Siebel Server Infrastructure Administration

This chapter describes how to administer the Siebel Server infrastructure and system management components. It includes the following topics:

About Server Request Broker (SRBroker)

Server Request Broker (alias SRBroker) is an interactive ode Siebel Server component that belongs in the System Management component group. By default, one SRBroker is started for each Siebel Server. SRBroker handles client component requests by acting as a request router. For example, if a client makes a request to a Siebel Server for a component that is not running on that Siebel Server, then the request is routed to another Siebel Server that is running the requested component.

Siebel Server requests from clients that have no end point get stored in the database until the request is completed. SRBroker works with the Server Request Processor (alias SRProc). For more information about this component, see About Server Request Processor (SRProc).

SRBroker also controls how many component requests by clients can be serviced by a Siebel Server at one time. Each client connection and component connection counts as one task. The number of tasks that can be handled by a single SRBroker is determined by the Maximum Tasks (alias MaxTasks), Maximum MT Servers (alias MaxMTServers), and Minimum MT Servers (alias MinMTServers) component parameters. Keep MaxMTServers and MinMTServers at their default value of 1 for SRBroker. For more information about these parameters, see Siebel Performance Tuning Guide. For information about how to set the number of tasks for SRBroker, see Configuring Tasks for Server Request Broker.

Do not configure run-time parameters for SRBroker. If you have to support more client and component connections, then increase the number of tasks that can be handled by the SRBroker component.

Configuring Tasks for Server Request Broker

This topic describes how to configure the number of tasks for the Server Request Broker (alias SRBroker) component. For more information about SRBroker, see About Server Request Broker (SRBroker).

To change the number of tasks that can be handled by Server Request Broker

Navigate to the Administration - Server Configuration, and then the Servers view.
In the Siebel Servers list, select the Siebel Server of interest.
Click the Components view tab.
In the Components list, query for Server Request Broker (alias SRBroker) in the Component field.
Select the Parameters view tab below the Components list.
In the Parameters list, query for the Maximum Tasks (alias MaxTasks) parameter.
In the Value on Restart field, type in the number of tasks.
The default value is 100. For more information about this parameter, see the parameter definition in Generic Parameters. For more information about values to set this parameter, see Siebel Performance Tuning Guide.
For changes to take effect, restart the Siebel Server system service.
For more information about restarting the Siebel Server system service, see Administering the Siebel Server System Service.

About Server Request Processor (SRProc)

The Server Request Processor (alias SRProc) and the Server Request Broker (alias SRBroker) components are jointly responsible for the processing of both synchronous and asynchronous requests from a variety of Siebel Server components. SRProc is a background mode component that handles requests between the Siebel Server and the database. There can only be one instance of SRProc for each Siebel Server. The following components rely on a functioning SRProc and SRBroker:

Assignment Manager
Communications Manager
Enterprise Application Integration
EIM
Field Service (all components)
Interactive Assignment
Workflow Management

If either SRBroker or SRProc become unavailable for any reason, then the ability to execute intercomponent requests is severely affected. The request mechanism (component jobs) of the Server Manager GUI relies on a functioning SRBroker and SRProc to schedule and process requests. However, the server manager command-line interface program bypasses this request mechanism permitting the user to start (but not schedule) a component task by using the command-line interface if either or both the SRBroker or SRProc components are unavailable (or, alternatively, restarting the SRBroker or SRProc components). For more information about using the server manager command-line interface program, see Using the Siebel Server Manager Command-Line Interface.

Several parameters are available that ensure that these components automatically restart in the event of a failure, so the components experience as little downtime as possible. For information about the parameters Default Tasks (alias DfltTasks) and Auto-Restart (alias AutoRestart), see Siebel Server Components and Parameters.

About Siebel Connection Broker (SCBroker)

The Siebel Connection Broker (alias SCBroker) component is a background mode server component that provides intraserver load balancing. By default, it is always enabled and online. At least one instance of SCBroker must be running on any Siebel Server hosting interactive components.

Note: If a Siebel Server hosts only batch mode components, then SCBroker can be disabled to prevent it from listening on a TCP port.

SCBroker listens on a configurable, static port for new connection requests from the Siebel Application Interface. The parameter Static Port Number (alias PortNumber) defines the port that SCBroker monitors. You specify this port number for SCBroker in the Siebel Management Console, when you configure the Siebel Server and when you configure the Siebel Application Interface. The default value is 2321. After a request is received, SCBroker distributes it to the appropriate instance of an Application Object Manager running on the Siebel Server.

The SCBroker component uses a connection forwarding algorithm to forward the socket to the Application Object Manager processes. The component parameter Connection Forward Algorithm for SCBroker (alias ConnForwardAlgorithm), which is a hidden parameter, has two possible settings:

LL (for least-loaded algorithm)
RR (for round-robin algorithm)

LL is the default value. The least-loaded algorithm balances incoming Application Object Manager login requests. It identifies which Application Object Manager process is handling the least number of tasks and assigns that process to handle the session. The round-robin algorithm distributes all of the Application Object Manager login requests to the next Application Object Manager process in a round-robin fashion, that is, equal loads distributed in order and without priority.

For more information about SCBroker and about load balancing, see Siebel Deployment Planning Guide and the Siebel Installation Guide for the operating system you are using.

About Other System Management Components

This topic describes the other server components that make up the System Management (alias System) and the Auxiliary System Management (SystemAux) component groups. It includes the following information:

About Server Tables Cleanup (SvrTblCleanup)

Server Tables Cleanup (alias SvrTblCleanup) is a component that deletes the completed and expired Server Request records. The parameter Sleep Time (alias SleepTime) controls how often the cleanup occurs. The default value for Sleep Time is 300 seconds (5 minutes).

By default, the Server Tables Cleanup component is enabled on all of the Siebel Servers in your Siebel Enterprise Server. However, you only have to run one instance of this component, because it deletes the completed and expired server request records for all of the Siebel Servers in the Siebel Enterprise Server from the S_SRM_REQUEST table in the Siebel database. For this reason, you can disable other instances of this component on other Siebel Servers in the Siebel Enterprise Server.

The Server Tables Cleanup component is part of the Auxiliary System Management component group.

About Siebel Administrator Notification (AdminNotify)

Siebel Administrator Notification (alias AdminNotify) is a batch mode component that notifies the Siebel administrator when problems are detected on the Siebel Server or its running components. For more information about component notification, see About System Alert Notification.

The Siebel Administrator Notification component is part of the Auxiliary System Management component group.

About Siebel Server Scheduler (SrvrSched)

Siebel Server Scheduler (alias SrvrSched) is a background mode component supports the running of the Siebel Server and server components by spawning component processes as requested by the Siebel Server. No entries for the Siebel Server Scheduler component appear in the Siebel Server log file. Instead, entries appear for the component for which Siebel Server Scheduler spawns a process. A network message eventually assigns the process to the component it is supposed to run. The process loads the component and runs it. The Siebel Server Scheduler component is part of the System Management component group.

Caution: Do not modify the Siebel Server Scheduler component without instructions from Global Customer Support. For help modifying Siebel Server Scheduler, create a service request (SR) on My Oracle Support. Alternatively, you can phone Global Customer Support directly to create a service request or get a status update on your current SR. Support phone numbers are listed on My Oracle Support.

Note: Because of the nature of the Siebel Server Scheduler component, the Siebel Server Scheduler task IDs that appear in the log files do not have an appropriate entry in the Administration - Server Management screen. For the same reason, no entry appears in the Components view of the Administration - Server Configuration screen for this component.

Backing Up and Restoring the Siebel Gateway Registry

This topic describes how to back up the Siebel Gateway registry and restore the Siebel Gateway registry from a backup. For more information about the Siebel Gateway registry, see About the Siebel Gateway. See also the Siebel Installation Guide for the operating system you are using.

This topic contains the following information:

Backing Up the Siebel Gateway Registry

To back up the Siebel Gateway registry, perform the steps in the procedure that follows. It is strongly recommended to back up the Siebel Gateway registry periodically for safety reasons, so that, if something goes wrong, you will be able to restore the Siebel Gateway registry from a backup, as described in Restoring the Siebel Gateway Registry.

It is recommended to back up the Siebel Gateway registry at regular intervals, and before deploying Siebel Gateway clustering into your topology. Retain the backups, which you can use to restore the Siebel Gateway registry data in scenarios identified as caused by corruption in the Siebel Gateway registry. One of the known scenarios where a restore is required is if Siebel Gateway cluster deployment fails and the administrator experiences inconsistency in accessing profiles and deployments in Siebel Management Console. Where you have already deployed Siebel Gateway clustering, then you create the backup for only one of the deployed instances. For more information about Siebel Gateway clustering, see the Siebel Installation Guide for the operating system you are using.

This topic is part of Backing Up and Restoring the Siebel Gateway Registry.

To back up the Siebel Gateway registry

(Optional) Shut down the Siebel CRM deployment, as described in Starting and Shutting Down a Siebel CRM Deployment.
For the instance of Siebel Gateway for which you are backing up the registry, do the following:
1. Navigate to the directory GTWYSRVR_ROOT\zookeeper\conf.
2. Note the dataDir path in the zoo1.cfg file.
3. Navigate to the path noted in the previous step (usually, this is GTWYSRVR_ROOT\zookeeper).
4. Copy the version-2 directory to another name, such as version-2_preclusterdeploy or a name that includes the current date.
(If necessary) Restart the Siebel CRM deployment, as described in Starting and Shutting Down a Siebel CRM Deployment.

Restoring the Siebel Gateway Registry

To restore the Siebel Gateway registry from the backup, perform the steps in the procedure that follows.

Caution: Do not perform the task of restoring the Siebel Gateway registry unless it is strictly necessary to resolve a problem. You must have first created a valid backup, as described in Backing Up the Siebel Gateway Registry.

Where you have already deployed Siebel Gateway clustering, then you restore the same backup for all of the deployed instances. For more information about Siebel Gateway clustering, see the Siebel Installation Guide for the operating system you are using.

This topic is part of Backing Up and Restoring the Siebel Gateway Registry.

To restore the Siebel Gateway registry

Shut down the Siebel CRM deployment, including all Siebel Gateway cluster nodes, as described in Starting and Shutting Down a Siebel CRM Deployment.
For each deployed instance of Siebel Gateway, including all Siebel Gateway cluster nodes, do the following:
1. Navigate to the directory GTWYSRVR_ROOT\zookeeper\conf.
2. Note the dataDir path in the zoo1.cfg file.
3. Navigate to the path noted in the previous step (usually, this is GTWYSRVR_ROOT\zookeeper).
4. Remove the version-2 directory (if it exists).
5. Into this same location, copy the backup directory that you created in Backing Up the Siebel Gateway Registry.
6. Rename this directory, such as to rename version-2_preclusterdeploy to version-2.
Restart the Siebel CRM deployment, as described in Starting and Shutting Down a Siebel CRM Deployment.

Administering the Siebel File System

This topic provides background information and administration tasks applicable to the Siebel File System. This topic contains the following information:

About the Siebel File System

The Siebel File System is a shared directory or a set of directories that contain the physical files used by the Siebel clients. All of the File System directories must be network-accessible to the Siebel Server. You can create each File System directory on a server computer where you have installed a Siebel Server, or on another network server that can share the directory, so that the File System directories are available to the Siebel Server. For more information about the requirements for networked file systems, see the third-party documentation.

To gain access to files, Web clients connect to the appropriate Siebel Server to request file uploads or downloads. The Siebel Server then accesses the Siebel File System using the File System Manager (alias FSMSrvr) component. File System Manager processes these requests through interaction with the Siebel File System directories. For more information about data transfer, see About the File System Upload and Download Process.

At the server component level, most server components, including all Application Object Managers, access the Siebel File System through the File System Manager server component when administering attachments. Application Object Managers, however, access the Siebel File System directly when saving user preference files.

When using Siebel Developer Web Client for administrative tasks, you might want to connect directly to the Siebel File System without going through the File System Manager.

Files stored in the Siebel File System are compressed at the Siebel Server-level and appended with the extension .saf. (The file size displayed in the GUI represents the size of the compressed .saf file, not the actual file size.) The Siebel File System storage locations of the compressed files are set by the enterprise parameter Siebel File System (alias FileSystem). For more information about this parameter, see Siebel Enterprise Server Parameters. The files stored in the Siebel File System are not directly accessible by users and must be retrieved (and decompressed) by the user through normal Siebel Web Client operations only.

Files stored in the Siebel File System are always compressed. That is, you cannot disable the compression feature of the Siebel File System.

You can exclude certain types of files from being saved into the Siebel File System, based on their file extensions. For more information about setting system preferences to enable this feature and to specify the file extensions to be excluded, see Siebel Security Guide.

Note: Virus checking is not supported within the Siebel File System.

This topic is part of Administering the Siebel File System.

Related Books

For information about creating the Siebel File System, see the Siebel Installation Guide for the operating system you are using.

For information about deployment options for the Siebel File System, see Siebel Deployment Planning Guide.

For information about securing the Siebel File System, see Siebel Security Guide.

About the File System Upload and Download Process

This topic describes what happens when files are uploaded to or downloaded from the Siebel File System.

This topic is part of Administering the Siebel File System.

About the File System Upload Transfer Process

When a user saves a file or attachment to be written to the Siebel File System, the file is copied from the user's hard drive and transferred to the Siebel Server. The data transfer protocol for file transfer matches that of the Web client browser to Siebel Application Interface, for example, HTTP or HTTPS. The File System Manager (alias FSMSrvr) component compresses the file, and then stores the compressed file in the Siebel File System. The compression and naming convention of the files is automated by FSMSrvr.

About the File System Download Transfer Process

When a Siebel application user accesses a file (for example, a PDF document) that is stored in the Siebel File System, a file or attachment download request is received by the FSMSrvr component of the Siebel Server. This component interacts with the Siebel File System directories to retrieve and send the compressed file back to the user's Web browser. As with the file upload process, the data transfer protocol for file transfer matches that of the Web client browser to Siebel Application Interface. The compressed file is decompressed by the user's Web browser, where the file can be reviewed or saved.

In some cases, the file is decompressed by the FSMSrvr component and sent to the user's Web browser in an uncompressed format. An uncompressed file is sent back to the Web browser in the following cases:

The parameter Compressed File Download (alias CompressedFileDownload) is set to False. You configure this parameter in the Siebel Server Component Parameters view. For information about this task, see Configuring Siebel Server Component Parameters.
The CompressedFileDownload parameter is set to False in the application configuration file for a Siebel Mobile Web Client. (If this parameter is not already in the configuration file, then you can add it to the [InfraUIFramework] section of the file.)
The Web browser does not support compressed files, which is determined by looking at the request header.
The file has the extension .zip, .z, .tgz, .gz, .gif, .jpg, or .jpeg.

Partitioning the Siebel File System

This topic describes how to perform the optional task of partitioning the Siebel File System.

This topic is part of Administering the Siebel File System.

This topic contains the following information:

About Partitioning the Siebel File System

This topic describes how to perform the optional task of partitioning the Siebel File System.

This topic is part of Partitioning the Siebel File System.

Partitioning the Siebel File System allows you to store larger volumes of data on multiple devices. The original Siebel File System might use a single directory or might already use multiple directories on multiple devices or partitions.

In general, the term partitioning, as used in this topic, refers to running the sfspartition utility, which is provided for the purpose of adding one or more network directories to an existing Siebel File System and distributing the existing files among all of the participating directories. You can add each new directory on the same device as an existing directory or add it on a different device or partition in order to expand the overall capacity of the Siebel File System. (You must consider the future growth of the volume of data when you plan how to organize the file system directories.)

You can also use sfspartition to remove one or more existing directories from service for the Siebel File System, provided that the overall file system capacity remains sufficient. To partition your Siebel File System, you first update the value of the enterprise parameter Siebel File System (alias FileSystem) so it specifies all of the network directories that you want to use for the Siebel File System, delimited by commas. You then run the partitioning utility and specify both the original directories containing the existing files and the updated target directories, corresponding to the updated FileSystem parameter value. The sfspartition utility distributes the files in the Siebel File System evenly across the target directories. The utility logs information into a file named sfspartition.log, which is located in the log directory within the Siebel Server root directory.

When the File System Manager component (alias FSMSrvr) starts, it verifies the existence of all of the file system directories specified using the FileSystem parameter. When new file attachments are inserted, FSMSrvr distributes them across these directories. If a file system directory is unavailable, then FSMSrvr logs an error message in the FSMSrvr log file and tries to write the file attachment to the next available directory. If no file system directory is available, then FSMSrvr terminates and writes an error message to the FSMSrvr log file.

In order to maintain the even distribution of files across file system directories, you must run the partitioning utility every time that you update the value of the FileSystem parameter, for example, if you add or remove a file system directory. The procedure in this topic describes in detail how to perform this task.

Before you partition your Siebel File System, note the following additional deployment options:

Mobile Web Client. A Mobile Web Client's configuration file must refer to a single directory location, unless you configure it to use the server-based data source.
Replication Manager requirements. Partitioning is supported on replicated nodes. For more information about replication, see Siebel Remote and Replication Manager Administration Guide.

Partitioning the File System Directories Using the sfspartition Utility

To partition the Siebel File System directories, run the partitioning utility sfspartition, as described in the following instructions.

This topic is part of Partitioning the Siebel File System.

The partitioning utility is named sfspartition.exe on Microsoft Windows or sfspartition on UNIX operating systems. This utility is located in the bin directory within the Siebel Server root directory.

Note: Where necessary, before you run the sfspartition utility, you must manually create any file system directories (such as the examples siebelFS1, siebelFS2, and siebelFS3, and so on) and subdirectories (such as att, attmp, and so on) on each target file system and grant the appropriate permissions to all of these directories. All of the file system directories must be accessible to all of the applicable Siebel Servers using the notation by which they are represented in the value of the FileSystem parameter.

For more information about creating the Siebel File System, see the Siebel Installation Guide for the operating system you are using.

For more information about the parameters you can use for the sfspartition utility, see Parameters for the sfspartition Utility.

To partition the file system directories using sfspartition

Where necessary, create any new directories or partitions that you will use with the Siebel File System, create required subdirectories, and grant the appropriate permissions to all of the directories. Then verify access to these directories.
Note the current value of the enterprise parameter Siebel File System (alias FileSystem) for later reference.

Note: You must note this value because you will use this information later when you specify the source directories by using the /O parameter when you run the sfspartition utility.
Set the value of the FileSystem parameter to include all of the directories that you want to use for the Siebel File System. Separate each directory with a comma (with no spaces), as in the examples that follow. Note the updated parameter value for later reference.

Note: You must note this value because you will use this information later when you specify the target directories by using the /F parameter when you run the sfspartition utility.
For example, on UNIX, you might specify this value:
```
/export/home/siebelFS1,/export/home/siebelFS2,/export/home/siebelFS3
```
For example, on Microsoft Windows, you might specify this value:
```
\\\\server1\\siebelFS1,\\\\server1\\siebelFS2,\\\\server2\\siebelFS3
```
Note: In this example, note that each backslash is doubled compared to the usual notation for such shared directories. For example, \\server1\siebelFS1 must be represented as \\\\server1\\siebelFS1. Alternatively, each file system directory can be represented using a mapped drive letter by which the directory can be accessed from each Siebel Server, such as D:\\siebelFS1 (note that each backslash must be doubled in this scenario also).

Caution: You must specify the file system directories in the same order for the FileSystem parameter and for the sfspartition utility. If you specify the directories using a different order, then the Siebel File System files might not be accessible after you use sfspartition.
If it is not already set, then set the ServerDataSrc named subsystem parameter DSFileSystem to *FSM*.
For information about configuring named subsystem parameters, see Configuring Siebel Enterprise Server Named Subsystem Parameters.
Restart the Siebel Server after updating the FileSystem and DSFileSystem parameter values.
Open a command prompt and change the directory to the bin subdirectory within the Siebel Server root directory.
Run sfspartition using parameters listed in this topic, as in the examples that follow.
The following example for UNIX distributes the files from one file system directory into three directories, corresponding to the updated value of the FileSystem parameter from topicPartitioning the Siebel File System:
```
sfspartition /O /export/home/siebelFS /F /export/home/siebelFS1,/export/home/siebelFS2,/export/home/siebelFS3 /H Y
```
Tip: Depending on how these network directories were created or mounted for use in UNIX environments, they might be on the same server or on different servers.
The following example for Microsoft Windows distributes the files from one file system directory on server1 into three directories on server1 and server2, corresponding to the updated value of the FileSystem parameter from the first step in this procedurePartitioning the Siebel File System:
```
sfspartition /O \\server1\siebelFS /F \\server1\siebelFS1,\\server1\siebelFS2,\\server2\siebelFS3 /H Y
```

Parameters for the sfspartition Utility

The following table describes the parameters for the sfspartition utility.

This topic is part of Partitioning the Siebel File System.

Table Parameters for sfspartition Utility

Parameter	Value	Description	Required?
/O	Paths for existing source directories	Set this value to the paths of the existing source directories for the file system. Separate multiple directories using commas, with no spaces. If any of the paths themselves contain a space, then enclose the parameter value in double quotes. Whether you must append `att` to each source directory depends on how you use the /H parameter: If you use /H Y, then do not append `att` to each source directory that you specify by using /O. (The purpose of /H Y is to eliminate the need to append `att`.) If you use /H N (or omit /H), then you must append `att` to each source directory that you specify by using /O. The utility looks both in the specified source directories and in the `att` subdirectories to find the files to be distributed to the specified target directories.	Yes
/F	Paths for target directories	Set this value to the paths of the target directories for the file system. Separate multiple directories using commas, with no spaces. If any of the paths themselves contain a space, then enclose the parameter value in double quotes. (Use the same value as the value of the FileSystem parameter.) Whether you must append `att` to each target directory depends on how you use the /H parameter: If you use /H Y, then do not append `att` to each target directory that you specify by using /F. (The purpose of /H Y is to eliminate the need to append `att`.) If you use /H N (or omit /H), then you must append `att` to each target directory that you specify by using /F. Otherwise, files will not be distributed to the `att` subdirectories of the target directories and will be inaccessible to clients.	Yes
/H	Y or N	Set /H Y if you want the utility to automatically append `att` to each source and target directory that you specify by using /O and /F. (Do not append `att` when you specify these directories.) Set /H N (or omit /H) if you do not want the utility to automatically append `att` to each source and target directory that you specify by using /O and /F. (Append `att` when you specify these directories.)	No

Parameter

Value

Description

Required?

/O

Paths for existing source directories

Set this value to the paths of the existing source directories for the file system. Separate multiple directories using commas, with no spaces. If any of the paths themselves contain a space, then enclose the parameter value in double quotes.

Whether you must append att to each source directory depends on how you use the /H parameter:

If you use /H Y, then do not append att to each source directory that you specify by using /O. (The purpose of /H Y is to eliminate the need to append att.)
If you use /H N (or omit /H), then you must append att to each source directory that you specify by using /O. The utility looks both in the specified source directories and in the att subdirectories to find the files to be distributed to the specified target directories.

Yes

/F

Paths for target directories

Set this value to the paths of the target directories for the file system. Separate multiple directories using commas, with no spaces. If any of the paths themselves contain a space, then enclose the parameter value in double quotes. (Use the same value as the value of the FileSystem parameter.)

Whether you must append att to each target directory depends on how you use the /H parameter:

If you use /H Y, then do not append att to each target directory that you specify by using /F. (The purpose of /H Y is to eliminate the need to append att.)
If you use /H N (or omit /H), then you must append att to each target directory that you specify by using /F. Otherwise, files will not be distributed to the att subdirectories of the target directories and will be inaccessible to clients.

Yes

/H

Y or N

Set /H Y if you want the utility to automatically append att to each source and target directory that you specify by using /O and /F. (Do not append att when you specify these directories.)

Set /H N (or omit /H) if you do not want the utility to automatically append att to each source and target directory that you specify by using /O and /F. (Append att when you specify these directories.)

No

Cleaning Up the Siebel File System

This topic describes how to clean up the Siebel File System by removing orphan records using the Siebel File System cleanup utility, sfscleanup. Orphan records are those that remain if a user deletes a parent record in the application that has associated child records. The child records are not deleted from the Siebel File System with the parent record and so you must remove them by using file system cleanup utility.

The Siebel File System cleanup utility is named sfscleanup.exe on Microsoft Windows or sfscleanup on UNIX operating systems. This utility is located in the bin directory within the Siebel Server root directory.

The sfscleanup utility processes records for every file in the file attachment directories (the att subdirectories) of the specified Siebel File System directories and performs one of several operations to each record and file, depending on the file type and on the parameters that you set. Optionally, you can run sfscleanup for a limited period of time and resume the operation again later.

For information about running sfscleanup, including descriptions of the file types and the associated operations performed by sfscleanup during processing, and the run-time parameters you can use for the utility, see the information that follows.

This topic is part of Administering the Siebel File System.

This topic contains the following information:

Cleaning Up the File System Directories Using the sfscleanup Utility

This topic describes how to clean up the Siebel File System directories by removing orphan records using the Siebel File System cleanup utility, sfscleanup.

This topic is part of Cleaning Up the Siebel File System.

To clean up the file attachment directory using sfscleanup

At the command prompt, change directory to the bin subdirectory within the Siebel Server root directory.
Run sfscleanup using parameters listed here, in a command like the following example:
```
sfscleanup /U sadmin /P pwd /F \\server1\files /X \\server1\logs\sfscleanup.log
```
For more information about the parameters you can use for the sfscleanup utility, see Parameters for the sfscleanup Utility.

About the sfscleanup Log File

If you specified an output file using the /X parameter, then sfscleanup generates a log file listing the operations that were performed. The output file is a tab-delimited text file that contains the following columns:

File Name. This column lists the name of each file that was processed.
File Type. This column lists the type of each file that was processed.
Operation. This column lists the type of operation that was performed during processing. A table later in this topic provides descriptions of each operation.

File Types and Operations for the sfscleanup Utility

The following table lists the possible file types and the associated operation performed by sfscleanup during processing. The Operation column lists the type of operation that was performed during processing.

Table File Types and Associated Operations for sfscleanup Utility

File Type	Description	Operation
CURRENT	The file has a corresponding record in the file attachment database table.	KEPT
NEW	The file is less than one hour old. The `sfscleanup` utility does not check for the file in the file attachment database table.	KEPT
ORPHAN	The file does not have a corresponding record in the file attachment database table. If you used the /M parameter to set a move directory, then the operation performed is MOVED, not DELETED.	DELETED
INVALID	The file (or directory) is not a file attachment. If `sfscleanup` tries to delete a subdirectory that is not empty, then the operation errors out. Review the files contained within the directory before deleting them. If you set the /G parameter to Y, then the operation performed is DELETED, not KEPT.	KEPT
ANCIENT	The file has an associated record in the database with a different revision number. If you set the /N parameter to Y, then the operation performed is either MOVED (if you used the /M parameter to set a move directory) or DELETED, not KEPT.	KEPT

The operations mentioned in the previous table are described in the following table.

Table Operations for sfscleanup Utility

Operation	Description
KEPT	The file was kept.
DELETED	The file was deleted.
MOVED	The file was moved to the directory specified by the /M parameter. Files are moved if you used the /M parameter.
KEPT_DIR	The item was kept because it was a directory and requires manual processing.
KEPT_ERROR	The file was kept because an error occurred while trying to move or delete the file.

Parameters for the sfscleanup Utility

The following table describes the parameters for the sfscleanup utility. More information about some of the parameters is provided after the table.

This topic is part of Cleaning Up the Siebel File System.

Table Parameters for sfscleanup Utility

Parameter	Value	Description	Required?
/U	Username	User name ID.	Yes
/P	Password	User name password.	Yes
/C	ODBC_data_source	Set this value to the ODBC data source. The default value is the setting of the environment variable SIEBEL_DATA_SOURCE.	No
/D	Siebel_table_owner	Set this value to the Siebel table owner. The default value is the setting of the environment variable SIEBEL_TABLE_OWNER.	No
/F	Paths for file system directories	Set this value to the paths for the file system directories. Separate multiple directories using commas, with no spaces. If any of the paths themselves contain a space, then enclose the parameter value in double quotes. (Use the same value as the value of the FileSystem parameter.) Whether you must append `att` to each directory depends on how you use the /H parameter: If you use /H Y, then do not append `att` to each directory that you specify using /F. (The purpose of /H Y is to eliminate the need to append `att`.) If you use /H N (or omit /H), then you must append `att` to each directory that you specify using /F. The utility looks both in the specified directories and in the `att` subdirectories to find the files to be cleaned up.	Yes
/X	Path for output file	Set this value to the path for the output file.	No
/M	Path for move directory	Set this value to the path for the directory where discarded files are to be moved.	No
/N	Y or N	Determines whether old versions of file attachments are to be removed. To remove old versions, set this value to Y. The default value is N.	No
/R	Y or N	Set this value to Y to generate only a report file. If it is set to Y, then the report file contains only the columns File Name and File Type. The default value is N.	No
/H	Y or N	Set /H Y if you want the utility to automatically append `att` to each directory that you specify using /F. (Do not append `att` when you specify these directories.) Set /H N (or omit /H) if you do not want the utility to automatically append `att` to each directory that you specify using /F. (Append `att` when you specify these directories.)	N
/G	Y or N	Set this value to remove garbage files or non-Siebel files. The default value is N.	No
/Q	Y or N	Set /Q Y if you want the utility to perform a query by file attachment records. This parameter allows you to run the utility for a limited period of time and provides other ways to manage how the utility runs. The default value is N. Note: When you are using /Q Y, you can also use the parameters /I, /O, /S, and /T. Otherwise, these parameters have no effect.	N
/I	Number of file IDs	Set /I to the number of file attachment records to query. The default value is 300. The utility processes records and files in batches based on the specified number of records. After those files have been processed, the utility processes more records and files in another batch. Note: This parameter has an effect only if you are using /Q Y.	N
/O	Y or N	Use /O Y when you want the utility to use an OR clause to constrain the query row IDs, like this: `(ROW_ID = 'Id1' OR ROW_ID = 'Id2' OR ...)`. The default value is Y. Use /O N to instead use a clause like this: `ROW_ID IN ('Id1','Id2',...)`. Note: This parameter has an effect only if you are using /Q Y. It determines the internal query executed in Siebel database. No row ID is entered as an argument.	N
/S	Y or N	Use /S Y to resume the previous run, from the next unprocessed record, where information about the last processed record is available in a temporary directory. The default value is N. Use /S N (or omit /S) to instead start a new run. Note: This parameter has an effect only if you are using /Q Y.	N
/T	Number of minutes	Set /T to the number of minutes to run the query. When the utility reaches that time, the last processed file attachment record is noted in a temporary directory and the utility exits. Later, you can resume the previous run by using /S Y. By default, the utility runs to completion, until all of the records and files are processed. Note: This parameter has an effect only if you are using /Q Y.	N

More Information About Some sfscleanup Parameters

The following provides more information about some of the parameters for the sfscleanup utility.

/N. By default, old file revisions are kept. Such files are marked ANCIENT in the log, and represent old revisions of an existing attachment record. That is, their row ID matches with the database record but not the file revision number. To delete such files, set the /N parameter to Y.
/G. If the file system contains files that were not created by the File System Manager component (alias FSMSrvr), then their deletion or move is controlled by the /G parameter. This parameter includes non-Siebel files or directories. By default these files are not deleted. The directories are not affected or moved by sfscleanup.
/Q. By default, the sfscleanup utility processes all of the files in the file attachment directories in a single long-running operation. For a Siebel File System that includes a very large number of files, such an operation might affect performance or inconvenience production users.
Alternatively, the /Q parameter allows you, for example, to process files in batches based on a query of a given number or file attachment records (by using /I), to modify how the utility queries these records (by using /O), to run the utility for a specific period of time (by using /T), and to resume a run later where it left off (by using /S). For example, you might decide to run the sfscleanup utility only when most of your users are not logged in.

8Siebel Server Infrastructure Administration