When you first install NetWare 6.5, the operating system is tuned by default so that its performance is optimized for most systems. Over time, the server optimizes itself according to the network usage it encounters. Occasionally, however, you might find that you want to make manual changes to your system in some aspect of the server's operation. This is particularly true if you are reconfiguring a server specifically to run a different set of service than what is currently configured. For this reason, there are numerous server parameters you can set to change the way the server handles things such as memory or file locks. These server parameters, also called SET parameters, can be set in several ways: -
NoRM: Open the SET parameters page to access all NetWare 6.5 SET parameters, by category. -
MONITOR.NLM: Provides menu-driven access to SET parameters from the server console. Select Server Parameters from the main menu. -
SET console command: You can change any SET parameter individually by using the following general syntax: SET < parameter name > = < value > In most cases, you might not need to change any server parameters. However, if you need to change them, each server parameter is explained in this appendix. There are 14 SET parameter categories in NetWare 6.5. Within each category there are hidden and unhidden parameters. Hidden parameters are viewable through NoRM by selecting the View Hidden SET Parameters setting on the Set Parameters page. However, information is limited for these parameters and they should be adjusted only if you are absolutely sure of the consequences. The following sections list some, but not all, of the hidden parameters. Use NoRM for a complete listing of potential settings. Common File System This category sets basic file system parameters common to both NSS and traditional volumes , including compression characteristics. Table C.10 describes the Common File System parameters. Table C.10. Common File System SET Parameters PARAMETER | VALUES | DESCRIPTION | Maximum transactions | 100 to 10000 (Default: 10000) | Specifies how many transactions can occur simultaneously across all connections. | Maximum concurrent directory cache writes | 5 to 500 requests (Default: 75) | Sets the maximum number of write requests that can be stored before the disk head begins a sweep across the disk. | Minimum file delete wait time | 0 sec7 days (Default: 1 min. 6 sec) | Specifies how long a deleted file must be stored before it can be purged. | Immediate purge of deleted files | ONOFF (Default: OFF) | Specifies whether files are purged immediately when they are deleted or stored in a salvageable state. If turned ON, this parameter purges deleted files immediately, and they cannot be salvaged. | Compression daily check stop hour | 023 (Default: 6) | Specifies the hour, using a 24-hour clock, when the file compressor stops searching volumes for files that need to be compressed. If this value is the same as the Compression daily check starting hour value, the search starts at the specified starting hour and goes until all compressible files have been found. | Compression daily check starting hour | 023 (Default: 0) | Specifies the hour, using a 24-hour clock, when the file compressor begins searching volumes for files that need to be compressed. | Minimum compression percentage gain | 050 (Minimum 20) | Specifies the minimum percentage that a file must be able to be compressed in order to remain compressed. | Enable file compression | ONOFF (Default: ON) | When set to ON, file compression is allowed to occur on volumes that were previously enabled for compression (during installation). Just because a volume is enabled for compression doesn't mean compression will actually occur. This parameter must be turned ON for compression to occur. For more information about file compression, see Chapter 8. | Maximum concurrent compressions | 18 (Default: 2) | Specifies how many volumes can compress files at the same time. Increasing this value might slow down server performance during compression times. | Convert compressed to uncompressed option | 02 (Default: 1) | Specifies how a compressed file is stored after it has been accessed. 0 = always leave the file compressed; 1 = leave the file compressed after the first access within the time frame defined by the Days untouched before compression parameter and then leave the file uncompressed after the second access; 2 = change the file to uncompressed after the first access. | Decompress percent disk space free to allow commit | 075 (Default: 10) | Specifies the percentage of free disk space that is required on a volume before committing an uncompressed file to disk. This helps you avoid running out of disk space by uncompressing files. | Decompress free space warning interval | 0 sec29 days 15 hours, 50 min 3.8 sec. (Default: 31 min 18.5 sec) | Specifies the interval between warnings when the volume doesn't have enough disk space for uncompressed files. | Deleted files compression option | 02 (Default: 1) | Specifies how the server handles deleted files. 0 = don't compress deleted files; 1 = compress deleted files during the next day's search; 2 = compress deleted files immediately. | Days untouched before compression | -0100000 (Default: 14) | Specifies how many days a file or directory must remain untouched before being compressed. | Communications Communication parameters configure the way the operating system handles communication buffers. Table C.11 describes the communication parameters that are available. Table C.11. Communication SET Parameters PARAMETER | VALUES | DESCRIPTION | IPX CMD Mode Routing | ONOFF (Default: OFF) | When set to ON, turns on IPX Compatibility Mode routing. | SPX Maximum Window Size | 016 (Default: 0) | Sets the maximum SPXS window size. | Load Balance Local LAN | ONOFF (Default: OFF) | Turns load balancing on or off. | Maximum Pending TCP Connection Requests | ONOFF (Default: OFF) | Determines whether IPX Compatibility Mode routing is enabled. | BSD Socket default Buffer Size in Bytes | 4096 to 1073741824 bytes (Default: 32768) | This parameter is used to set the BSD socket default send and receive buffer size. The constraint is given in bytes. | Discard Oversize PING Packets | ONOFF (Default: ON) | Determines whether to discard ping packets larger than Largest Ping Packet Size . | Largest Ping Packet Size | The range is 0 to 65535 | This option sets the largest ping packet size that can be received by the server. | TCP Defend Land Attacks | ONOFF (Default: ON) | Sets whether TCP defends against land attacks. | IP Wan Client Validation | ONOFF (Default: OFF) | Sets whether IP WAN clients are validated when they dial in to the network remotely via NetWare Connect. | TOS for IP Packets | 0 to 127 (Default: 0) | This SET option is used to set the Type of Service (TOS) in the IP header for all outgoing packets for all available interfaces. | Arp Entry Update Time | 240 to 14400 (Default: 300) | This parameter is used to change the update time of Address Resolution Protocol (Arp) entries in seconds. It is recommended that the Arp entry update time be greater than or equal to the Arp entry expiry time . | Arp entry expiry time | 240 to 14400 (Default: 300) | This option is used to change the expiry time of Arp entries in seconds. It is recommended that the Apr entry expiry time be less than or equal to the Arp entry update time. | ICMP Redirect Timeout | 0525600 (Default: 3) | Allows ICMP learned routes to expire. 146 = 1 sec. | Largest UDP Packet Size | 036992 (Default: 16384) | Set maximum allowable UDP packet for this server. | Discard Oversized UDP Packets | ONOFF (Default: ON) | Discards UDP packets larger than Largest UDP Packet Size . | TCP UDP Diagnostic Services | ONOFF (Default: OFF) | Allows you to start or stop diagnostic servers over TCP and UDP. | TCP Path MTU Black Hole Detection and Recovery | ONOFF (Default OFF) | This option is used to enable or disable Path Maximum Transmission Unit (MTU) Black Hole Detection and Recovery feature. This feature aids in resolving the problems of datagrams being sent out, but no replies come back. | TCP Max Port Limit | 499954999 (Default: 54999) | Defines the upper limit for ephemeral, or user -configurable, TCP ports. | TCP Sack Option | ONOFF (Default: ON) | This SET option is used to enable or disable Selective Acknowledgment (Sack) support. This set option will not change the behavior of existing connections. | TCP Large Window Option | ONOFF (Default: ON) | Use this parameter to enable or disable Large Window support. This set option will not change the behavior of existing connections. | Maximum RIP SAP Events | 1 to 100000 (Default: 100) | Limits the SAP and RIP traffic on network. | IPX Router Broadcast Delay | 0, 1, 2 (Default: 0; Adjust based on size of RIP/SAP tables) | Sets how long the IPX router waits between SAP/RIP broadcast packets. | IPX NETBios replication option | 03 (Default: 2) 0 = no replication of broadcasts; 1 = replicate broadcasts; 2 = replicate broadcasts, but suppress duplicate broadcasts; 3 = same as 2, but do not replicate to WAN links | Sets how the IPX router handles replicated NetBIOS broadcasts. | Use old watchdog packet type | ONOFF (Default: OFF) | Sets server to use type 0 instead of type 4 watchdog packets. Use this option if you use older router hardware that filters out type 4 IPX packets. | Reply to get nearest server | ONOFF (Default: ON) | When set to ON , this server will respond to workstations that request a connection to their nearest server. | Number of watchdog packets | 5100 (Default: 10) | Sets the number of watchdog packets the server sends to an unresponsive workstation before clearing the workstation's connection. | Delay between watchdog packets | 9.9 sec to 10 min 26.2 sec (Default: 59.3) | Sets the time the server waits before sending each watchdog packet. | Delay before first watchdog packet | 15.7 sec to 14 days (Default: 4 min 56.6 sec) | Sets the time the server waits before sending the first watchdog packet to an unresponsive workstation. | Console display watchdog logouts | ONOFF (Default: OFF) | When set to ON , a console message is displayed when the watchdog clears a workstation's connection. | Maximum packet receive buffers | 50 to 4294967295 (Default: 500) | Sets the maximum number of packet receive buffers the server can allocate. | Minimum packet receive buffers | 10 to 4294967295 (Default: 128) | Sets the minimum number of packet receive buffers that the server can allocate. This number is allocated automatically when the server is booted . | Maximum physical receive packet size | 618 to 24682 bytes (Default: 4224) | Sets the largest size of packets that can be received by an MLIB. Default size is acceptable for Ethernet and token ring boards. If some boards on the network can transmit more than 512 bytes of data per packet, use the largest packet size. | New packet receive buffer wait time | 0.1 to 20 sec (Default: 0.1) | Sets the time the operating system waits after allocating the minimum number of buffers before granting the next packet receive buffer. | Maximum interrupt events | 1 to 1000000 (Default: 10) | Sets the maximum number of interrupt time events (such as IPX routing) that occurs before a thread switch is guaranteed to have occurred. | Directory Services Directory Caching parameters enable you to configure how directory cache buffers are used to optimize access to frequently used directories. A directory cache buffer is a portion of server memory that holds a directory entry that is accessed frequently. A directory entry held in memory is accessed faster than a directory entry stored on the hard disk. Table C.12 describes the Directory Caching parameters that are available. Table C.12. Directory Services SET Parameters PARAMETER | VALUES | DESCRIPTION | NDS trace to screen | ONOFF (Default: OFF) | When set to ON , the NDS trace screen, which displays information about NDS events, is turned on. | NDS trace to file | ONOFF (Default: OFF) | When set to ON , the NDS trace information is sent to a file in the SYS: SYSTEM directory, named DSTRACE.DBG by default. | NDS trace filename | Path and filename (Default: SYS: SYSTEM\DSTRACE.DBG ) | Specifies a different pathname or filename for the NDS trace file. | Bindery context | Name of Bindery Context(s) (Default: None) | Specifies which containers and their objects will be used as the server's "bindery" when the server provides bindery services. You can include up to 16 containers as part of this server's bindery context. Separate each context with a semicolon. | NDS external reference life span | 1384 hours (Default: 192) | Sets the number of hours that unused external references (local IDs assigned to users when they access other servers) can exist before they are removed. | NDS inactivity synchronization interval | 21440 minutes (Default: 60) | Sets how many minutes can elapse between exhaustive synchronization checks. Set high (up to 240 minutes) if replicas have to synchronize across WAN connections to reduce network traffic. | NDS synchronization restrictions | ONOFF (Default: OFF) | When set to OFF , the server synchronizes with all versions of NDS that are available on the network. When this parameter is ON , the server synchronizes only with the versions of NDS specified (such as "On, 489, 492, 599"). | NDS servers status | UPDOWN (Default: None) | Sets the status of all Server objects in the local NDS database as either up or down, so you can force the network to recognize that a particular server is up when the network thinks it is down. | NDS janitor interval | 110080 minutes (Default: 60) | Specifies how often, in minutes, the janitor process runs. The janitor process cleans up unused records, reclaims disk space, and purges deleted objects. | NDS Distributed reference link interval | 110080 minutes (Default: 780) | Specifies how often, in minutes, the NDS distributed reference link consistency check is performed. | NDS Bindery QoS Mask | 04294967295 (Default: 8) | Selects which eDirectory objects should be subject to the bindery quality of service delay. | NDS Bindery QoS Delay | 060000 msec (Default: 150) | Specifies the delay between successfully finding a requested bindery object and returning it to the requesting process. | NDS backlink interval | 110080 minutes (Default: 780) | Specifies how often, in minutes, backlink consistency is checked. Backlinks indicate that an object in a replica has an ID on a server where the replica does not exist. | NDS trace file length to zero | ONOFF (Default: OFF) | When set to ON , the server deletes the contents of the NDS trace file (but does not delete the trace file itself). To delete the file contents, also set the NDS Trace to File parameter to ON , so that the file will be open for the deletion process. | NDS bootstrap address | IP address (Default: None) | Specifies the address of a remote server with which this server can perform tree connectivity operations. Set only if this server does not hold a replica. | Disk Disk parameters allow you to control several aspects of your disk drive operation and configuration, including disk mirroring for fault tolerance. Table C.13 describes the SET parameters available in the Disk category. Table C.13. Disk SET Parameters PARAMETER | VALUES | DESCRIPTION | Auto Scan of CDM Modules | ONOFF (Default: ON) | When set to ON, the server automatically scans and loads defined generic Custom Device Modules (CDMs). If turned off, the desired CDMs must be loaded manually, or added to the STARTUP.NCF file. | Sequential elevator depth | 0294967295 (Default: 8) | Sets the number of sequential requests that the Media Manager will send to the same device. If another device in the mirror group is idle when the first device contains this number of requests, Media Manager will begin sending requests to the idle device. | Enable IO handicap attribute | ONOFF (Default: OFF) | When set to ON , it allows drivers and applications to set and use an attribute to inhibit (or handicap) read requests from one or more devices. When turned off, NetWare can treat the device like any other device. Do not turn this parameter on unless instructed to by the manufacturer. | Mirrored devices are out of sync message frequency | 59999 minutes (Default: 28) | Specifies how often devices are checked for out-of-sync status. | Remirror block size | 18 (Default: 1) | Specifies the remirror block size in multiples of 4K. 1 = 4K, 2 = 8K, 3 = 12K, and so on. | Ignore Partition Ownership | ONOFF (Default: OFF) | When set to ON , partitions owned by other servers in a cluster can be activated on this server. | Concurrent remirror requests | 232 (Default: 32) | Sets the number of simultaneous remirror requests that can occur per logical disk partition. | Ignore disk geometry | ONOFF (Default: OFF) | When set to ON , allows the creation of nonstandard and unsupported partitions on the server's hard disk. | Enable Hardware Write Back | ONOFF (Default: OFF) | Enables storage drivers to use the hardware write-back feature if they support it. This allows I/O write requests to be cached and processed before the data is actually written to the disk, which can increase performance. | Enable disk read after write verify | ONOFF (Default: OFF) | Specifies whether data written to disk is compared with the data in memory to verify its accuracy. If set to ON , this parameter tells the driver to perform the highest level of read-after-write verification that it can. If set to OFF , this parameter turns off any form of read-after-write verification that the driver performs . If the disk controller has a built-in function for read-after-write verification, leave this parameter OFF . | Error Handling Error-handling parameters let you manage log files and error recovery options in NetWare 6.5. Table C.14 describes the error-handling parameters available in NetWare 6.5. Table C.14. Error-Handling SET Parameters PARAMETERS | VALUES | DESCRIPTION | Server log file state | 02 (Default: 1) | Specifies what action to take when the server log, SYS$LOG.ERR, reaches its maximum size. 0 = take no action; 1 = delete the log file; 2 = rename the log file. | Server log file overflow size | 655364294967295 bytes (Default: 4194304) | Specifies the maximum size of SYS$LOG.ERR. | Boot error log file state | 03 (Default: 3) | Specifies what action to take when the boot log, BOOT$LOG.ERR, reaches its maximum size. 0 = take no action; 1 = delete the log file; 2 = rename the log file; 3 = start a new log file each time the server reboots. | Boot error log file overflow size | 65536 to 4294967295 bytes (Default: 4194304) | Specifies the maximum size of BOOT$LOG.ERR. | Boot error log | ONOFF (Default: ON) | When turned on, specifies that all console error messages are saved in the BOOT$LOG.ERR file. When turned off, only error messages that occur during the boot procedure are saved in this file. | Hung unload wait delay | 0 sec1 min 58.3 sec (Default: 30 sec) | Specifies how long the system will wait for a hung NLM to unload before asking the user whether to shut down the NLM's address space. Used only for NLMs loaded in protected memory spaces. | Auto Restart Down Timeout | 0600 sec (Default: 180) | Specifies the time that a server will wait before automatically restarting after an ABEND . This prevents the server from hanging if a problem occurs while it is trying to go down. | Auto restart after ABEND delay time | 260 minutes (Default: 2) | Sets the time that the server will wait before automatically shutting down and restarting after an ABEND occurs. | Auto restart after ABEND | 03 (Default: 1) | Determines whether the server automatically attempts to recover if an ABEND occurs. 0 = the server does not restart itself; 1 = the server determines the ABEND cause, and then either keeps the computer running or shuts it down and restarts; 2 = the server attempts to recover, and then shuts down and restarts the server in the configured amount of time; 3 = the server immediately shuts down and restarts. | Licensing Services Licensing Services parameters provides a few Novell Licensing Services (NLS) configuration parameters. Table C.15 describes the licensing services parameters available in NetWare 6.5. Table C.15. Licensing Services SET Parameters PARAMETER | VALUES | DESCRIPTION | NLS search type | 01 (Default: 0) | Specifies where to stop the upward tree search for license objects. 0 = stop search at tree root, 1 = stop search at partition root. Setting this parameter to 1 will limit the scope of a license search to the current partition. | Store NetWare 5 Conn SCL MLA usage in NDS | ONOFF (Default: OFF) | Setting this to ON moves MLA-type server connection licenses (SCL) for NetWare 5 servers into eDirectory where they can be more easily managed. | NLSTRACE | 02 (Default: 0) | Specifies if/how NLS trace files should be managed. 0 = no trace, 1 = trace to screen, 2 = trace to screen and file ( SYS:SYSTEM\NLSTRACE.DBG ). | Memory Memory parameters enable you to configure how the server's memory is managed. There are also a bunch of hidden SET parameters in this category, which are viewable through NoRM but not Monitor. Table C.16 describes the memory parameters available in NetWare 6.5. Table C.16. Memory SET Parameters PARAMETER | VALUES | DESCRIPTION | VM garbage collector period | 10 sec23 h 59 m 29 s (Default: 5 min) | Specifies how often the Virtual Memory (VM) garbage collector will run. The VM garbage collector analyzes virtual memory, identifying memory contents that are no longer in use, cleaning them up. | VM garbage collector looks | 11048576 memory pages (Default: 10000) | Each time the VM garbage collector runs, this variable determines how many memory pages are examined. | Interactive screen timeout | 0 sec2730+ days (Default: 1 hour) | Specifies the amount of time that the current interactive screen will wait for some type of user input. | Average page in alert threshold | 1 to 4294967295 memory pages (Default: 2000) | Specifies the average number of incoming virtual memory pages that will trigger an alert to the system console. | Restart server on address space cleanup failure | ONOFF (Default: OFF) | When set to OFF , the server will isolate address spaces that cannot be cleaned up completely. When set to ON the server will react according to the Auto Restart settings in the Error, Handling SET Parameters. See the Error Handling section for more information. | Memory protection ABEND after restart count | ONOFF (Default: OFF) | If set to ON , the server will ABEND if an address space restarts more than the number of times specified in Memory protection restart count , within the time interval specified by Memory protection no restart interval . | Memory protection restart count | 11000 (Default: 1) | Specifies the number of restarts allowed during Memory Protection No Restart Interval. An address space will restart if a protection fault occurs. This parameter limits the number of the number of errors that can occur during the defined interval. | Memory protection no restart interval | 060 minutes (Default: 1) | Specifies the interval over which the number of address space restarts are tracked. If the number of restarts exceeds Memory protection restart count during the interval, the address space will not be restarted. Setting the parameter to turns the interval off, allowing the address space to restart regardless of faults. | Memory protection fault cleanup | ONOFF (Default: ON) | When set to ON , specifies that an address space is removed and its NLMs unloaded if an NLM in the space attempts to violate memory protection. When set to OFF , the address space isn't cleaned up, and the problem is left to be handled by the normal ABEND recovery process. | Garbage collection interval | 1 min1 hour (Default: 5 min) | Sets the maximum time between garbage collections. | Alloc memory check flag | ONOFF (Default: OFF) | When set to ON , the server is set to do corruption checking in the alloc memory nodes. | Reserved buffers below 16 meg | 82000 (Default: 300) | Reserves the specified number of cache buffers in lower memory for device drivers that do not support memory access above 16MB. | Miscellaneous The Miscellaneous category includes all those "other" parameters that don't fit in a specific category. Table C.17 describes the Miscellaneous SET parameters available in NetWare 6.5. Table C.17. Miscellaneous SET Parameters PARAMETER | VALUES | DESCRIPTION | Display incomplete IPX packet alerts | ONOFF (Default: ON) | When set to ON , alert messages are displayed when IPX receives incomplete packets. | Enable SECURE.NCF | ONOFF (Default: OFF) | When set to ON , causes the SECURE.NCF file to be executed when the server boots. This file is used to set configuration parameters necessary for a C2 security-compliant system. See the NetWare security documentation for more information. | Allow audit passwords | ONOFF (Default: OFF) | When set to ON , allows audit password requests to be used. | Command line prompt default choice | ONOFF (Default: ON) | Sets the default input for the conditional execution ( "?" ) console command. ON = Yes, OFF = No. | Command line prompt time out | 0 to 4294967295 sec (Default: 10) | Sets the number of seconds the conditional execution ( "?" ) will wait before using the default answer. | Sound bell for alerts | ONOFF (Default: ON) | When set to ON , a sound emits whenever an alert message appears on the server's console screen. | Replace console prompt with server name | ONOFF (Default: ON) | When set to ON , the server's name is displayed as the console prompt on the server command line. | Alert message nodes | 10256 (Default: 20) | Determines how many alert message nodes are pre-allocated. This sets aside memory to receive alert messages. | Classic work to do Pre Check Flag | ONOFF (Default: OFF) | If an event is scheduled using the classic NetWare WorkToDo method, setting this to ON causes the process scheduler to see if the event is already on the scheduler list. | Worker thread execute in a row count | 120 (Default: 10) | Determines how many times in a row the scheduler dispatches new work before allowing other threads to execute. | Halt system on invalid parameters | ONOFF (Default: OFF) | When set to ON , the system stops whenever invalid parameters or conditions are detected . When this parameter is set to OFF , the system displays an alert message but continues running if an invalid parameter is detected. | Display relinquish control alerts | ONOFF (Default: OFF) | When set to ON , messages are displayed when an NLM uses the server's processor for more than 0.4 seconds without giving up control to other processes. This command is useful if you are writing your own NLMs and want to see if your NLM is using the CPU correctly. | Display old API names | ONOFF (Default: OFF) | When set to ON , messages are displayed when API calls from older versions of NetWare are used by an NLM. If you receive these messages, you might want to contact the NLM manufacturer for an upgrade that uses current the current API set, which will provide better performance. | CPU hog timeout amount | 0 sec1 hour (Default: 1 min) | Sets the amount of time the system waits before terminating a process that has not relinquished control of the CPU. If set to , this option is disabled. | Developer option | ONOFF (Default: OFF) | When set to ON , options that are associated with a developer environment are enabled. | Display spurious interrupt alerts threshold | 11000000 per sec (Default: 200) | Specifies how many spurious interrupts per second must occur before a spurious interrupt alert is sent to the console. | Display lost interrupt alerts threshold | 11000000 per sec (Default: 10) | Specifies how many lost interrupts per second must occur before a lost interrupt alert is sent to the console. | Display spurious interrupt alerts | ONOFF (Default: OFF) | When set to ON , error messages are displayed when the server hardware creates an interrupt that has been reserved for another device. If you receive this error, remove all add-on boards and run SERVER.EXE (the NetWare operating system). If the message doesn't appear after you remove everything, add the boards one at a time until you locate the board that is generating the message. Then contact the board's vendor or manufacturer for assistance. | Display lost interrupt alerts | ONOFF (Default: OFF) | When set to ON , error messages are displayed when a driver or board makes an interrupt call but drops the request before it's filled. To identify the problem driver, unload all drivers, and then reload them one at a time. When you locate the driver that is generating the message, contact the driver's manufacturer. | Pseudo preemption count | 14294967295 (Default: 40) | Sets how many times threads can make file read or write system calls before they are forced to relinquish control. | Global pseudo preemption | ONOFF (Default: ON) | When set to ON , all threads are forced to use pseudo preemption. | Minimum service processes | 10500 (Default: 25) | Sets the minimum number of service processes the server can allocate without waiting for the amount of time specified in the New service process wait time parameter. | Maximum service processes | 51000 (Default: 500) | Sets the maximum number of service processes the server can create. Increase this parameter if the number of service processes (as shown in NoRM) is at the maximum | New service process wait time | .3 sec20 sec (Default: 2.2 sec) | Determines how long the server waits to allocate another service process after receiving an NCP request. If a service process is freed up during this time, a new one will not be allocated. | Allow unencrypted passwords | ONOFF (Default: OFF) | When set to ON , the server will accept unencrypted passwords. This is only required for ancient versions of NetWare, which you shouldn't be using anymore anyway. | Multiprocessor Multiprocessor parameters enable you to configure settings for servers with multiple processors. Table C.18 describes the Multiprocessor SET parameters in NetWare 6.5. Table C.18. Multiprocessor SET Parameters PARAMETER | VALUES | DESCRIPTION | System threshold | 0102400 (Default: 1536) | Specifies the threshold for calculating thread shedding for load balancing. Thread shedding is the process of transferring process threads to another processor when a processor starts to get overloaded. | Auto clear interrupt statistics | ONOFF (Default: ON) | When set to ON , the detailed statistics for an offline processor or removed interrupt handles are removed from memory. When set to OFF , the statistics are retained in memory. Must be set in STARTUP.NCF. Changes will take effect when server is rebooted. | Auto start processors | ONOFF (Default: ON) | When set to ON, secondary processors are automatically started when the PSM (Platform Support Module) is loaded. When set to OFF, you must use the console command START PROCESSORS to start secondary processors. | NCP NCP (NetWare Core Protocol) parameters enable you to configure NCP packets, control boundary checking, and change NCP Packet Signature security levels on the server. Table C.19 describes the NCP SET parameters in NetWare 6.5. Table C.19. NCP SET Parameters PARAMETER | VALUES | DESCRIPTION | NCP exclude IP addresses | max 191 char (Default: None) | Specifies IP addresses over which NCP is disabled. All other IP addresses on this server will accept NCP traffic. NONE indicates that no bound IP addresses on this server will disable NCP. ALL indicates that NCP is disabled on all bound IP addresses. Addresses listed here have priority over the SET NCP Include IP Addresses command. | NCP include IP addresses | max 191 char (Default: All) | Specifies IP addresses over which NCP is enabled. NONE indicates that NCP is disabled on all bound IP addresses. ALL indicates that NCP is enabled on all bound IP addresses. | NCP over UDP | ONOFF (Default: ON) | This parameter enables NCP packets over the UDP transport. | NCP TCP keep alive interval | 0 sec16 hours (Default: 9 min 53 sec) | Specifies how long before "TCP keep-alive" closes idle NCP connections. A value of indicates that idle connections are never closed. | Minimum NCP TCP receive window to advertise | 25616384 (Default: 4096) | Sets the minimum receive window to advertise on NCP connections. | NCP TCP receive window | 140065535 (Default: 23360) | Sets the advertised receive window on NCP connections. | Enable UDP checksums on NCP packets | 02 (Default: 1) | Specifies how checksumming of NCP UDP packets is handled. 0 = no checksums; 1 = checksums performed if enabled at the client; 2 = checksums required. | Enable IPX checksums | 02 (Default: 1) | Sets the IPX checksum level. 0 = no checksums; 1 = checksums performed if enabled on the client; 2 = checksums required. | Enable task zero checking | ONOFF (Default: OFF) | This is a read-only parameter and is provided for informational purposes only. | NCP packet signature option | 03 (Default: 1) | Sets the server's NCP packet signature security level. 0 = don't do packet signatures; 1 = do packet signatures only if the client requires them; 2 = do packet signatures if the client can, but don't require them if the client doesn't support them; 3 = require packet signatures. After startup, you can only increase the level. | Client file caching enabled | ONOFF (Default: ON) | When set to ON , allows clients to cache open files locally, thereby increasing file access performance. | NCP protocol preferences | IPX, TCP, UDP (Default: None) | Specifies the preferred protocol order of the loaded transports. List each transport separated by a space. | NCP file commit | ONOFF (Default: OFF) | When set to ON , an application is allowed to issue a File Commit NCP and flush the file immediately from cache to disk. | Display NCP bad component warnings | ONOFF (Default: OFF) | When set to ON , NCP bad component alert messages are displayed. | Reject NCP packets with bad components | ONOFF (Default: OFF) | When set to ON , the server rejects NCP packets that fail component checking. | Display NCP bad length warnings | ONOFF (Default: OFF) | When set to ON , NCP bad-length alert messages are displayed. | Reject NCP packets with bad lengths | ONOFF (Default: OFF) | When set to ON , the server rejects NCP packets that fail boundary checking. | Maximum outstanding NCP searches | 101000 (Default: 51) | Determines the maximum number of NCP directory searches that can be performed at the same time. | Allow change to client rights | ONOFF (Default: ON) | When set to ON , a job server is allowed to assume a client's rights for NCP Packet Signature. If you are concerned that a job or print server may forge packets, turn off this parameter. | Allow LIP | ONOFF (Default: ON) | When set to ON , support for Large Internet Packets is enabled. | Novell Storage Services A new category, added for NetWare 6.5, is Novell Storage Services (NSS). This category introduces 23 new parameters to allow customization of NSS. Table C.20 describes the NSS parameters available in NetWare 6.5. Table C.20. NSS SET Parameters PARAMETER | VALUES | DESCRIPTION | NSS auto locks CD-ROM disc in device | ONOFF (Default: OFF) | Disables the ability to eject a CD-ROM that is mounted as an NSS logical volume. | NSS auto update CD9660 volume objects to NDS | ONOFF (Default: OFF) | Automatically mounts a 9660 format CD-ROM when it is inserted into the server's CD-ROM reader. | Emulate classic NetWare directory quotas | ONOFF (Default: OFF | Setting this parameter ON instructs the server to emulate preNetWare 6.5 disk quotas. | NSS work to do count | 5100 entries (Default: 40) | This parameter determines the number of WorkToDo entries that can be concurrently executing. | NSS maximum cache balance buffers per session | 161048576 buffers (Default: 1024) | This parameter limits the number of cache buffers used during a cache balance operation. | NSS cache balance timer | 13600 sec (Default: 30) | This parameter sets the cache balance timer. | NSS cache balance percent | 199 percent (Default: 60) | This parameter sets what percentage of free memory NSS will use for its buffer cache. | NSS cache balance enable | ONOFF (Default: ON) | This parameter sets the dynamic balancing of free memory for the NSS buffer cache. | NSS buffer flush timer | 13600 sec (Default: 1) | Specifies the delay before flushing modified cache buffers to disk. | NSS minimum OS cache buffers | 2561048576 entries (Default: 256) | Specifies the minimum number of NetWare buffer cache entries. | NSS minimum cache buffers | 2561048576 (Default: 256) | Specifies the minimum number of NSS buffer cache entries. | NSS file flush timer | 13600 sec (Default: 10) | Specifies the delay before flushing modified open files to disk. | NSS closed file cache size | 01000000 (Default: 50000) | Specifies the number of closed files that can be cached in memory. | NSS open file hash shift | 825 (Default: 16) | Specifies the size of the Open File hash table in powers of 2. For example, a setting of 10 indicates 210, or 1024 bytes. | NSS auth cache size | 1650000 entries (Default: 1024) | Specifies the number of authorization cache entries. | NSS ASCII name cache enable | ONOFF (Default: ON) | This parameter enables or disables ASCII name caching. | NSS name cache enable | ONOFF (Default: ON) | This parameter enables or disables name caching. | NSS name cache size | 365521 entries (Default: 2111) | This parameter sets the number of name cache entries. | NSS low volume space alerts | ONOFF (Default: ON) | This parameter enables or disables sending messages warning of low storage space to all users. | NSS low volume space warning reset threshold | 01000000 MB (Default: 15) | This parameter sets the threshold, above which a low storage space warning is reset. | NSS low volume space warning threshold | 01000000 (Default: 10) | This parameter sets the threshold below which a low storage space warning will be issued. | Service Location Protocol Service Location Protocol (SLP) parameters enable you to configure SLP values. Table C.21 describes the SLP parameters in NetWare 6.5. Table C.21. Service Location Protocol SET Parameters PARAMETER | VALUES | DESCRIPTION | SLP exclude IP addresses | Max 191 char (Default: NONE) | Specifies IP addresses over which SLP is disabled. All other IP addresses on this server will accept SLP traffic. NONE indicates that SLP is enabled on all addresses. ALL indicates that SLP is disabled on all bound IP addresses. This setting overrides SLP include IP addresses . | SLP include IP addresses | Max 191 char (Default: ALL) | This parameter allows you to specify IP addresses over which SLP is enabled. All other IP addresses on this server will ignore SLP traffic. If you specify NONE , it indicates that SLP is disabled on all IP addresses. Specifying ALL enables SLP over all bound IP addresses on this server. | SLP agent IP address | Max 31 char (Default: FIRST) | This SET parameter specifies the IP address of the SLP Directory Agent (DA) on this server. You specify FIRST to use the first bound non-excluded IP address. FOLLOW may be specified to use the IP address of the received DA service request. | SLP API handicap | 1100000 (Default: 50) | The maximum number of UA requests before SLP starts slowing them down. | SLP nwserver NLM names | ONOFF (Default: OFF) | This parameter enables or disables the nwserver.novell service, which includes NLM names as an SLP attribute. | SLP maximum WTD | 164 (Default: 10) | Sets the maximum number of controlled WorkToDos supported on this server. | SLP reset | ONOFF (Default: OFF) | Forces the service agent to send new service registers. Forces the directory agent to send DA Advertise. After execution, flag is reset to OFF . | SLP scope list | Max 185 char (Default: NA) | Holds a comma-delimited list of Service Agent scopes. | SLP close idle TCP connections time | 086400 sec (Default: 300) | Specifies how many seconds the system must wait before closing idle TCP connections. | SLP DA heart beat time | 065535 sec (Default: 10800) | Specifies how many seconds the system must wait before sending the next DA heartbeat packet. | SLP DA event timeout | 0120 sec (Default: 15) | Specifies how many seconds the system must wait before timing out a DA packet request. | SLP event timeout | 0120 sec (Default: 3) | Specifies how many seconds the system must wait before timing out a multicast packet request. | SLP SA default lifetime | 065535 sec (Default: 3600) | Specifies the default lifetime for service registers. | SLP retry count | 0128 (Default: 3) | Specifies the maximum number of retry attempts. | SLP debug | 065535 (Default: 0) | When set to , turns off debug mode. When set to a number, turns on debug mode. Bit 0x01 = COMM; 0x02 = TRAN; 0x04 = API; 0x08 = ERR; 0x20 = SA. | SLP rediscover inactive directory agents | 086400 (Default: 60) | Specifies how many seconds the system must wait before issuing service requests to rediscover inactive DAs. | SLP multicast radius | 032 (Default: 32) | Specifies the multicast radius. | SLP DA discovery options | 08 (Default: 7) | Sets the DA discovery option bit flag. Bit 0x01 = use multicast DA advertisements; 0x02 = use DHCP discovery; 0x04 = use static file SYS:ETC\SLP.CFG ; 0x08 = scopes required. | SLP MTU size | 024682 bytes (Default: 1450) | Specifies the maximum transfer unit size. | SLP broadcast | ONOFF (Default: OFF) | When set to ON , tells SLP to use broadcast packets instead of multicast packets. When set to OFF, tells SLP to use multicast packets. | SLP TCP | ONOFF (Default: OFF) | When set to ON , tells SLP to use TCP packets instead of UDP packets when possible. When set to OFF, tells SLP to use UDP packets. | Time Initially, you set up time services on the server during installation. Time services are controlled by TIMESYNC.NLM, which is loaded automatically when the server is started up. To modify time synchronization after installation, use the Time SET parameters, which are described in Table C.22. All parameters that start with the word TIMESYNC are elements of the TIMESYNC.CFG file. NetWare 6.5 now updates the TIMESYNC.CFG file automatically when you make changes to these settings. You can use EDIT.NLM to modify the TIMESYNC.CFG file directly. Table C.22. Time SET Parameters PARAMETER | VALUES | DESCRIPTION | TIMESYNC configuration file | path\filename (Default: SYS: SYSTEM\TIMESYNC.CFG ) | Specifies the directory and filename of the time synchronization configuration file. | TIMESYNC configured sources | ONOFF (Default: OFF) | When set to ON , the server ignores SAP (Service Advertising Protocol) time sources and instead accepts time sources configured with the TIMESYNC Time Source parameter. When this parameter is turned OFF , it causes the server to listen to any advertising time source. | Start of daylight savings time | Max 79 char (First Sunday in April @0200) | Indicates the day that daylight saving time begins locally. To specify the beginning of daylight saving time so it recurs every year, enclose the date and time in parentheses, and use the following format: (April Sunday First 2:00:00 a.m.) . If you do not enclose the date in parentheses, the change will occur only in the current year. "April Sunday First" indicates that the change occurs on the first Sunday in April. | End of daylight savings time | Max 79 char (Default: Last Sunday in October @0200) | Indicates the day that daylight saving time ends locally. To specify the end of daylight saving time so it recurs every year, enclose the date and time in parentheses, and use the following format: (October Sunday Last 2:00:00 a.m.) . If you do not enclose the date in parentheses, the change will occur only in the current year. "October Sunday Last" indicates that the change occurs on the last Sunday in October. | Daylight savings time offset | +or- hour:minute:second (Default: +1:00:00) | Specifies the offset applied to time calculations when daylight saving time is in effect, causing UTC time to be recalculated from local time. | Daylight savings time status | ONOFF (Default: NA) | When set to ON , this parameter indicates that daylight saving time is currently in effect. Changing this parameter does not change the local time on the server. | New time with daylight savings time status | ONOFF (Default: ON) | When set to ON , the local time on the server is adjusted by adding or subtracting the time indicated in the Daylight saving time offset parameter. | TIMESYNC directory tree mode | ONOFF (Default: ON) | When set to ON , time synchronization ignores SAP (Service Advertising Protocol) packets that don't originate from within the server's Directory tree. When this parameter is set to OFF , the server accepts SAP packets from any time source on the network, regardless of the tree from which it originates. If SAP is turned on, this parameter should also be set to ON . | TIMESYNC polling count | 11000 packets (Default: 3) | Specifies the number of time packets to exchange while polling. Increasing this number might increase unnecessary traffic on the network. | TIMESYNC polling interval | 10 to 2678400 sec (Default: 600) | Determines the polling interval, in seconds. All servers in the tree must use the same polling interval. Maximum works out to 31 days. | TIMESYNC RESET | ONOFF (Default: OFF) | When set to ON , all servers are removed from the time source list, and time synchronization is reset. This parameter automatically resets itself to OFF after execution. | TIMESYNC restart flag | ONOFF (Default: OFF) | When set to ON , the TIMESYNC service automatically restarts. This parameter automatically resets itself to OFF after execution. | TIMESYNC service advertising | ONOFF (Default: ON) | When set to ON , SAP (Service Advertising Protocol) is turned on, meaning Single Reference, Reference, and Primary time sources advertise using SAP. Set this parameter OFF if you are using a configured list of time sources. | TIMESYNC synchronization radius | 0 to 2147483647 msec (Default: 2000) | Determines the maximum time that a server is allowed to drift from network time while still being synchronized. Do not set this parameter for under two seconds (2000 milliseconds ) unless you have an application that uses synchronized time stamps that will not tolerate a two-second deviation between time sources. | TIMESYNC time adjustment | +or- hour:minute:second [at month/day/year hour:minute:second] | Determines when a time adjustment will take place. This parameter does not apply to Secondary time servers. Use sparingly to correct network-wide time errors. Overuse can corrupt time synchronization. The default date and time is six polling intervals or one hour (whichever is longer) from the current time. | TIMESYNC time sources | Max 150 char (Default: ;) | Specifies a server, or servers, as accepted time sources. If used at the console, and no server name is entered, the parameter displays the list of configured servers. To specify multiple servers, separate each one by a semicolon (;). The list of servers must contain at least one semicolon, even if only one server is listed. A semicolon alone clears the list of servers. | TIMESYNC type | Single, Reference, Primary, Secondary | Specifies the type of time sync role this server will play in the network. Default is Single for the first server in the tree, and Secondary for all subsequent servers. See Chapter 6 for more information on time synchronization. | Time zone | Max 80 char (Default: NA) | Specifies the abbreviation for the server time zone, its offset from UTC (Universal Coordinated Time, which used to be called Greenwich Mean Time), and the daylight savings abbreviation, if applicable . For example: zone is MST7MDT for mountain standard time in the U.S.A. Mountain standard time is offset seven hours from UTC, and the abbreviation used when daylight saving time is in effect is MDT. | Default time server type | Max 50 char (Default: Secondary) | Specifies the type of time server for this server. | Traditional File System A new category with NetWare 6.5, Traditional File System, includes all kinds of parameters relating to traditional NetWare volumes. If you are familiar with SET parameters from NetWare past, this category combines parameters from several different areas. So, if you can't find it, it might very well be right here. Table C.23 describes the traditional file system parameters in NetWare 6.5. Table C.23. Traditional File System SET Parameters PARAMETER | VALUES | DESCRIPTION | Purge files on dismount | ONOFF (Default: OFF) | Automatically purges all deleted files on a volume when it is dismounted. | Volume log file state | 02 (Default: 1) | Specifies what action to take when VOL$LOG.ERR reaches its maximum size. 0 = take no action; 1 = delete the log file; 2 = rename the log file. | Volume TTS log file state | 02 (Default: 1) | Specifies what action to take when TTS$LOG.ERR reaches its maximum size. 0 = take no action; 1 = delete the log file; 2 = rename the log file. | Volume log file overflow size | 65536 to 4294967295 bytes (Default: 4194304) | Specifies the maximum size of VOL$LOG.ERR. | Volume TTS log file overflow size | 65536 to 4294967295 bytes (Default: 4194304) | Specifies the maximum size of TTS$LOG.ERR. | Auto TTS backout flag | ONOFF (Default: ON) | When set to ON , incomplete transactions can be backed out automatically when a downed server is rebooted. | TTS abort dump flag | ONOFF (Default: OFF) | When set to ON , the TTS$LOG.ERR file is created to record backout data in the event of a failure. | TTS unwritten cache wait time | 11 sec10 min 59 sec (Default: 59 min 19 sec) | Sets the time that a block of transactional data can be held in memory. | TTS backout file truncation wait time | 1 min 6 sec26 hours 21 min 51 sec (Default: 59 min 19 sec) | Sets the minimum amount of time that allocated blocks remain available for the TTS backout file. | Dirty directory cache delay time | 010 sec (Default: 0.5 sec) | Specifies how long a directory table write request is kept in memory before it is written to disk. | Directory cache allocation wait time | 0.5 sec2 min (Default: 2.2 sec) | Specifies how long the server waits after allocating one directory cache buffer before allocating another one. | Directory cache buffer nonreferenced delay | 1 sec1 hour (Default: 5.5 sec) | Sets how long a directory entry is held in cache before it is overwritten. | Directory cache allocation wait time | 0.52 min (Default: 2.2 sec) | Specifies how long the server waits after allocating one directory cache buffer before allocating another one. | Directory cache buffer nonreferenced delay | 1 sec1 hour (Default: 5.5 sec) | Sets how long a directory entry is held in cache before it is overwritten. | Maximum directory cache buffers | 20200000 (Default: 500) | Sets the maximum number of directory cache buffers that the server can allocate. Prevents the server from allocating so many directory cache buffers that other server processes run out. | Minimum directory cache buffers | 10100000 (Default: 150) | Sets the minimum number of directory cache buffers to be allocated by the server before the server uses the Directory cache allocation wait time to determine if an other directory cache buffer should be allocated. Allocating buffers too quickly will cause the server to eat up memory resources during peak loads. Waiting too long can cause a delay in file searches. This wait time creates a leveling factor between peak and low access times. | Maximum number of internal directory handles | 401000 (Default: 100) | Sets the maximum number of directory handles that are available to internal NLMs that use connection 0. A directory handle is allocated each time an NLM accesses a file or directory. Allocating directory handles decreases the time required to gain access rights. | Maximum number of directory handles | 201000 (Default: 20) | Sets the maximum number of directory handles that each connection can obtain. | Maximum record locks per connection | 10100000 (Default: 500) | Sets the number of record locks a workstation can use simultaneously. | Maximum file locks per connection | 101000 (Default: 250) | Sets the number of opened and locked files a workstation can use simultaneously. | Maximum record locks | 100400000 (Default: 20000) | Sets the number of record locks the server can support simultaneously. | Maximum file locks | 100100000 (Default: 10000) | Sets how many opened and locked files the server can support simultaneously. | Read ahead enabled | ONOFF (Default: ON) | When set to ON , background reads can be done during sequential file access so that blocks are placed into the cache before they are requested. | Read ahead LRU sitting time threshold | 0 sec1 hour (Default: 10 sec) | Sets the time the server will wait before doing a read ahead. (LRU means Least Recently Used .) | Minimum file cache buffers | 202000 (Default: 20) | Sets the minimum number of cache buffers that must be reserved for file caching. | Maximum concurrent disk cache writes | 104000 (Default: 750) | Sets the maximum number of write requests that can be stored before the disk head begins a sweep across the disk. | Dirty disk cache delay time | 0.1 sec10 sec (Default: 3.3 sec) | Sets how long the server will keep a write request in memory before writing it to the disk. | Minimum file cache report threshold | 02000 (Default: 20) | Sets how close to the minimum number of allowed buffers the system can drop before a warning message is sent. | Auto Mount mirrored volume containing inactive device | ONOFF (Default: OFF) | Automatically mounts a volume detected to have a mirrored partition with an inactive device associated with it. | Automatically repair bad volumes | ONOFF (Default: ON) | When set to ON , automatically runs VREPAIR.NLM when a volume fails to mount. | File delete wait time | 0 sec7 days (Default: 5 min 29.6 sec) | Sets the maximum amount of time a deleted file must be stored in a salvageable state. After this time has elapsed, the file can be purged if the space is needed. | Allow deletion of active directories | ONOFF (Default: ON) | When set to ON , a directory can be deleted even if a user has a drive mapped to it. | Maximum percent of volume space allowed for extended attributes | 550 percent (Default: 10) | Limits the percentage of disk space that can be used to store extended attributes. | Maximum extended attributes per file or path | 4512 attributes (Default: 16) | Specifies the maximum number of extended attributes that can be assigned to a file or a subdirectory (path) on any of the server's volumes. | Fast volume mounts | ONOFF (Default: ON) | When set to ON , allows the server to mount volumes more quickly by not checking certain less-important fields. This parameter should be on only if the volume was dismounted normally the last time. | Maximum percent of volume used by directory | 585 percent (Default: 13) | Limits the percentage of disk space that can be used as directory space. | Maximum subdirectory tree depth | 10100 subdirectories (Default: 25) | Sets the maximum level of subdirectories the server can support. | Volume low warn all users | ONOFF (Default: ON) | When set to ON , all users are notified when the free space on a volume reaches a minimum level. | Volume low warning reset threshold | 0100000 blocks (Default: 256) | Specifies the number of disk blocks above the Volume low warning threshold value that must be freed up to reset the low volume warning. This parameter controls how often you receive the low volume warning if your free space is fluctuating around the threshold. | Volume low warning threshold | 01000000 blocks (Default: 256) | Sets the minimum amount of free space (in blocks) that a volume can have before it issues a warning that it is low on space. | Turbo FAT re-use wait time | 0.3 sec1 hour 5 min 54.6 sec (Default: 5 min 29.6 sec) | Sets how long a turbo FAT (File Allocation Table) buffer stays in memory after an indexed file is closed. | Allow unowned files to be extended | ONOFF (Default: ON) | When set to ON, files can be changed even if their owner has been deleted. | |