Applies to:
Siebel System Software - Version: 7.7.2 [18325] and later [Release: V7 and later ]
HP-UX Itanium
Product Release: V7 (Enterprise)
Version: 7.7.2 [18325]
Database: Oracle 9.2
Application Server OS: HP-UX 11.11
Database Server OS: HP-UX 11.11
This document was previously published as Siebel SR 38-1967171771.
Symptoms
SBL-GEN-05004, SBL-ICF-00018
Our database file was filled up around 1:07AM last night possibly
causing our Siebel Server to fail. While troubleshooting this morning,
we’ve shutdown the Siebel Server and Gateway Server and received (hand
translated from Chinese) “Not Started: Unable to copy service item files
into memory” when attempting to restart Gateway server.
The database size has since been extended, however, we still continue to receive this error.
We have tried the following:
- Delete and rebuild svc* file
- Verify all files are owned by Siebel user (and have read write permission to siebns.dat)
- Tried using an older version of siebns.dat
- Verify port 2320 is available
I have uploaded:
- screen shot of the error (in Chinese)
- Gateway logs
- enterprise server log
Any suggestions would be much appreciated
Cause
Configuration/ Setup
Solution
For the benefit of other users, the customer received the following
error message when attempting to restart their Siebel Gateway Server
“Not Started: Unable to copy service item files into memory”
Following
the failed restart, they recreated the *.svc files by running a siebctl
command but still could not start the Siebel Gateway Server and
received “NameSrvr - [5018] (No message)” errors in the NameSrvr.log.
Summary:
Siebel
does not recommend that customers should re-create the Siebel Gateway
or Server services if there are problems starting any service. The
reason Siebel does not recommend re-creating the Siebel Services is
because they could be created incorrectly which then cause other issues.
Customers are advised to raise a Service Request with Siebel Technical
Support if assistance is required diagnosing why a Siebel service failed
to start.
Further research found the siebctl syntax the customer
used to create the Gateway Server service is the one specified in
Service Request #38-1960105998 to create the Siebel Server service. The
steps used to resolve this Service Request was:
1) Ensure the Siebel Server Service and Web Server services are offline
2) Run gtwysrvr\siebenv.sh
3) Stop the Gateway Server service (stop_ns)
4) Copy svc.gtwyns. in $SIEBEL_ROOT/gtwysrvr/sys to backup.svc.gtwyns.old
5) Delete the svc.gtwyns. file
6) Run
the command siebctl -S gtwyns -a -g "/f $SIEBEL_ROOT/sys/siebns.dat /t
2320" substituting the port number for the Gateway Server service in
place of 2320.
7) Start the Siebel Gateway service (start_ns)
8) Type list_ns
Regards,
Applies to:
Siebel System Software - Version: 7.7.2.4 [18365] and later [Release: V7 and later ]
IBM AIX on POWER Systems (64-bit)
Product Release: V7 (Enterprise)
Version: 7.7.2.4 [18365]
Database: IBM DB2 8.2
Application Server OS: IBM AIX 5L 5.2
Database Server OS: IBM AIX 5L 5.2
This document was previously published as Siebel SR 38-3265850867.
Symptoms
SBL-GEN-05004, SBL-ICF-00018
We are in the process of setting up a disaster recovery
environment. While installing the first server we are running into
issue. I have installed the gateway server without issue. I have
installed the Siebel server 5 times now and get the same error every
time we try to start the server.
bash-2.05b$ start_server all
Siebel Server "rirdrs7gw01" (Enterprise "siebelpro")
OSDmmap: mmap(addr = 0x00000000, len = 3888496, prot = 0x00000003, flags = 0x00000001) failed: Invalid argument
Error 1300003 in OSDInit
bash-2.05b$ start_server all
Siebel Server "rirdrs7gw01" (Enterprise "siebelpro")
OSDmmap: mmap(addr = 0x00000000, len = 3888496, prot = 0x00000003, flags = 0x00000001) failed: Invalid argument
Internal:
The parameter values in the header of file
/siebelgw/siebsrvr/sys/osdf.siebelpro.rirdrs7gw01 do not correspond to
the current parameter values:
nFileSize 0 3888496
nSiebNLatch 0 5039
nSiebLatch 0 1000
nSiebRWLock 0 503
nSiebNSema 0 1021
nSiebLatchNameLen 0 256
nSiebSemaNameLen 0 64
nSiebProcExit 0 256
Error 1300003 in OSDInit
As
you can see, we start the server, the startup fails but an osdf file is
created. if we try to start again we get the same error as the first
time but with an Internal error as well.
Cause
The filesystems were mounted with CIO option.
CIO is concurrent IO -
it's a mount option for the filesystem. It has some implications about
what can be done with the filesystem, including hindering the ability to
run executables from the filesystem.
Solution
Message 1
For the benefit of other readers, the troubleshooting steps and resolution are summarized below:
1) The customer was advised to run the following:
- Run siebsrvr\siebenv.sh
- Make sure that $SIEBEL_ROOT is correctly defined.
- Stop the Siebel Server service (stop_server ALL)
- Remove *.osdf in the $SIEBEL_ROOT/sys
- Remove the svc.* files in the $SIEBEL_ROOT/sys (after backing it up)
- Run
the siebctl command below but with the actual names of the Siebel
Enterprise, Server and Gateway Server names substituted for the
variables.
siebctl -S siebsrvr -i "$SIEBEL_ENTERPRISE:$SIEBEL_SERVER" -a -g "-g $SIEBEL_GATEWAY -e $SIEBEL_ENTERPRISE -s $SIEBEL_SERVER"
- Start the Siebel Server service (start_server ALL)
This did not resolve the issue.
2) We checked that the customer’s environment met all the requirements in the following SupportWeb documents:
a) System Requirements and Supported Platforms, Version 7.7;
b)
Siebel 7.7 Bookshelf > Performance Tuning Guide > Tuning UNIX
Operating Systems for Performance > Tuning Siebel eBusiness
Applications for AIX >Tuning Kernel Settings for AIX
3) The test in the following SupportWeb Knowledge Item was successful:
FAQ 1113: How Do You Test Connectivity Through ODBC Data Sources Created by Siebel Systems?
4)
After reviewing the truss output from running start_server, Siebel
Technical Support recommended that the customer contact IBM AIX Support
who identified that the filesystems were mounted with CIO option.
When the filesystems were remounted without CIO option, the Siebel Server started successfully.
Best Regards,
No comments:
Post a Comment