:        
smartNET.policy.unitTestSubscribers 
Subscriber Management: Not Active

Probable cause:

The following causes are possible:

•The IP address is incorrect.

•The element YYYY is down.

Solution:

The Service Control solutions for the above causes are:

•Ensure that the IP address is correct.

•Ensure that the element YYYY is online and is connected via its management port.

•Ensure that the PRPC adapter is online on the port that the status indicates.

Error message:

The following sequence of output appears (in response to the command p3net --show-ne ):

>p3net --show --ne-name=se0 
Network Element Information: 
============================ 
Name:        YYY 
Description: testing element 
Host:        X.X.X.X 
Ip:          X.X.X.X 
Port:        14374 
Status:      Connection ready 
Type:        SCE1000 
Domain:        
smartNET.policy.unitTestSubscribers 
Subscriber Management: Not Active

Probable cause:

The Subscriber Management field indicates whether the SM successfully performed SM-SCE subscriber synchronization. If the value of the field is Not Active , it is possible that the SM failed to synchronize the SCE.

Solution:

One possible solution is to force SM-SCE resynchronization by using the CLU command p3sm --resync .

Error message:

The following sequence of output appears (in response to the command p3subsdb --import ):

>p3subsdb --import 
--file=/export/home/pcube/XXX.csv 
Error - Failed to validate the file XXX.csv 
See import-results.txt for detailed errors 
description. 
>cat import-results.txt  
x.csv:1: expected 2 items but got 4 items. 
1 subscribers, 1 errors. 
NO APPLICATION INSTALLED, MAKE SURE TO 
INSTALL PQI BEFORE IMPORTING CM WITH 
TUNEABLES.

Probable cause:

You tried to import a four-field csv file to the SM, but no application (SCA BB) was installed.

For example, the following csv file for a SCA BB application contains four fields:

# CSV line format: subscriber-id, domain, 
mappings, package-id 
JerryS,subscribers,80.179.152.159,0 
ElainB,subscribers,194.90.12.2,3

However, the default definition file that defines csv file parsing rules contains only two fields: name and ip mapping.

Solution:

Do one of the following:

•Install an application (SCA BB) on the SM (for details, see the Cisco Service Control Application for Broadband User Guide ).

•Import a csv file that has just two fields.

Error message:

The following sequence of output appears (in response to command p3cable --import-cm ):

>p3cable --import-cm 
--file=/export/home/pcube/XXX.csv 
Importing cable modems ... 0% 
Importing cable modems ... 100% 
Error - Errors during import from 
'H:\work\Mng\dev\install\ems\bin\win32\x.cs
v': 
Imported 1 CM(s).  1 Error(s). 
See cm-import-results.txt for detailed 
errors description. 
>cat cm-import-results.txt 
x.csv:1: expected 2 items but got 4 items. 
1 cable modem(s); 1 error(s). 
NO APPLICATION INSTALLED, MAKE SURE TO 
INSTALL PQI BEFORE IMPORTING CM WITH 
TUNEABLES.

Probable cause:

You tried to import a four-field csv file to the SM, but no application (SCA BB) was installed.

For example, the following csv file for a SCA BB application contains four fields:

# CSV line format: subscriber-id, domain, 
mappings, package-id 
JerryS,subscribers,80.179.152.159,0 
ElainB,subscribers,194.90.12.2,3

However, the default definition file that defines csv file parsing rules contains only two fields: name and ip mapping.

Solution:

Do one of the following:

•Install an application (SCA BB) on the SM (for details, see the Cisco Service Control Application for Broadband User Guide ).

•Import a csv file that has just two fields.

Error message (1):

Probable cause:

[SCE.XXX] section cannot contain white spaces (SCE name cannot contain white space).

Solution:

Remove the white spaces.

Error message (2):

Probable cause:

Configuration cannot contain two SCEs with the same IP address.

Solution:

Change the IP address of one of the SCEs.

Error message (3):

Probable cause:

The section <domain name>includes, under the elements property, an SCE that is not defined in an [SCE.XXX] section.

Solution:

Add the missing [SCE.XXX] section to the file.

Error message (4):

Probable cause:

Same SCE cannot belong to more than one domain.

Solution:

Remove the SCE from all but one of the domains.

Error message (1):

Probable cause:

[Domain.XXX] section cannot contain white spaces (Domain name cannot contain white space).

Solution:

Remove the white spaces.

Error message (2):

Probable cause:

The alias name cannot start with CM_ because this is the prefix of hidden domain generated by the SM when working with CMs (see CPE as Subscriber in Cable Environment, page C-1 ).

Solution:

Use a different prefix for the alias.

Error message (3):

Probable cause:

Each alias can appear in only one [Domain.XXX] section.

Solution:

Alias mentioned in the error message should be removed from all but one [Domain.XXX] section.

Error message (4):

Probable cause:

Domain mentioned in given section does not have a [Domain.XXX] section.

Solution:

Domain mentioned in error message should be given a [Domain.XXX] section.

Error message (5):

Probable cause:

Properties in [Domain.XXX] sections do not have integer values.

Solution:

Properties mentioned in error message should be given integer values.

Error message (6):

Probable cause:

Property defined in [Domain.XXX] section is not found in properties list defined by installed application.

Solution:

Delete the properties mentioned in the error message (or define them in the installed application).

Error message (7):

Probable cause:

This is actually a warning: its purpose is to warn that the user removed from the p3sm.cfg file a domain that contained SCEs (which probably contained subscribers, etc.), and that the user will lose all subscriber data relevant for that domain. This warning appears only after the CLU command p3sm --load-config is activated.

Solution:

To avoid this warning, use the --ignore-warnings option.

Error message:

Probable cause:

This is actually a warning: it is displayed after the CLU command p3sm -load-config is activated when the PRPC configuration in the p3sm.cfg file has been changed.

Solution:

Use the --ignore-warnings option to complete the operation.

Error message:

Probable cause:

<nasID>is not unique.

Solution:

Change one of the <nasID>so that both are unique.

Error message (1):

Probable cause:

Property written in the p3sm.cfg file is unknown to the SM. Maybe the name is misspelled or the property belongs in a different section.

Solution:

Ensure that the name is spelled correctly and that the property resides in the correct section.

Error message (2):

Probable cause:

The section written in the p3sm.cfg file is unknown to the SM. Maybe the name is misspelled.

Solution:

Ensure that the section name is spelled correctly.

Error message (3):

Probable cause:

Value of the property is invalid. The <values range>field contains the valid values.

Solution:

Specify any valid value for the property.

Error message (4):

Probable cause:

The property <property name>is mandatory and must appear in the section <section name>.

Solution:

Set a value for the requested property in the specified section.

Error message (5):

Probable cause:

The value is an invalid IP address.

Solution:

Specify a valid IP address.

Error message (6):

Probable cause:

Value of the <property name>is empty; for example, prop=

Solution:

Specify a non-empty value for the property.

Error message (7):

Probable cause:

The section with <section name>appears more than once. This error is most likely to occur for the [SCE.XXX] and [Domain.XXX] sections.

Solution:

Use the specified section name only once.