MAI COMPANY CONFIDENTIAL FIELD INFORMATION BULLETIN SERVICE GROUP(S): MINI MFG: BASIC FOUR NACS TYPE(S): N039 MFG. MODEL(S): MAI Basic Four SPx Systems CATEGORY: HANDBOOKS DESC: Handbook *** TABLE OF CONTENTS AS OF 11/19/94 *** FIB 00001.Software Functions..........................................12/06/89 FIB 00002.Diagnostics Information.....................................09/15/89 FIB 00003.Lowest Acceptable Revision Levels (LARL)....................07/17/90 FIB 00004.Error Halt Codes............................................12/19/88 FIB 00005.Disk Drive Information......................................12/05/90 FIB 00006.CMB PCBA - 2000 System - Early Style........................09/21/94 FIB 00007.CMB PCBA - 2000 System - Late Style.........................06/05/89 FIB 00008.CMB PCBA - 2500, 3000 & 4000 Systems........................01/08/90 FIB 00009.Eight-way Controller PCBA - 2500,3000 & 4000 Systems........09/30/93 FIB 00010.Four-way Controller PCBA....................................04/25/89 FIB 00011.LAN Controller PCBA.........................................02/10/90 FIB 00012.Memory PCBA - 128K/256K - 2000 System.......................09/01/87 FIB 00013.Memory PCBAs - 1 MB & 4 MB - 2500,3000 & 4000 Systems.......06/26/89 FIB 00014.MTCS Controller PCBA........................................12/06/88 FIB 00015.MTS - Formatter Controller PCBA.............................09/01/87 FIB 00016.MTS - SCSI Bus Adapter PCBA.................................05/25/89 FIB 00017.WDC Adaptec and Adapter PCBA set............................09/01/87 FIB 00018.WD Controller PCBA - single board...........................04/24/89 FIB 00019.System Serial Number change information.....................12/06/89 FIB 00020.Expansion Interface A PCBA - 4000 System....................02/03/89 FIB 00021.Expansion Interface B PCBA - 4000 System....................03/10/88 FIB 00022.2500/3000/4000 Boot Error Messages..........................01/27/89 FIB 00023.Notes on margining the 3000/4000 power supply...............02/16/90 FIB 00024.Device to Major,Minor Number & Port Assignments.............01/17/92 FIB 00025.Hard Disk Bad Block Retirement..............................04/10/89 FIB 00026.Crash Recovery and Dump Analysis Procedures.................08/02/91 FIB 00027.Adding.Winchester disk drive(s) to the SPx system...........01/12/90 FIB 00028.Dual SCSI Disk Controller PCBA..............................03/16/90 FIB 00029.1/2" Tape Controller - Single Board.........................09/17/90 FIB 00030.Helpful Hints...............................................05/09/94 FIB 00031.SPx Data File Integrity Check & File Repair Procedure.......02/18/92 HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----Table Of Contents Pg01 of 01 FIB 00001 12/06/89 *** Software Functions *** Due to ongoing changes in BOSS/IX this section will not be maintained with exact, step-by-step O.S. installation instructions. The flow of activity is very similar on all releases and the procedure will generally suffice. If you encounter a particular problem during this procedure, refer to the Software Announcement document for the BOSS/IX release you are working with at the time. The Software Announcements can be found in the BOSS/IX section of the F.I.B.s SOFTWARE category. Whenever you must install the O.S. on an Spx System, be certain to install the diagnostics on the disk first. Since a diagnostics partition must be created on the disk, if you fail to do this prior to installation of the O.S. and the customer files, the diagnostics cannot be installed on the disk without backing up and re-installing the O.S and customer files. * INSTALLATION AND UPGRADE OF THE OPERATING SYSTEM, SOFTWARE AND DIAGNOSTICS. Diagnostic programs for your system may be contained on the system fixed disk. The system diagnostics are contained on a dedicated diagnostic partition. This diagnostic partition is inaccessible to user programs, and is actually beyond the reach of the user disk area. For this reason, the diagnostic programs cannot be accessed by the file system, yet they are maintained by a Diagnostic Executive (EXEC). If your system will not boot diagnostics from disk and the O.S. is already installed, the WD0 file system must first be backed up. Once backed up, a diagnostic partition can be reserved. Once the diagnostic partition is reserved, diagnostics can be installed from the bootable diagnostic tape. Afterwards the file system must be restored. Use the following procedure to install the diagnostics on your fixed disk. *** DISK DIAGNOSTIC INSTALLATION PROCEDURE *** If user data is already installed on the disk, you must do a complete file by file backup. If user data has not been installed, install the diagnostics first and then install the BOSS/IX. Mount the diagnostic tape. From the bootstrap menu enter 2 to alt load. For the boot device enter cs , for MCS or ts , for MTS and press CR. At the System File: prompt, press CR. When the EXEC is loaded, it will auto-size the system and display the program prompt: To begin the installation of diagnostics, you must first enable the service commands. Enter: service and press CR EXEC will then ask for the service password, enter: b4bus and press CR (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB001 Pg001 The prompt will again be displayed. The superblock on the disk must be updated to reserve a section of the disk for the diagnostics. Enter the superblock initialization command: sbinit and press CR EXEC will ask you if you want to destroy the existing disk partition information and the disk directory name. Any (y/n) responses are single character responses, and do not require a CR after entry. EXEC will prompt: Destroy: "volume ID" (y/n)? In place of "volume ID" will be the current volume ID contained in the superblock. y (do not press CR) EXEC will ask for the number of sectors to reserve on the disk for the diagnostics, enter: 1200 and press CR EXEC will prompt: Destroy: "partition name" (y/n)? The actual current partition name will be displayed in place of "partition name" (if a partition exists). Enter: y (do not press CR) EXEC will then ask: Disk name: Enter the disk name, up to 12 characters in length (usually WDx where x=driven number), and press CR. This name will be displayed every time you display the directory. EXEC will then display: Initializing: "disk name" Displayed in place of "disk name" will be the new name you just entered. The prompt will be displayed. To install the diagnostics on the fixed disk, enter: install and press CR EXEC will prompt: Destroy: "disk name" (y/n)? In place of "disk name" will be displayed the name of the disk entered during the "sbinit" command sequence. Enter: n (do not press CR) (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB001 Pg002 As the diagnostics are installed on the disk, EXEC will display the current program being copied from the tape, and the size in sectors. After the sector size display, the file is written to the disk and the directory updated. When all the files are installed, EXEC will again display the prompt: After the diagnostics are installed on the disk, you may want to list the diagnostics on the disk to verify proper installation, enter: ldisk and press CR A listing of the diagnostic files on disk will be displayed. You should test the bootstrap capability now, to see that you can boot diagnostics from the disk. Enter: shutdown and press CR. .This will get the '' prompt. Any file name listed by the 'ldisk' command can now be loaded by entry of the command: load diagname and press CR (in place of diagname enter the desired diagnostic) Once you are satisfied that everything is OK, shutdown the system. *** O.S. INSTALLATION INSTRUCTIONS *** Insert the O.S. base tape into the MCS drive or MTS drive. Do an alternate load. At the Boot Device: prompt, type cs for MCS, or ts for MTS, then press CR. Press CR again at System File: prompt. The system will display loading messages, then the Installation Menu. Enter: A and press CR Option A re-defines all the wd0 partitions and copies a new boot partition to disk. After the copy is complete, you are instructed to reload. After the boot partition is installed the configuration record must be installed. When the message 'OPTION>' is re-displayed mount the configuration record tape in the tape drive. Use option 'D' to install the configuration record. After the configuration record is installed enter: CTRL D and then shutdown and press CR After the configuration record, the EOS product must be installed. Reloading the system at this time automatically reloads and roots to the boot partition. Enter the correct date, the screen will then display the ADMIN> prompt. Enter: osinstall cs (for MCS) or osinstall ts (for MTS) and press CR This command sets up the rest of the partitions on the disk and installs the EOS product. When completed, you will be instructed to reload. Reloading of the system at this time automatically reloads and roots to the root partition. Either enter the current date or press CR if the date is correct. (continued on next page) At the ADMIN> prompt enter: HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB001 Pg003 install cs EUT EBS (for MCS) or install ts EUT EBS (for MTS) and press CR The BASIC and Utility files will be installed and the ADMIN> prompt will be displayed. The O.S. installation is now complete. Test the system for proper operation. In most cases, the etc/conf file (located in the boot partition) must be customized at this time. Then the system is ready for port configuration and down-loading of the user's files. These functions are the responsibility of the customer. *** MISCELLANEOUS SOFTWARE INFORMATION *** A copy of the System Errors List should be kept at each site, in the System Log book or other readily accessable location. To print out a copy of this list, at the ADMIN> prompt, enter: cd / and press CR pr /etc/comerr.list > /dev/lp and press CR To display the System Errors List enter: p /etc/comerr.list and press CR .............................................................................. In the case where it is desired to re-install the OS while leaving the user's files intact, do the following: 1. Perform an alternate load, Boot Device: wd0 System File: ,/etc/boot.conf 2. Mount the O.S. base tape and enter: osupdate /dev/cs (for MTC) or osupdate /dev/ts (for MTS) and press CR 3. Shutdown the system and perform a normal boot. 4. At the ADMIN> prompt, enter: . install cs EUT EBS (for MCS) or install ts EUT EBS (for MTS) and press CR ............................................................................ This proceedure will save the entire root partition to tape. 1. Perform an alternate load, Boot Device: wd0 System File: ,/etc/boot.conf 2. Label a scratch tape, enter: mcslabel set=01 id=root (for MCS) or mtslabel set=01 id=root (for MTS) and press CR (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB001 Pg004 3. To perform the backup, enter: mcssave -filesystem /dev/root name=backup 4. A compare of the files is recommended, after the backup, enter: mcscompare filesystem=/dev/root (for MCS) - or - mtscompare filesystem=/dev/root (for MTS) *** SYSTEM ERROR LOG INFORMATION *** To CREATE and INITIALIZE the ERROR LOG FILE: cd / and press CR /sys/errlog /etc/error.log -initial .............................................................................. To DISPLAY the ERROR LOG FILE to the TERMINAL: /sys/errlog /etc/error.log ............................................................................ To PRINT the ERROR LOG to the LP PRINTER: pr /sys/errlog /etc/error.log > dev/lp .............................................................................. To DISPLAY the last 8 ERROR LOG entrys from the memory tables: /sys/errlog .............................................................................. ORIGINATOR: H. Mitchell HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB001 Pg005 FIB 00002 09/15/89 *** Diagnostics Information *** o There is a very wide range of diagnostics available for these systems. o Each system is supplied with a DIAG tape. o Systems 3000 and 4000 also have fascilities to allow diagnostics to reside on the Winchester disk (wd0). o The Manual "BOSS/IX Diagnostics and Error Log Manual" is an excellent reference for all SPx diagnostics. If you service SPx systems and do not have a personal copy of this manual, M6204B, order one immediately! Be certain your copy has the B suffix. The B level manual was released to the field in November of 1988. o Because of the availability of the above mentioned manual and the large volume of data involved, this section of the handbook will not be maintained. o Using a macro to printout the errorlog to LP printer A macro can be made and saved on every 2000/3000 that will save time on getting a hard copy printout of the system errorlog. When at the ADMIN prompt type the following. ADMIN>#b errorlog /sys/errlog /etc/error.log |pr>/dev/lp The above command will create a macro named "errorlog", which can be used on future service calls, or by the customer before the service representative arrives to obtain a hard copy of the errorlog. If the printout is desired to be rerouted to a different printer than LP change the macro to suit your needs for the paticular /dev/xx (xx = printer ID) The above macro will work with 7.1 thru 7.3 O.S. If on 7.3 OS usage of options such as -mem, -tape, -term, -printer, -comm, -disk -text. could be added to the macro which would print specific type errors Example: ADMIN>#b errorlog /sys/errlog /etc/error.log -disk -text |pr>/dev/lp This would create a macro named "errorlog" that would print out disk errors in text format. This Macro procedure is a contribution of Allan Mollyhorn. o Instructions for ordering SSN tapes and diskettes for SPX products Contact Cay (Caroline) Williams in software distribution when in need of new or replacement SSN tapes and diskettes for SPX products. Send OMS to attention of (WILLIC) or call at 714-730-3085. ORIGINATOR: H. Mitchell HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB002 Pg001 FIB 00003 07/17/90 *** Lowest Acceptable Revision Levels (LARL) *** LARL is the recommended level at which field personnel should maintain their spare parts inventory. This level will insure correct system operation when a board is replaced. LARL is NOT the lowest operational level. The lowest revision level at which a system will operate may be lower than the LARL. A system operating properly should not be upgraded to LARL unless a board fails. USED ON SYSTEMS 2 3 4 2 0 0 0 5 SPARES 0 0 0 0 ICN VENDOR P/N DESCRIPTION LARL NOTES 0 0 0 0 -------- ---------- -------------------- ---- ------ ---------- . MM531010 903441-001 PCBA, CMB (old style) BH x MM531011 903441-002 PCBA, CMB (new style) BH x MM531030 903368-001 PCBA, 256K Memory A x MM531040 903390-001 PCBA, 4-Way Controller W x x x x MM531050 907649-001 PCBA, Adaptec Controller E x MM531060 903405-001 PCBA, Lan Controller D x x x x MM531070 903406-001 PCBA, MCS Controller Y NOTE 1 x x x 903406-002 PCBA, MCS Controller AB NOTE 1 x x x x MM531100 903439-001 PCBA, WDC Adapter Bd. B x MM531150 903496-001 PCBA, 1-board WD Controller R x x x x MM532010 903528-001 PCBA, 1/2 Streamer Bus Adapter A x x x x MM532020 400591-001 PCBA, 1/2 Streamer Formatter C x x x x MM533000 907472-001 Power Supply Assembly 2000 E x MM532525 907472-003 Power Supply Assembly 2500 K x MM535010 907897-001 Power Supply Asm 3000/4000 E x x MM535025 903593-001 PCBA, CMB 3000 (old style) AA* x MM537025 903631-001 PCBA, CMB 3000 (new style) U x x MM532520 903631-003 PCBA, CMB 2500/3000/4000 AC* x x x MM532550 903631-004 PCBA, CMB 2500/3000/4000 AE x x x MM532551 903631-005 PCBA, CMB 2500/3000/4000 AE NOTE 2 x x x MM535030 903533-001 PCBA, 1MB MEMORY 3000 D x x x MM535040 903577-001 PCBA, 8-WAY CONTROLLER G x x x MM537030 903575-001 PCBA, 4MB Memory D x x MM537055 903645-001 PCBA, Expansion Interface EIA C x MM537050 903643-001 PCBA, Expansion Interface EIB A x MM535063 903639-001 PCBA, 1/2 Tape Controller F x x MM535080 903667-001 PCBA, Dual SCSI Disk Contrl A x x NOTE 1: Board revision "AB" should contain prom P/N 165047-086 in location 4N to support the Tandberg 1/4 cartridge streamer drive. NOTE 2: P/N 903631-005, Revision "AB" or higher is required to support SCSI disk drives. * Indicates changes/additions since last report. ORIGINATOR: Harvey Mitchell HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB003 Pg001 HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB003 Pg002 FIB 00004 12/19/88 *** Error Halt Codes *** Original Document Name - Worldwide Product Support Field Bulletin 286 Error halt codes are error values that correspond to conditions that cause an abnormal system shutdown (i.e. error halt or crash). Below is a list of error halt codes. Each code has a name (e.g. HFKWRPROC), a reason and cause of the error halt and a possible workaround if one is available. This bulletin is only a guide, and in most cases a dump of the system is required to fully determine the cause of the error halt. For this reason a dump should be taken of the system and analyzed to determine the exact cause of the error halt. NOTE: Missing numbers are reserved for future error halt codes. Code Name Reason, Cause, and Possible Workaround -------------- ----------------------------------------------------------------. 0 This halt is caused by the CPU. It can be a problem with either the hardware, (bad terminal, board or other hardware components) or software (Operating System type software problem, not customer application software). WORKAROUND: Check the system for hardware problems and get a copy of the system dump. 5 HFKWRPROC If while writing a parent process to the swapper, because of not enough space in memory for both the parent and child and a disk error occurs, then the system will halt. WORKAROUND: Check the disk for possible errors/problems. This halt can be caused by a bad spot in the swapper partition requiring the disk to be reformatted. 6 HCOUNTERR If while copying a parent process's eventcount entries and there are not any the system will halt. WORKAROUND: Increase the 'events=' parameter in the 'vconf' system file. 7 HGROW If while writing a process to the swapper during the growing of its text, data, or stack, a disk error occurs, the system will halt. WORKAROUND: Check the disk for possible errors/problems. This halt can be caused by a bad spot in the swapper partition requiring the disk to be reformatted. 9 HEXECFAIL If execution of '/etc/start' fails during system boot, the system will halt. WORKAROUND: Root to the boot partition and re-install '/etc/start' to the root partition. (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB004 Pg001 10 HNORMAL During graceful shutdown after all processing is stopped, the system will halt. WORKAROUND: This is a normal halt. 11 HLISTOVEL If while freeing a chunk of memory space or swap space the free chunk list becomes full the system will halt. WORKAROUND: Allocate more 'procs='; the max number of free chunks are determined by the number of 'procs='. 14 HSWAPRD If any errors occur while reading in swapper processes the system will halt. WORKAROUND: Check the disk for possible errors/problems. This halt can be caused by a bad spot in the swapper partition requiring the disk to be reformatted. 15 HSWAPWR If any errors occur while writing a process to the swapper the system will halt. WORKAROUND: Check the disk for possible errors/problems. This halt can be caused by a bad spot in the swapper partition requiring the disk to be reformatted. 16 HSWAPSTART If during the boot process the swapper is unable to start, because the system cannot make a new process due to not enough process configured, then the system will halt. WORKAROUND: Boot to the boot partition and increase the number of 'procs=' or boot off the base tape and use option 'B' to re-install the boot partition. 17 HBASYNCON If a system buffer that was used for async I/O is put back into the buffer cache the system will halt. WORKAROUND: This should never happen; if so a system dump is required. 19 HECWTOVEL If a process goes to sleep waiting on more than three events then the system will halt. WORKAROUND: This should never happen; if so, a system dump is required. 20 HREQERR If adding an eventcount to the eventcount chain, and that eventcount is already in the chain the system will halt. WORKAROUND: This should not happen; system dump required. 21 HALLOCMGM If the free chunks of free memory or swap space overlap while merging two adjacent pieces of these the system will halt. WORKAROUND: This should not happen; system dump required. 23 HBOIWT If a process waiting on I/O wakes up before the I/O has completed the system will halt. WORKAROUND: This should not happen; system dump required. (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB004 Pg002 24 HIOWTERR If while waiting for buffered I/O, the buffer's flag does not indicate that it is in progress the system will halt. WORKAROUND: This should not happen; system dump required. 25 HSYSPROC If during the boot process, the first process (i.e."user()"), is unable to start because of no more process then the system will halt. WORKAROUND: Boot to the boot partition and increase the 'procs=' parameter in the 'vconf' system file. Or boot off the base tape and use option 'B' to re-install the boot partition. 26 HONIOERR If the system is unable to read a character from the system console the system will halt. WORKAROUND: Check the system console for possible errors/problems. Also, check the NVRAM and terminal to insure they are set up properly. 27 HTTYOPEN If during the boot process the system is unable to open the system console the system will halt. WORKAROUND: Check the sytem console for possible errors/problems. Also, check or re-install the '/etc/ports' and '/etc/terminals' files. Check to see if '/dev/tty0' exists. 28 HUNMOUNTPRNT If the parent directory of a file or directory is not mounted the system will halt. WORKAROUND: This should not happen; system dump required. 29 HTRAPERR If the "setperr()" system routine was called to set a process error flag and there was no error the system will halt. WORKAROUND: This should not happen; system dump required. 30 HBADCOPY If while copying a process in memory due to a fork or grow, the parent's uarea magic number is corrupted the system will halt. WORKAROUND: Excessive terminal I/O interrupts, hardware/software memory corruption are the usual causes. 31 HBADRDPROC If while reading a process into memory from the swapper, its uarea magic number is corrupted the system will halt. WORKAROUND: Excessive terminal I/O interrupts, hardware/software memory corruption are the usual causes. 32 HBADWRPROC If while writing a process to the swapper, its uarea magic number is corrupted the system will halt. WORKAROUND: Excessive terminal I/O interrupts, hardware/software memory corruption are the usual causes. (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB004 Pg003 33 HBDEVOVFL If while adding BFS devices, during boot, there was not enough memory allocated for all the devices the system will halt. WORKAROUND: Deconfigure some of your ports. Also, check 4 ways/ 8 ways for proper connectivity. 35 HBADDR If the system has an invalid confiuration the system will halt. WORKAROUND: This should not happen; system dump required. 36 HBADCONF If during boot time the configuration record does not decode correctly the system will halt. WORKAROUND: Re-install the system's security record ensuring it matches the system's hardware configuration. NOTE: Level 7.2A would sometimes not decode the configuration record properly. In these cases the system should be upgraded to a level 7.2B or greater. 37 HBADRDYQ If there are not any prcesses ready to execute during dispatch the system will halt. WORKAROUND: This should not happen; system dump required. 38 HDCFREE Directory Cache Entry already free: This error occurs if the operating system attempts to free a subpath entry from the directory cache entries which are already free. WORKAROUND: A system dump is required, also contact Product Support. 39 HBADRDYQ Bad ready queue. You should not receive this code. WORKAROUND: Contact Product Support. 40 HEVNOTPND Eventcount not pending. This error occurs when an eventcount is not found. WORKAROUND: A system dump is required, also contact Product Support. 42 HBUFFREE Cached buffer already free. This error occurs if the operating system attempts to free an already free buffer. WORKAROUND: A system dump is required, also contact Product Support. 43 HVC This code represents a possible memory management problem. WORKAROUND: A system dump is required, also contact Product Support. 44 HTIMEOUT This code occurs when a serial port controller fails to respond to the system programming commmands within an expected time limit. WORKAROUND: Check the switches and run the diagnostics on the serial port controller. The board number on which the timeout occurred is displayed in the error message. (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB004 Pg004 45 HBUSY This code is caused by incorrect DMA arbitration on one or more of the controllers (4-way, 8-way, or WDC controller). WORKAROUND: Check the switches on the 4-way, 8-way, and WDC controllers. 46 HOVERLAP This code occurs when the operating system attempts to send a command to a 4-way or 8-way controller while another command is already in progress. WORKAROUND: Check the switches on 4-way, 8-way, and WDC controlers. ORIGINATOR: W. Moore/J. Shreve/N. Jones HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB004 Pg005 FIB 00005 12/05/90 *** Disk Drive Information *** The purpose of this section of the handbook is to provide handy reference to commonly needed information for disk drives attached to the 2000/3000 systems. * General information: MFG. FORMATTED NACS # OF MANUFACTURER MODEL CAPACITY I.C.N. HEADS ------------ ------ --------- ------ ----- 20MB Rodime RO 204 22MB TT524100 8 40MB Rodime RO 203E 33MB MM534055 6 53MB Rodime RO 204E 44MB MM534060 8 50MB Micropolis 1304 43MB MM534020 6 85MB Maxtor XT 1085 71MB MM534070 8 105MB Maxtor XT 1105 88MB MM534030 11 140MB Maxtor XT 1140 120MB MM534040 15 190MB Maxtor XT 2190 160MB MM534080 15 380MB Maxtor (SCSI) XT 8380S 354MB MM295100 8 o RODIME o For additional information, see FIB group 4122. o To decode the flashing LED failure code, see FIB #2 of the 4122 FIB group o MICROPOLIS o For additional information, see FIB group 4143. o MAXTOR o For additional information, see FIB group 4120. * Drive addressing and termination information: *** NOTE *** A Bus terminator is required on the last drive attached to each WD controller o Rodime Drive addressing is accomplished via a switch accessable through the left side of the drive. Use the following chart: SWITCH DRIVE DRIVE POSITION 0 1 -------- ----- ----- 1 CLOSED OPENED 2 OPENED CLOSED 3 OPENED OPENED 4 OPENED OPENED The bus terminator is a resistor pack located near connector P/J1. (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB005 Pg001 Rodime RO 204E switches are spring wires which are placed under a hook to close the "switch", the following figure shows the switch and terminator locations. ___________________________________________________________________ | | | Rodime RO 204E End View | | | | | | | | S S S S _______________ | | 4 3 2 1 |1 Terminator | _____ | | /?/?/?/? |_______________| | P3 | | _________________________________________________________|_____|___ | P2 1|___| P1 1|_______________________| |_____________| |_________________________| o Micropolis Drive addressing is accomplished via jumpers W1 through W8 which are located near connector P1. Use the following chart: JUMPER DRIVE DRIVE POSITION 0 1 -------- ----- ----- W1 IN IN W2 IN IN W3 IN OUT W4 OUT IN W5 OUT OUT W6 OUT OUT W7 OUT OUT W8 IN IN The bus terminator is located near connector J1. It is called RN1 and the part is SIP 166-0011-1. The PCBA must be swung up in order to set addresses and check bus termination. o Maxtor (All models except the XT 8380S 380MB SCSI drive) Drive addressing is accomplished via jumpers which are located near connector P1. Use the following chart: * PRE-SERIES 2 DRIVES * | * SERIES 2 DRIVES * | JUMPER DRIVE DRIVE | JUMPER DRIVE DRIVE POSITION 0 1 | POSITION 0 1 -------- ----- ----- | -------- ----- ----- 1-C IN OUT | 6-5 IN OUT 2-C OUT IN | 4-5 OUT IN 3-C OUT OUT | 2-3 OUT OUT 4-C OUT OUT | 1-2 OUT OUT The bus terminator is a resistor pack located near the address jumpers. (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB005 Pg002 o Maxtor XT 8380S 380MB SCSI drive The drive address on the embedded SCSI drive is always set to 5, regardless to the drive position in the SPx System. Set the address as follows: JP-35 - IN JP-36 - OUT JP-37 - IN All SCSI drives must contain terminators. These are located adjacent to connector J1 and are labled RN17, RN18 and RN19. When oriented properly, the Pin 1 end (has a dot on the terminator chip) will be the end nearest to pin 1 of J1. The following is a list of the correct jumpering on this drive: JP-10 - IN JP-11 - IN JP-14 - IN JP-15 - OUT JP-16 - OUT JP-17 - OUT JP-18 - OUT JP-26 - OUT JP-32 - IN JP-33 - IN JP-34 - OUT JP-35 - IN JP-36 - OUT JP-37 - IN JP-38 - OUT JP-39 - IN JP-40 - IN JP-41 - IN ORIGINATOR: Harvey Mitchell MODIFIED BY: Norm Jones HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB005 Pg003 FIB 00006 09/21/94 *** CMB PCBA - 2000 System - Early Style *** The following information applies to this PCBA: MM531010 - Early Style CMB, 2000 System - 903441-001 * Serial ports jumpers settings: PORT 0 Terminal/Printer jumper setting: JMP A: 1-3, 2-4, 5-6, 7-9, 8-10, 11-13, 12-14 Modem jumper settings: JMP A: 1-2, 3-4, 7-8, 9-10, 11-12, 13-14 PORT 1 - RS 232 Terminal/Printer jumper settings: JMP B: 1-3, 2-4, 5-6, 7-9, 8-10, 13-15, 14-16 JMP G: 15-16, 17-18, 19-20, 23-24 JMP K: 1-2 JMP H: none Modem jumper setting: JMP B: 1-2, 3-4, 7-8, 9-10, 13-14, 15-16 JMP G: 15-16, 17-18, 19-20, 23-24 JMP K: 1-2 JMP H: none PORT 1 - RS 422 JMP G: 1-2, 3-4, 5-6, 7-8, 9-10, 11-12, 13-14, 21-22 JMP B: 11-12 JMP H: 2-3 The following jumpers must be disconnected: JMP G: 15-16, 17-18, 19-20 JMP K: 1-2 * Miscellaneous jumpers: C: none (1-2 etched) D: 2-3 E: 1-2, 7-8 F: 1-2 L: 1-2 M: 1-2 N: 1-2 P: none (1-2 etched) R: none S: none * JMP-S1, located near the RESET/NMI switch on the back of the CMB can be set as follows: 1-2 causes an interrupt (dump) when the RESET/NMI switch is pressed. 2-3 causes a RESET when the RESET/NMI switch is pressed. ORIGINATOR: H. Mitchell HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB006 Pg001 FIB 00007 06/05/89 *** CMB PCBA - 2000 System - Late Style *** The following information applies to this PCBA: MM531011 - Late Style CMB, 2000 System - 903441-002 Serial Ports configuration between a TERMINAL/PRINTER or MODEM is determined by the orientation of the "zero-ohm-packs". * Serial port configuration: PORT 0 TERMINAL/PRINTER jumper set-up: Pin 9 of zero-ohm-pack is in pin 1 of socket A. MODEM jumper set-up: Pin 1 of zero-ohm-pack is in pin 1 of socket A. PORT 1 - RS 232 TERMINAL/PRINTER jumper setting: Pin 9 of zero-ohm-pack is in pin 1 of socket B Pin 1 of zero-ohm-pack is in pin 1 of socket K Nothing in sockets G & H MODEM jumper set-up: Pin 1 of zero-ohm-pack is in pin 1 of socket B Pin 1 of zero-ohm-pack is in pin 1 of socket K Nothing in sockets G or H PORT 1 - RS 422 Pin 1 of the zero-ohm-pack is in pin 1 of socket G Pin 1 of the zero-ohm-pack is in pin 1 of socket H Nothing in sockets B or K * Miscellaneous jumpers: C: none (1-2 etched) D: 2-3 E: 1-2, 7-8 F: 1-2 L: 1-2 M: 1-2 N: 1-2 P: none (1-2 etched) R: none S: none V: 1-2 W: none X: 1-2 T: none (1-2 etched) on Rev. AN and lower T: 2-3 on Rev. AP and higher * JMP-S1, located near the RESET/NMI switch on the back of the CMB can be set as follows: 1-2 causes an interrupt (dump) when the RESET/NMI switch is pressed. 2-3 causes a RESET when the RESET/NMI switch is pressed. HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB007 Pg001 ORIGINATOR: H. Mitchell HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB007 Pg002 FIB 00008 01/08/90 *** CMB PCBA - 2500, 3000 & 4000 Systems *** The following information applies to these PCBAs: MM535025 - CMB, 3000 System - (early style) 903593-001 MM537025 - CMB, 3000/4000 System - (late style) 903631-001 MM532520 - CMB, 2500/3000/4000 System - 903631-003 MM532550 - CMB, 2500/3000/4000 System - 903631-004 MM532551 - CMB, 2500/3000/4000 System - 903631-005 *** JUMPERS TO CONTROL THE SERIAL PORTS *** Jumpers 29A, 28C, 26B, and 29B are zero-ohm-packs where the orientation of the zero-ohm-packs determine the configuration of the serial ports A and B. Jumpers 29A, 26B and 29B are inserted in standard dip sockets with standard numbering system. Jumper 28C is inserted in special socket 28C that occupies two locations, 28C and 29C, and will be referred to as 28/29C. Socket 28/29C has a unique 3-row configuration, see diagram below: As viewed from the Rear: 17 24 o o o o o o o o Row 28 9 16 o o o o o o o o Row 29 1 8 o o o o o o o o Column C * Configure the serial ports as follows: SERIAL PORT A: MODEM TERMINAL/PRINTER Pin 1 of the zero-ohm-pack is Pin 1 of the zero-ohm-pack is inserted in pin 1 of socket 29A. inserted in pin 9 of socket 29A. SERIAL PORT B - RS 232: MODEM TERMINAL/PRINTER Pin 1 of the zero-ohm-pack is Pin 1 of the zero-ohm-pack is inserted in pin 1 of socket 28/29C inserted in pin 16 of socket 28/29C Pin 1 of the zero-ohm-pack is Pin 1 of the zero-ohm-pack is inserted in pin 1 of socket 26B inserted in pin 1 of socket 26B (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB008 Pg001 Nothing in 29B Nothing in 29B SERIAL PORT B - RS-422 (X.21) Pin 1 of the zero-ohm-pack is inserted in pin 9 of socket 28/29C Pin 1 of the zero-ohm-pack is inserted in pin 1 of socket 29B Nothing in 26B * Set the miscellaneous function jumpers as follows: Loc. Loc. 16F JMP 1-2 29G JMP 25-26 (see NOTE 1) 14G JMP 3-4 1S JMP 29-30 17S JMP 6-7 13N JMP 31-32 17S JMP 12-13 (May be hard wired) 13N JMP 33-34 17S JMP 14-15 (May be hard wired) 13N JMP 36-37 17S JMP 16-17 (May be hard wired) 6R JMP 39-40 17S JMP 18-19 (May be hard wired) 12H JMP IC PAD 4-11 *** NOTE 1 *** Jumpers 24, 25 and 26 are used to define the function of the NMI/RESET switch located on the front panel: o When points 25 and 26 are jumpered together, a system reset will occur when the NMI/RESET switch is depressed. o When points 24 and 25 are jumpered together, a Non-Maskable Interrupt will occur when the NMI/RESET switch is depressed. This will normally force a system dump. * On 903631-001 and 903631-003 only: JMP 47-48 Loc. 7K * On 903631-004 and 903631-005 only: JMP 46-47 Loc. 7K * P/N 903631-005 is required when SCSI drives are in use. * Jumpers EX1, EX2 and EX3 must have no jumpers installed. (903631-001 will not have EX1, EX2, EX3) * If MMB board is installed: JMP 50-52 Loc. 19S * If PMMU IC is installed: JMP 50-51 Loc. 19S *** FUNCTIONS OF CMB DIP-SWITCH - SW2 *** The CMB dip-switch, SW2 can be set to perform various unique operations or to loop in a specific self-test diagnostic section. The normal setting for this switch is SW2-9 ON ,ALL OTHERS OFF. The following is a chart showing the functions of this switch: Location 17H -----------------SWITCH SETTING LEGEND----------------- : : : C = Closed or On O = Open or Off X = Variable : ------------------------------------------------------- HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB008 Pg002 (continued on next page) SW2 SWITCH POSITIONS 10 9 8 7 6 5 4 3 2 1 RESULTING ACTION -- - - - - - - - - - ---------------- O C O O O O O O O O Normal self-test, initialization and boot O C O O O O O O O C Enter keyboard control mode for command entry O C O O O O O O C O Enter MDS keyboard control mode O C O O O O O O C C Alternate bootstrap from a different device O C O O O O O C O O Enter config mode to configure the NVRAM O C O O O O O C O C Force extended memory tests to be executed O C O O O X X X X X All settings not shown above are reserved O C O O C O O O O O Pass control to the EDC test module O C O O C O O O O C Execute EDC test 01 O C O O C X X X X X Execute EDC test # set in switches 1-5 O C O C O O O O O O Pass control to Memory test module O C O C O O O O O C Execute Memory test 01 O C O C O X X X X X Execute Memory test # set in switches 1-5 O C O C C O O O O O Pass control to MMU test module O C O C C O O O O C Execute MMU test 01 O C O C C X X X X X Execute MMU test # set in switches 1-5 O C C O O O O O O O Pass control to CMB test module O C C O O O O O O C Execute CMB test 01 O C C O O X X X X X Execute CMB test # set in switches 1-5 O C C O C O O O O O Pass control to WDC control module O C C O C O O O O C Execute WDC test 01 O C C O C X X X X X Execute WDC test # set in switches 1-5 O C C C O O O O O O Pass control to Cache test Module O C C C O O O O O C Execute CMB test 01 O C C C O X X X X X Execute CMB test # set in switches 1-5 O C C C C O O O O O Pass control to Controllers test module O C C C C O O O O 1 Execute Controllers test 01 O C C C C X X X X X Execute Controllers test # set in switches 1-5 NOTE: CMB MM535025/903593-001 will not work in the 4000 system. * The Amber LED, DS1 indicates a CPU halted condition. M/OORIGINATOR: H. Mitchell/N. Jones /Allan Mollyhorn HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB008 Pg003 FIB 00009 MAI COMPANY CONFIDENTIAL 09/30/93 *** Eight-way Controller PCBA - 2500,3000 & 4000 Systems *** The following information applies to this PCBA: MM535040 - Eight-way controller - 903577-001 -----------------SWITCH SETTING LEGEND----------------- : : : C = Closed or On O = Open or Off X = Not Used : ------------------------------------------------------- ------------------------------------------------------------------------------- 2500 & 3000 SYSTEM 8-WAY SWITCH SETTINGS * Switch SW-1, PCBA Address and DMA Arbitration: PCBA ADDRESS DMA ARBITRATION BOARD S1-1 S1-2 S1-3 S1-4 S1-5 S1-6 S1-7 S1-8 S1-9 S1-10 ----- ---- ---- ---- ---- ---- ---- ---- ---- ---- ----- 0 C O C O O O O C C C 1 O O C O O O C C C C 2 C C O O O C O C C C 3 O C O O O C C C C C 4 C O O O C O O C C C 5 O C C O C O C C C C ------------------------------------------------------------------------------- 4000 SYSTEM 8-WAY SWITCH SETTINGS * Switch SW-1, PCBA Address and DMA Arbitration: PCBA ADDRESS DMA ARBITRATION ------------------- ------------------------------ BOARD S1-1 S1-2 S1-3 S1-4 S1-5 S1-6 S1-7 S1-8 S1-9 S1-10 LOCATION MINOR #s ----- ---- ---- ---- ---- ---- ---- ---- ---- ---- ----- -------- -------- . 0 C O C O O O O C C C BU 0-7 1 O O C O O O C C C C BU 8-15 2 C C O O O C O C C C BU 16-23 3 O C O O O C C C C C BU 24-31 4 C O O O C O O C C C BU 32-39 5 O C C O C O C C C C BU 40-47 (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB009 Pg001 PCBA ADDRESS DMA ARBITRATION ------------------- ------------------------------ BOARD S1-1 S1-2 S1-3 S1-4 S1-5 S1-6 S1-7 S1-8 S1-9 S1-10 LOCATION MINOR #s ----- ---- ---- ---- ---- ---- ---- ---- ---- ---- ----- -------- -------- . 6 C O C O O O O C C C EU 48-55 7 O O C O O O C C C C EU 56-63 8 C C O O O C O C C C EU 64-71 9 O C O O O C C C C C EU 72-79 10 C O O O C O O C C C EU 80-87 11 O C C O C O C C C C EU 88-95 4000 System NOTES: 1. The addressing and DMA arbitration settings are the same for DIFFERENT board numbers depending upon the placement in the BASE UNIT (BU) or in the EXPANSION UNIT (EU). 2. If there are less than six 8-ways in the base unit and there are 8-ways in the expansion unit, the first 8-way in the expansion unit will still be board number six. There will be a gap in the board numbers and in the minor numbers if this situation exists. ------------------------------------------------------------------------------- 2500,3000 & 4000 SYSTEM COMMON INFORMATION * Individual Port Program Jumpering, PGM 0 through PGM 7: TERMINAL PRINTER MODEM -------- ------- ----- 1 to 3 1 to 3 1 to 2 2 to 4 2 to 4 3 to 4 5 to 6 5 to 6 5 to 6 7 to 8 7 to 8 7 to 8 9 to 11 9 to 11 9 to 10 10 to 12 10 to 12 11 to 12 13 to 15 13 to 15 13 to 14 14 to 16 14 to 16 15 to 16 * PROM/RAM/CLOCK Jumpering JMPs A,B,C and D: JUMPER SETTING DESCRIPTION ------ ------- ----------- JMP A 1 to 2 PROM size: 16K bytes (default) 2 to 3 PROM size: 32K bytes JMP B 1 to 2 RAM size: 8K bytes 2 to 3 RAM size: 32K bytes (default) JMP C No Connection Eight-way configuration (default) 1 to 2 Reserved JMP D 1 to 2 8 MHz Z80H clock (default) 2 to 3 4 MHz Z80H clock * The green LED illuminates to indicate normal operation. * See 4116 FIB #14 for additional information. (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB009 Pg002 * CABLE, DRIVER/RECEIVER IC INFORMATION: ICs used 8-way 8-way QUADPUS Driver Receiver Port Connector Cable End 1488 1489 ------------------------------- -------- ---- 0 J4 (LOWER) PB 2C,1B,2D 2F 1 J4 (LOWER) PC 1B 1A 2 J4 (LOWER) PD 2D 2E 3 J4 (LOWER) PE 2C 2B 4 J3 (UPPER) PB 6B,4B,5B 3B 5 J3 (UPPER) PC 4B 3A 6 J3 (UPPER) PD 5B 5A 7 J3 (UPPER) PE 6B 6A NOTE: When multiple driver ICs listed for one port, the first one listed controls transmit data. 8 WAY SERIAL INTERFACE SIGNAL PIN-OUTS SIG NAME PORT 0 PORT 1 PORT 2 PORT 3 DB-25 PCBA OCTPUS PCBA PCTPUS PCBA OCTPUS PCBA OCTPUS -------- ------ ------ ------ ------ ------ ------ ------ ------ ----- TX DATA J4-15 PB-01 J4-11 PC-01 J4-06 PD-01 J4-02 PE-01 2 RX DATA J4-16 PB-02 J4-12 PC-02 J4-07 PD-02 J4-03 PE-02 3 RTS J4-17 PB-03 J4-13 PC-03 J4-08 PD-03 J4-04 PE-03 4 CTS J4-18 PB-04 J4-14 PC-04 J4-09 PD-04 J4-05 PE-04 5 DSR J4-34 PB-05 J4-29 PC-05 J4-25 PD-05 J4-20 PE-05 6 GND J4-35 PB-06 J4-30 PC-06 J4-26 PD-06 J4-21 PE-06 7 DCD J4-36 PB-07 J4-31 PC-07 J4-27 PD-07 J4-22 PE-07 8 DTR J4-37 PB-09 J4-32 PC-09 J4-28 PD-09 J4-23 PE-09 20 SIG NAME PORT 4 PORT 5 PORT 6 PORT 7 DB-25 PCBA OCTPUS PCBA PCTPUS PCBA OCTPUS PCBA OCTPUS -------- ------ ------ ------ ------ ------ ------ ------ ------ ----- TX DATA J3-15 PB-01 J3-11 PC-01 J3-06 PD-01 J3-02 PE-01 2 RX DATA J3-16 PB-02 J3-12 PC-02 J3-07 PD-02 J3-03 PE-02 3 RTS J3-17 PB-03 J3-13 PC-03 J3-08 PD-03 J3-04 PE-03 4 CTS J3-18 PB-04 J3-14 PC-04 J3-09 PD-04 J3-05 PE-04 5 DSR J3-34 PB-05 J3-29 PC-05 J3-25 PD-05 J3-20 PE-05 6 GND J3-35 PB-06 J3-30 PC-06 J3-26 PD-06 J3-21 PE-06 7 DCD J3-36 PB-07 J3-31 PC-07 J3-27 PD-07 J3-22 PE-07 8 DTR J3-37 PB-09 J3-32 PC-09 J3-28 PD-09 J3-23 PE-09 20 ORIGINATOR: Norm Jones HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB009 Pg003 FIB 00010 04/25/89 *** Four-way Controller PCBA *** The following information applies to this PCBA: MM531040 - Four-way controller - 903390-001 -----------------SWITCH SETTING LEGEND----------------- : : : C = Closed or On O = Open or Off X = Not Used : ------------------------------------------------------- * Switch SW-1, PCBA Address and DMA Arbritration: DMA ARBITRATION PCBA ADDRESS BOARD S1-1 S1-2 S1-3 S1-4 S1-5 S1-6 S1-7 S1-8 S1-9 S1-10 ----- ---------------------------------- ----------------------- 1 C O O O O O O C O C 2 C O O C O O O C O O 3 C O O C C O O O C C 4 C O O C C C O O C O * Individual Port Program Jumpering PGM1 through PGM4: PCBA Artwork 904741-001: TERMINAL PRINTER MODEM -------- ------- ----- 1 to 7 1 to 7 1 to 2 2 to 8 2 to 8 3 to 4 3 to 4 3 to 4 7 to 8 9 to 10 9 to 10 9 to 10 13 to 15 13 to 15 13 to 14 14 to 16 14 to 16 15 to 16 PCBA Artwork 904943-001: TERMINAL PRINTER MODEM -------- ------- ----- 1 to 3 1 to 3 1 to 2 2 to 4 2 to 4 3 to 4 9 to 11 9 to 11 9 to 10 10 to 12 10 to 12 11 to 12 13 to 15 13 to 15 13 to 14 14 to 16 14 to 16 15 to 16 (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB010 Pg001 P/N 904943-001 has two additional jumpers: Jump 7-8 to pull up Data Carrier Detect Jump 5-6 to pull up Data Set Ready These two jumpers may be used to satisfy Modem configurations when they cannot be supplied through the communications cable. Normally the jumpers should be left in place. * The green LED illuminates to indicate normal operation. ORIGINATOR: H. Mitchell HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB010 Pg002 FIB 00011 02/10/90 *** LAN Controller PCBA *** The following information applies to this PCBA: MM531060 - Local Area Network controller - 903405-001 -----------------SWITCH SETTING LEGEND----------------- : : : C = Closed or On O = Open or Off X = Variable : ------------------------------------------------------- * Address Switch, SW1: SW1-1 is the BIAS switch for the network. This switch must be in the OFF position on one and ONLY one of the LAN controllers connected in the network. All othe LAN controllers within the network must have SW1-1 set to ON. SW1-2 is the TERMINATOR switch for each end of the network. Only two controllers, one at each end of the network, may have SW1-2 in the ON position. All other LAN controllers must have SW1-2 set to OFF. This switch and it's associated circuitry eliminates the need for the 100 Ohm terminating resistors within the tap boxes. SW1-3 through SW1-8 are the LAN PCBA Station Address (node) switches. SW1-8 is the least significant bit and SW1-3 is the most significant bit. Each LAN controller connected in the network must have a unique address. Address 00000 is invalid. Use the chart shown below to set the LAN PCBA Station Address: BIT VALUE 32 16 8 4 2 1 --------- ----- ----- ----- ----- ----- ----- ADDRESS SW1-3 SW1-4 SW1-5 SW1-6 SW1-7 SW1-8 ------- ----- ----- ----- ----- ----- ----- 01 C C C C C O 02 C C C C O C 03 C C C C O O 04 C C C O C C 05 C C C O C O 06 thru 62 X X X X X X (counts up binarily) 63 O O O O O O (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB011 Pg001 * BUS Arbitration and Board Address Switch, SW2: There may be up to two LAN controllers installed in the CCA. However, the second LAN PCBA is normally only used for diagnostics. Set the SW2 switches as follows: BUS ARBITRATION BOARD ADDRESS (Within CCA) SW2-1 SW2-2 SW2-3 SW2-4 SW2-5 SW2-6 SW2-7 SW2-8 O O O C C O C O 1st controller O O O C C C O C 2nd controller * 100 Ohm resistors are not required in the tap boxes. * The green LED illuminates to indicate normal operation. * The red LED is an activity indicator. ORIGINATOR: H. Mitchell MODIFIED by: N. Jones HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB011 Pg002 FIB 00012 09/01/87 *** Memory PCBA - 128K/256K - 2000 System *** The following information applies to these PCBAs: MM531030 - 256K Memory PCBA - 903368-001 128K Memory PCBA - 903368-002 -----------------SWITCH SETTING LEGEND----------------- : : : C = Closed or On O = Open or Off X = Not Used : ------------------------------------------------------- The following chart shows the settings for the Memory Address switch, S1: CAPACITY -> 128K 256K 256K --------------------------------------------------------------- PCB DETAIL -> Rev X3/A Rev X2 Rev X3/A --------------------------------------------------------------- SW1 POSITIONS -> 1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6 Address Range 000K to 128k O C C C C C 128K to 256K O O C C C C C C C C C C C C C C C C 256K to 384K O C C C C O 384K to 512K O O C C C O C C C C C O C C C C C O 512K to 640K O C C C O C 640K to 768K O O C C O C C C O C C C C C C C O C 768K to 896K O C C C O O 896K to 1024K O O C C O O C C O C C O C C C C O O 1024K to 1152K O C C O C C 1152K to 1280K O O C O C C C C C C O C C C C O C C 1280K to 1408K O C C O C O 1408K to 1536K O O C O C O C C C C O O C C C O C O ORIGINATOR: H. Mitchell HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB012 Pg001 FIB 00013 06/26/89 *** Memory PCBAs - 1 MB & 4 MB - 2500,3000 & 4000 Systems *** The following information applies to this PCBA: MM535030 - 1 Megabyte Memory PCBA - 903533-001 MM537030 - 4 Megabyte Memory PCBA - 903575-001 -----------------SWITCH SETTING LEGEND----------------- : : : C = Closed or On O = Open or Off X = Not Used : ------------------------------------------------------- o The chart below shows the settings for the Memory Address switch, S1: - 1 Megabyte PCBA - SETTINGS OF ADDRESS SWITCH, S-1 ADDRESS RANGE 8 7 6 5 4 3 2 1 ------------- - - - - - - - - 0 MB to 1 MB C C C C C C C C 1 MB to 2 MB C C C C C C C O 2 MB to 3 MB C C C C C C O C 3 MB to 4 MB C C C C C C O O 4 MB to 5 MB C C C C C O C C 5 MB to 6 MB C C C C C O C O 6 MB to 7 MB C C C C C O O C 7 MB to 8 MB C C C C C O O O 8 MB to 9 MB C C C C O C C C 9 MB to 10 MB C C C C O C C O 10 MB to 11 MB C C C C O C O C 11 MB to 12 MB C C C C O C O O 12 MB to 13 MB C C C C O O C C 13 MB to 14 MB C C C C O O C O 14 MB to 15 MB C C C C O O O C 15 MB to 16 MB C C C C O O O O (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB013 Pg001 - 4 Megabyte PCBA - SETTINGS OF ADDRESS SWITCH, S-1 ADDRESS RANGE 6 5 4 3 2 1 --------------- - - - - - - 0 MB to 4 MB........C C C C C C 4 MB to 8 MB........C C C C C O 8 MB to 12 MB........C C C C O C 12 MB to 16 MB........C C C C O O ....................................NOTES...................................... o 4 MB boards are not used in the 2500 system. o A maximum of four 1 MB boards may be installed on a 2500 system. o A maximum of six memory PCBAs may be installed in 3000 and 4000 systems. o 4 MB and 1 MB boards may be mixed in a system, but the 4 MB boards must be installed on 4 MB boundaries. ORIGINATOR: H. Mitchell HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB013 Pg002 FIB 00014 12/06/88 *** MTCS Controller PCBA *** The following information applies to these PCBAs: MM531070 - 1/4" Magnetic Tape Cartridge Streamer Controller - 903406-001 See NOTE 1 --> 903406-002 -----------------SWITCH SETTING LEGEND----------------- : : : C = Closed or On O = Open or Off X = Not Used : ------------------------------------------------------- * Switch SW-1, Mode and DMA Arbritration: BUS ARBITRATION MODE SW1-1 SW1-2 SW1-3 SW1-4 SW1-5 SW1-6 SW1-7 SW1-8 C C C C O O X C * The MODE switch, SW1-8 may be set to the OPEN position for drive burn-in. * The green LED illuminates to indicate normal operation. *** NOTE 1 *** PCBA 903406-001 will support only the MT-4403 Cipher Streamer. PCBA 903406-002 must be used to support the MT-4406 - Tandberg Streamer. This PCBA will also support the MT-4403 - Cipher Streamer. If the part number of the PROM at location 4N is 165047-086 then the PCBA is 903406-002. ORIGINATOR: H. Mitchell/N. Jones HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB014 Pg001 FIB 00015 09/01/87 *** MTS - Formatter Controller PCBA *** The following information applies this PCBA: MM532020 - MTS Formatter Controller - 400591-001 This PCBA is installed externally from the CCA, within the Formatter Controller box, MBF Model 4602. The proper set-up is as follows: * Dip SW. 1 at location 7W: all positions must be set to CLOSED. * Jumpers W22, W23, W24 at location 6J must be OPEN. * Jumper J16 at location 4G must be OPEN. * All other jumpers on the board must be ON. Jumpers are numbered from 1 to 33, jumper position 12 is not used. * Terminator Packs must be in place at locations 6H and 7B. * The PROMs in locations 6N and 6S on the 2000 CMB (903441-002) must be P/N 165047-084 and P/N 165047-085, respectfully, in order to support MCS boot *** NOTE *** Controller self-tests are started when the power switch is turn ON in the controller box. Self-tests may be re-started by pressing the Reset Switch on the formatter board. There are 8 Leds marked from '7' (MSB) thru '0'. The upper half should always display a '9' (bits 7 and 4). The lower half changes with the test being run. The LEDs should display: '9F' while doing Data Buffer Test. '90' at the end if the self tests ran properly. * See 2000 System FIB #24 and FIB #36 for additional information. ORIGINATOR: H. Mitchell HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB015 Pg001 FIB 00016 05/25/89 *** MTS - SCSI Bus Adapter PCBA *** The following information applies to this PCBA: MM532010 - SCSI Bus Adapter - 903528-001 -----------------SWITCH SETTING LEGEND----------------- : : : C = Closed or On O = Open or Off X = Not Used : ------------------------------------------------------- * Set the switches on S1 as follows: SW1-1 SW1-2 SW1-3 SW1-4 SW1-5 SW1-6 SW1-7 SW1-8 ----- ----- ----- ----- ----- ----- ----- ----- C C C C C C X X * There is a green LED which illuminates to indicate normal operation. * See 2000 System FIB #24 and FIB #36 for additional information. ORIGINATOR: H. Mitchell HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB016 Pg001 FIB 00017 09/01/87 *** WDC Adaptec and Adapter PCBA set *** The following information applies to these PCBAs: MM531100 - WDC Adapter - 903439-001 MM531050 - Adaptec Controller - 907649-001 The Hard Disk Controller for the System 2000 contains jumpers to set write pre-compensation according to the drives being used: The 2-board hard disk controller consists of one Basic Four board and one Adaptec board. The pre-compensation jumper is located on the Adaptec board near connector J1. This one jumper sets pre-compensation for BOTH drives. * If the system has only one drive or two drives of the SAME MANUFACTURER, the jumper is set as follows: DRIVE TYPE JUMPER WRITE PRE-COMPENSATION ---------- ------ ---------------------- Rodime R to T always on Micropolis R to S same cylinder as reduced write current Maxtor R to U always off Note: Jumper position R-U is equivalent to no jumper installed. * If the system has two drives of DIFFERENT MANUFACTURERS, set the pre- compensation jumper as follows: 1. If either of the drives is a Maxtor, set jumper to R-U, and do not perform step 2. 2. If either of the drives is a Micropolis, set jumper to R-S. * The Adaptec board also contains jumper block J5. It must contain no jumpers. * A green LED located on the WDC Adapter indicates normal operation. * A red LED located on the Adaptec Controller indicates an access problem. ORIGINATOR: H. Mitchell HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB017 Pg001 FIB 00018 04/24/89 *** WD Controller PCBA - single board *** The following information applies to this PCBA: MM531150 - Winchester Disk Controller - 903496-00X The Hard Disk Controller contains jumpers to set write pre-compensation according to the manufacturer of the drives being used: * Jumpers are located next to connector J3. * Each drive is set independently by a separate jumper: Jumper B is for drive 0 Jumper C is for drive 1 * Set Jumper B and C according to drive manufacturer. DRIVE TYPE JUMPER WRITE PRECOMPENSATION -------------- -------- --------------------- Maxtor 1A to 2A always off Rodime 1B to 2B always on Micropolis 1C to 2C same cylinder as reduced write current * The controller address is determined by two things: 1. The plugging of JUMPER "A" 2. The PAL chip installed in location 1J If the P/N of the PAL in location 1J is 911017-001 (MM535065), the JUMPER "A" must be plugged from pin 1 to pin 2. With this PAL, the WDC can be used as the controller for disk drives 0 and 1 (when mounted on the stack of the base unit) or disk drives 4 and 5 (when mounted on the stack of the 4000 expansion unit). This will set the board address to CCXXXX. If the P/N of the PAL in location 1J is 911017-007 (MM535070), the JUMPER "A". must be plugged from pin 2 to pin 3. With this PAL, the WDC can be used as the controller for disk drive 2 (when mounted on the stack of the base unit) or disk drives 6 and 7 (when mounted on the stack of the 4000 expansion unit). This will set the board address to CDXXXX. *** NOTE *** Due to cooling requirements, no more than three disk drives may be installed in the 3000/4000 CCA if a MTC tape unit is installed in that CCA. ORIGINATOR: H. Mitchell HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB018 Pg001 FIB 00019 12/06/89 *** System Serial Number change information *** *** System Serial Number (SSN) installation procedure *** 1. Shutdown and power-off the system and install the Security PAL chip as follows: System 2000: If the CMB is the early style, P/N MM531010/903441-001, the Security PAL is not required on the SSN change. If the CMB is the late style, P/N MM531011/903441-002, remove the existing IC from location 8Z of the CMB PCBA and install Security PAL chip, P/N MM538180/911016-004, in it's place. System 2500, 3000 or 4000: Security PAL chip, P/N MM538181/911005-082, is used on these systems. There are five different CMBs currently in the field. Remove the chip from the applicable location shown below and install the Security PAL chip in it's place: CMB P/N 903593-001 - location 20F CMB P/N 903631-001 or -003 or -004 or -005 - location 20G 2. Power-on the system. 3. ALTLOAD the system as follows: System 2000: Press the CTRL key and the C key simultaneously (on the VDT attached to port 0) during the memory portion of the self-tests. System 2500,3000 or 4000: Select Option 2, ALT LOAD 4. At the 'Boot device:' prompt, mount the SSN tape on the MCS or MTS and make the drive Ready and Online. Then type 'cs' (for MCS) or 'ts' (for MTS) and press CR. 5. At the 'System file:' prompt, type 'ssn' (lower case only) and press CR. 6. The tape will be booted and the system will display the proprietary message. Press CR. 7. The system will display the currently installed SSN (if any) and prompt for the new serial number. Type the new serial number (exactly as found on the configuration record tape - example: 2000-12345 ) and press CR. (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB019 Pg001 8. The system will display an eight-character LOCK number and prompt for the KEY. Authorization codes for formatting fixed disks and keys for installing system serial number obtainable as follows: (714) 730-2444 6:30 AM - 5:00 PM PST (714) 731-5100 After 5:00 PM PST or weekends or holidays the guard will answer the phone. Ask for Lock & Key Service. Give name, phone number and system. A Basic Four rep. will call back with the information. Contact Basic Four Software Distribution to obtain the KEY and enter the KEY, double check it by reading it back to the Basic Four representative and then press CR. The 2500,3000 and 4000 system will prompt for the O.S. level. Type the O.S. . level as found on the O.S. tape ( example: 7.3B ) and press CR. Failure to enter the correct O.S. level may prevent proper operation of the O.S. 9. The new SSN will be installed and the system will notify you of completion and shutdown. 10. Power-off the system and remove the Security PAL installed in step 1 and re-install the chip that was removed (this step is not necessary on the 2000 system with the early style CMB). 11. Re-boot the system. ORIGINATOR: H. Mitchell HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB019 Pg002 FIB 00020 02/03/89 *** Expansion Interface A PCBA - 4000 System *** The following information applies to this PCBA: MM537055 - Expansion Interface A PCBA - 903645-001 -----------------SWITCH SETTING LEGEND----------------- : : : C = Closed or On O = Open or Off X = Not Used : ------------------------------------------------------- * This PCBA is the top board on the stack in the Base chassis. It provides two additional parallel ports and interfaces between the CMB (via the stack) and the Expansion chassis. * Jumper Setting JMP1 is used to set the base address of 4C00. Install a jumper connecting pin 1 to pin 2. * Switch Settings Switch SW1 is used for Bus Arbitration and Interrupt Acknowledge Arbitration. Set SW1 as follows: SW1-4 SW1-3 SW1-2 SW1-1 C O C O - This setting applies to PCBAs below REV B *** C C C O - Correct setting for PCBAs at or above REV B *** NOTE: The LARL for the EIA PCBA is REV B, all PCBAs should be at or above REV B * Parallel Printer Cable Information The lower connector on the EIA PCBA is J5 for connection to printer 1. Printer 1 is assigned device MAJOR,MINOR of 8,100. The upper connector on the EIA PCBA is J6 for connection to printer 2. Printer 2 is assigned device MAJOR,MINOR of 8,101. ORIGINATOR: H. Mitchell HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB020 Pg001 FIB 00021 03/10/88 *** Expansion Interface B PCBA - 4000 System *** The following information applies to this PCBA: MM537050 - Expansion Interface B PCBA - 903643-001 -----------------SWITCH SETTING LEGEND----------------- : : : C = Closed or On O = Open or Off X = Not Used : ------------------------------------------------------- * This PCBA is the mounted within the Expansion chassis. It provides two additional parallel ports, interfaces between the Expansion chassis and the Base chassis and provides for stacking of additional controllers. * Jumper Setting JMP1 is used to control whether the Base unit is reset when the Expansion chassis is powered up. JMP1 CONNECTION RESET BASE UNIT --------------------------------------- From 1 to 2 Yes No Jumper No (normal setting) JMP2 is used to control whether the CMB is interrupted when the Expansion chassis is powered up. JMP2 CONNECTION INTERRUPT CMB ------------------------------------- From 1 to 2 Yes (normal setting) From 2 to 3 No * Switch Settings Switch SW1 is used to set the interrupt sub-level. Set SW1 as follows: SW1-4 SW1-3 SW1-2 SW1-1 C O C O ORIGINATOR: H. Mitchell HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB021 Pg001 FIB 00022 01/27/89 *** 2500/3000/4000 Boot Error Messages *** The power on/reset self tests are divided into two sections. The pretests which output errors only to the two digit hex display on the CMB PCB with a flashing error code, and the self tests which output status and error messages to the system console device (if the console controller passes the pretests). Pretest Flashing LED Hex Displays Hex Units Display Tested Error Meaning (when flashing) -------.-------- ------------------------------------------------------------- 01 CMB The CPU instructions test failed. 02 CMB The prom-checksum test failed. 03 CMB The NVRAM test failed. 04 CMB/MEM A DMA serial controller is configured in NVRAM, and there is no memory. 05 CMB The SCC loopback test failed, and the NVRAM specifies the SCC as the console device. 06 CMB/MEM The first 4K of memory has failed. 07 CMB/MEM The selected test requires memory and the first 4K of memory failed. 08 CMB/CTLR Bus error detected with no console to display the error. XX Any other blinking value indicates the vector number of an unexpected interrupt received prior to the system console ready to receive output. .............................................................................. Self Test Error Messages Displayed on the System Console 1) csss Error When a bad status word is received from the 1/4" streamer controller the following error message is displayed: csss-xxxx Where xxxx is the status word received from the controller. (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB022 Pg001 MCS Controller Status - normal status = 8000 bit # 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0 | | | | | | | | | | | | | | | | value 8 4 2 1 8 4 2 1 8 4 2 1 8 4 2 1 | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | *-Tape error | | | | | | | | | | | | | | *---Tape unit not ready | | | | | | | | | | | | | *-----Power fail/reset | | | | | | | | | | | | *-------Tape overflow, wrote past EOT | | | | | | | | | | | *----------Data transfer error | | | | | | | | | | *------------Read/write abort | | | | | | | | | *--------------Write protected | | | | | | | | *----------------File mark detected | | | | | | | *--------------------No cartridge | | | | | | *----------------------Bad data transfer | | | | | *------------------------Tape overflow, read past EOT | | | | *--------------------------Filler sent | | | *-----------------------------Chain processing terminated | | *-------------------------------IOPB parameter error | *---------------------------------End of data *-----------------------------------Operation successfully completed 2) tscs Error When an unexpected status is received from the 1/2" streamer tape controller status register during a load the following error message is displayed: tscs-xx-cc-ssssssss Where: xx = ending status 00 = good 02 = Check condition 08 = Busy 18 = Reservation conflict cc = Command sent to the controller 00 = test unit ready command 01 = rewind command 08 = read command 11 = space one filemark forward command (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB022 Pg002 ssssssss = sense data from the tape controller The first byte is the primary byte or sense key, described as follows: bit 7: Filemark bit 6: End of tape bit 5: Requested block length mismatch bit 4: Not defined bits 3-0: This field is called the sense key and is defined as follows: 0: No sense 1: Recovered error 2: Not ready 3: Media error 4: Hardware error 5: Illegal request 6: Unit attention 7: Write protected 8: Blank tape 9: Not defined A: Not defined B: Aborted command C: Not defined D: Volume overflow E: Miscompare F: Not defined 3) tsss Error When a handshake problem occurs between the 1/2" streamer tape controller and the MDS (Micro Diagnostic System) driver routines, the following message is displayed on the system console: tsss-eee Where ee is one of the following MDS error codes: ERROR CODE ERROR DESCRIPTION 20........Timeout waiting for bus free status of 00. 21........Timeout waiting for the command phase after select. 22........Timeout waiting for status or message phase (stuck in cmd phase) 23........Skipped to the message phase over status phase. 24........Timeout waiting for the message phase. 25........The message byte received was not zero as expected. 26........A TS parity error occurred between the adapter board and the tape drive. 4) wdss Error When a handshake problem occurrs between the hard disk controller and the MDS (Micro Diagnostic System) driver routines, the following message is displayed on the system console: wdss-bb-ee Where: bb = board number (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB022 Pg003 ee = MDS error code ERROR CODE ERROR DESCRIPTION 20........Timeout waiting for bus free status of 00. 21........Timeout waiting for the command phase after select. 22........Timeout waiting for status or message phase (stuck in cmd phase) 23........Skipped to the message phase over status phase. 24........Timeout waiting for the message phase. 25........The message byte received was not zero as expected. 5) wdcs Error When an unexpected status is received from the hard disk controller status register the following error message is displayed on the system console: wdcs-bb-xx-cc-ssssssss Where: bb = board number xx = ending status WDC Ending Status bit # 7 6 5 4 3 2 1 0 | | | | | | | | value 8 4 2 1 8 4 2 1 | | | | | | | | | | | | | | | *-Reserved (not used) | | | | | | *---Check (the sense bytes define the error) | | | | | *-----Equal (set when a search is equal) | | | | *-------Unit busy | | | *----------Reserved (not used) | | *------------Reserved (not used) | *--------------Reserved (not used) *----------------Reserved (not used) cc = command byte issued to the controller 00 = test unit ready command 08 = read commmand ssssssss = sense data (valid if the check bit is on in the ending status) Hard Disk Sense Bytes ______________________________________ BIT # | 7 : 6 : 5 : 4 : 3 : 2 : 1 : 0 | |------------------------------------| VALUE | 8 : 4 : 2 : 1 : 8 : 4 : 2 : 1 | | : : : : : : : | BYTE |____________________________________| 0 |ADVAL: Error Code | |------------------------------------| 1 | Reserved :(MSB) Logical blk add | |------------------------------------| 2 | Logical block address | |------------------------------------| 3 | Logical block address | |____________________________________| If the ADVAL bit is on the logical block address is valid. (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB022 Pg004 For purposes of decoding the error code if the high order bit (ADVAL) is on disregard it, ex. byte 0 = A5, use bits 6 - 0, look up code 25 - invalid logical unit number. Error Code Description 00 No sense 01 No index signal 02 No seek complete 03 Write fault 04 Drive not ready 06 No track zero 10 I.D. CRC error 11 Uncorrectable data error 12 I.D. address mark not found 13 Data address mark not found 14 Record not found 15 Seek error 18 Data check in no retry mode 19 ECC error during verify 1A Interleave error 1C Unformatted or bad format on drive 1D Self test failed 1E Defective track (media errors) 20 Invalid command 21 Illegal block address 23 Volume overflow 24 Bad argument 25 Invalid logical unit number ORIGINATOR: Norm Jones/Javier Van Oordt HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB022 Pg005 FIB 00023 02/16/90 *** Notes on margining the 3000/4000 power supply *** There is a three-terminal header at the top opening of the power supply. There are marks on the PCBA near the pins on the header indicating +, N and -. If a jumper is installed between + and N, the +5 volts will increase by 5% (5.25 volts nominal). If the jumper is installed between the - and N, the +5 volts will decrease by 5% (4.75 volts nominal). While margining the +5 volts supply can be an excellent trouble-shooting tool, be aware of the following: Some CMBs will fail in self-test, during the NVRAM test, if the +5 volt supply is margined low. If you encounter this condition, replacement of the CMB is not necessarily needed. If the NVRAM test fails, you will be unable to boot the system with the +5 volts margined. In this case, remove the margin jumper, boot the system and anytime after the NVRAM portion of the self-test is complete, install the margin jumper. It will not hurt the power supply to install or remove the margin jumper with the system powered on. Another option is to install the jumper between the - and N pins, then adjust the +5 volts to around +4.9 volts and after booting, set the voltage to 4.75 and run the diagnostics, SIT, etc. ORIGINATOR: Harvey Mitchell HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB023 Pg001 FIB 00024 01/17/92 *** Device to Major,Minor Number & Port Assignments *** MAJOR |MINOR | NUMBER |NUMBER | ASSIGNED UNIT -------|-------|---------------------------------- 0 | 1 | /dev/kmem - User Memory | | 2 | 0 | /dev/ts - 1/2 Inch Tape Drive | | 5 | 0 | raw printer on CMB Parallel Port 5 | 1 | raw printer on EIA Parallel Port 1 5 | 2 | raw printer on EIA Parallel Port 2 | | 6 | 0 | /dev/cs - 1/4 Inch Tape Drive | | 7 | 0-31 | /dev/fd0 - Floppy Disk Drive 0 7 | 32-63 | /dev/fd1 - Floppy Disk Drive 1 | | 8 | 0 | Printer on CMB Parallel Port 8 | 1 | Printer on CMB Serial Port A 8 | 2 | Printer on CMB Serial Port B 8 | 3-10 | Printer on Base Unit 4WAY or 8WAY with address 0 8 | 11-18 | Printer on Base Unit 4WAY or 8WAY with address 1 8 | 19-26 | Printer on Base Unit 4WAY or 8WAY with address 2 8 | 27-34 | Printer on Base Unit 4WAY or 8WAY with address 3 8 | 35-42 | Printer on Base Unit 8WAY with address 4 8 | 43-50 | Printer on Base Unit 8WAY with address 5 8 | 51-58 | Printer on Exp Unit 8WAY with address 6 8 | 59-66 | Printer on Exp Unit 8WAY with address 7 8 | 67-74 | Printer on Exp Unit 8WAY with address 8 8 | 75-82 | Printer on Exp Unit 8WAY with address 9 8 | 83-90 | Printer on Exp Unit 8WAY with addrres 10 8 | 91-98 | Printer on Exp Unit 8WAY with address 11 8 | 100 | Printer on EIA Parallel Port 1 8 | 101 | Printer on EIA Parallel Port 2 | | 9 | 0 | /dev/(optional) - CMB Port 0 9 | 1 | /dev/(optional) - CMB Port 1 | | 10 | 0-7 | Terminal on Base Unit 4WAY or 8WAY with address 0 10 | 8-15 | Terminal on Base Unit 4WAY or 8WAY with address 1 10 | 16-23 | Terminal on Base Unit 4WAY or 8WAY with address 2 10 | 24-31 | Terminal on Base Unit 4WAY or 8WAY with address 3 10 | 32-39 | Terminal on Base Unit 8WAY with address 4 10 | 40-47 | Terminal on Base Unit 8WAY with address 5 10 | 48-55 | Terminal on Exp Unit 8WAY with address 6 10 | 56-63 | Terminal on Exp Unit 8WAY with address 7 10 | 64-71 | Terminal on Exp Unit 8WAY with address 8 10 | 72-79 | Terminal on Exp Unit 8WAY with address 9 10 | 80-87 | Terminal on Exp Unit 8WAY with addrres 10 10 | 88-95 | Terminal on Exp Unit 8WAY with address 11 | | 11 | 0 | Local Area Network - LAN | | 14 | 0 | /dev/wd0 and /dev/rwd0 14 | 1 | /dev/boot - wd0 BOOT PARTITION 14 | 2 | /dev/swap - wd0 SWAP PARTITION 14 | 3 | /dev/root - wd0 ROOT PARTITION (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB024 Pg001 MAJOR |MINOR | NUMBER |NUMBER | ASSIGNED UNIT -------|-------|---------------------------------- 14 | 32 | /dev/wd1 and /dev/rwd1 14 | 33 | /dev/drive1 and /dev/rdrive1 | | 14 | 64 | /dev/wd2 and /dev/rwd2 14 | 65 | /dev/drive2 and /dev/rdrive2 | | 14 | 96 | /dev/wd3 and /dev/rwd3 14 | 97 | /dev/drive3 and /dev/rdrive3 | | 14 | 128 | /dev/wd4 and /dev/rwd4 14 | 129 | /dev/drive4 and /dev/rdrive4 | | 14 | 160 | /dev/wd5 and /dev/rwd5 14 | 161 | /drive5 and /dev/rdrive5 | | 14 | 192 | /dev/wd6 and /dev/rwd6 14 | 193 | /dev/drive6 and /dev/rdrive6 | | 14 | 224 | /dev/wd7 and /dev/rwd7 14 | 225 | /dev/drive7 and /dev/rdrive7 -------|-------|---------------------------------- IDENTIFYING THE PORT ASSOCIATED WITH A MAJOR:MINOR NUMBER The /etc/ports file associates the port number used by the system to the device name, device type file, and major/minor number for all configured ports. The file is in the following format: :/etc/(printers or terminals):/dev/:: Sample /etc/ports: 0:/etc/terminals:/dev/tty0:9/0::Port A (console) 1:/etc/printers:/dev/p1:8/2:AA2000 8:/etc/terminals:/dev/tty7:10/5:EE2001 Command examples: To view type: >p /etc/ports To print type: >cat /etc/ports > /dev/lp ORIGINATOR: H. Mitchell MODIFIED by: Norm Jones HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB024 Pg002 FIB 00025 04/10/89 *** Hard Disk Bad Block Retirement *** Bad Block Retirement should be done with the assistance from MBF systems AND/OR the software vendor ! ! ! Effective with BOSS/IX release 7.3A, is the capability of retiring bad blocks on the Winchester disk drives - without the necessity of reformatting the disk. Since retirement of a bad block reduces the disk capacity by only one block, this is MUCH PREFERRED over backing up, reformatting and restoring because of the time savings. ************** CAUTION !! READ AND UNDERSTAND BEFORE PROCEEDING !! ************* *. * * DO NOT Retire Bad Blocks if Customer uses IMAGE BACKUPS ! ! ! ! * *. * * You CANNOT Retire Bad Blocks in the Boot or Swap partition, a problem in * * this area will require reformatting. * *. * * Currently you can only retire bad blocks on the first 4 drives (0 thru 3) * *. * * If a 'Retired' Bad Block is in a directory, you must run 'fschk -fix' to * * repair the NOW CORRUPT file system. Any files in the directory that used * * that block will become 'LOST FILES' (see user ref. manual m6211c pages * * 6-18,19 and 20 for recovery suggestions.) * * --------------------------------------------------------------------- * * -- THE FOLLOWING WILL OCCUR IF THE BAD BLOCK IS CONTAINED WITHIN A FILE: -- * *. * * The utility "frepair" will create a duplicate file, delete the original * * file and retire the bad block from future use. You will need sufficent free * * space (temporarily) for the duplication of the file. *. * * If the file is a DIRECT or MULTI-KEYED file, and the bad block is * * contained in the the KEY portion of the file, frepair will attempt to repair * * the file. The majority of the time the repair will be successful. * *. * * If the file is a DIRECT or MULTI-KEYED file, frepair will ask for a * * "template" file. The template file must be a file defined with exactly * * the same parameters as the file containing the bad block. It need not * * contain any data as frepair will use it only to determine key size, data * * size, etc. as reference for file repair. If a template file name is not * * specified, frepair will rebuild the keyset utilizing the data portion. This * * can take a long time depending on the file size. * *. * * If the file is a STRING, INDEXED, SERIAL or SORT type or it is a DIRECT * * or MULTI-KEYED type and the bad block is contained in the data portion of * * the file, frepair will copy the original file to the duplicate file and * * store zeros in the block of the duplicate file corresponding to the bad * * block of the original file. In this case, the zeroed out block MUST BE * * MANUALLY REPAIRED, probably by the system programmer. * *. * * Because of the facts stated above, it is HIGHLY RECOMMENDED that an attempt * * be made to copy the file PRIOR to retiring the bad block. After a copy or * * backup of the file is secured, you can delete the file containing the bad * * block and the retire the block with no consequence. Since most cases of a * * bad block involve intermittant failures, a backup or copy is easy to obtain. * *. * ********************* ( Continued on Next Page ) **************************** HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB025 Pg001 When a disk error occurs which is deemed by the O.S. to be an "unstable" disk surface problem, the failing block number is stored in the error log fascility. To display any bad blocks stored by the error logger, enter the following command (you must be logged on under "admin"): ADMIN> /sys/errlog /etc/error.log -disk -text (Record on paper any bad block(s) shown) After determining the bad block number(s), the block(s) can be retired (from usage by the O.S.) by invoking the "frepair" command, as follows: ADMIN> frepair this invokes the file repair/block retirement process type 1 twice this sets frepair to RETIRE BAD BLOCK type to indicate selections are correct type yes to proceed type disk no. problem disk drive number i.e, 0 Note: You cannot retire a bad block on a 'mounted' file sys. Drives 1,2 and 3 will have to be 'unmounted' first. type block no. bad block number (from the error log) The system will display "WORKING" while determining if the bad block is within a file. If the block is unused, it will display: Block xxxxxx is unused - Enter to map out bad block. If used it will report the directory and the file name that the bad block affects, It will display: Bad Block is in file /xxxx/xxxxxx in partition xx on disk xx Enter (note: copy file name for reference) System will then display: Enter 1- Repair the file 2- Delete the file {Caution! you should have a current backup of the file if you wish to proceed past this point, because the block you have selected will be set to nulls!!} If you select repair, the system will ask: Enter name of template file or if none. (Note: On large files, a Template will reduce repair time but is not needed) After a time the system will display: File Name : /xxxx/xxxxxx File Type : xxxxxxxxxx Repair Mode : Retire Bad Block and repair xx errors detected xx blocks retired xxxxxx | xxxxxx |------ will list bad blocks retired xxxxxx | ( continued on next page ) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB025 Pg002 **** NOTE 1 **** In BOSS/IX releases prior to 7.5A, there is a bug in frepair which causes it to display a block number different from that entered to be retired.(Only on drive 0). This is a display bug only, and 'frepair will retire the correct block. (FYI, The difference between the two figures 8192, which is the size of the boot and swap partition. 'frepair' references the start of the file system, (only when displaying), and the error log references the start of the disk. However, When entering the block to be retired, 'frepair' wants a block # starting from the begining of the disk,(which is what the error log gives you). To determine which blocks are retired on the system do the following: Type dispbb /dev/wdX where X = the disk number The system will display the bad block(s) on file. (Keep a hard copy of the bad blocks in the system log; if at a later date you wish to reformat the drive, you should add these blocks to the list. If you do not, these soft areas may show up again later !) **** NOTE 2 **** Retired bad blocks are no longer usable by the O.S. because they are flagged WITHIN O.S. as bad. Bad block retirement DOES NOT add the defective block to the defect map on the disk. Therefore whenever DUTIL is used to scan the disk, errors may occur when the bad block is read. Below is an method to scan the disk for errors online: ADMIN> debe if=dev/rwdX of=/dev/null bs=64b -pg -noerror (X = drive no.) Any failures will be logged in the error log. The debe command will read previously retired bad blocks and if an error occurs the error will be logged to the error log. You WILL have to compare any errors found to the blocks listed in the bad block table to 'discover' any 'new' errors. **** NOTE 3 **** If, for any reason, the O.S. must be re-installed, all retired bad blocks information will be lost. However, if the O.S. is only upgraded, the bad block information will not be lost. For this reason, a listing of all bad blocks should be kept in the system log. **** NOTE 4 **** If 'frepair' is used to check a file after it has been repaired, 'frepair' will show an error on multi-key files with COMPOSITE FIELDS, even after the repair...--- Please note that the file is OK, it's just that 'frepair' does not recognize composite fields as being correct. ORIGINATOR: H. Mitchell / D. Loiselle HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB025 Pg003 FIB 00026 08/02/91 *** Crash Recovery and Dump Analysis Procedures *** There are four steps which should be taken every time a system crash occures: (1) SAVE DUMP A dump of system memory should be taken every time the system crashes. To ensure a dump, make sure the "dump" option is selected in the 'vconf' configuration file. When a crash occurs, the system will ask which external media the dump should be saved to. Dumps may be saved to MCS, MTS or Floppy Diskettes (2000 system only). Once the dump has been saved, the system can be rebooted. (2) RE-BOOT and REPAIR/ANALYZE DAMAGE On the first boot after a system crash, an automatic filesystem repair will need to be done. A prompt to begin the filesystem repair process will appear during system boot (assuming 'noride' has been selected in the 'vconf' file); type to begin the repair process. Once the filesystem has been repaired, the system needs to be rebooted. NOTE: After the system has rebooted, files that were opened and modified during the time of the crash should be identified. These files lack integrity and need to be repaired. (3) RETRIEVE ERROR LOG FILE on DUMP NOTE: This command (dmplog) was implemented on OS level 7.3A. To retrieve the dump tape's error log file for examination: Type: ADMIN> dmplog /dev/cs > /dev/ to retrieve the error log file on MCS tape. If the dump is on MTS tape, replace "/dev/cs" with "/dev/ts". NOTE: See example # 1, a sample of a system error log printout. (4) Once the error log file has been retrieved, its contents can be examined to determine the cause of the crash. The last error in the file will be the cause of the crash, unless the dump was forced. A copy of the BOSS/IX Diagnostics and Error Log Manual (M6204) may be useful in deciphering information in the error log fields. The last eight errors recorded are contained in the errorlog report. The errors are listed in order from the most recent error (usually the error that caused the dump) to the oldest. On some dumps, errors that occur prior to the most recent error indicate a pattern that may be developing (ie. parity errors on a specific port, etc). HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB026 Pg001 When a dump is caused by Excessive Interrupts, other entries in the log may indicate the port(s) causing the problem. The other entries will have a major number of "10" which indicates a 4/way or 8/way port error. Each entry shows the date and time that the error was logged. When the first entry indicates Excessive Interrupts, look at the date and time that it was recorded and inspect other entries for similiar times. If the entries were logged near the same time, it may indicate the problem port(s). For more information on determining the exact terminal, see note 3. The following are known dumps and solutions. Entries where the SYSTEM field is shown as "3000*", indcates the MAI 2500/3000/4000 systems. O/S SYSTEM MAJ:MIN T FIX CAUSE ====================================================================== Any 2000 0:7 See Note 1 Parity Error on Memory Any 3000* 0:23 See Note 2 Parity Error on Memory Any 3000* 0:24 See Note 2 Parity Error on Memory Any 2000 0:18 Diagnostics/CMB Hardware/Illegal Instr. Any 3000* 0:18 Diagnostics/CMB Hardware/Illegal Instr. Any Any 0:0 Diagnostics/CMB Trap/Bus Error Any Any 0:1 Diagnostics/CMB Trap/Illegal Instruct. Any Any 0:4 Not Applicable Trap/Power Failure Any Any 0:5 Diagnostics/CMB Trap/System Call Any Any 0:10 Upgrade OS/CMB Trap/Memory Management Any Any 0:11 Diagnostics/CMB Trap/Odd Address Any Any 0:30 Expansion Unit Trap/Power Failure Any Any 2:0 Not Applicable 1/2 Inch Streamer Any Any 6:0 N/A Applicable Cartridge Tape Any Any 8:-- See Note 3 Printer Filter Any Any 9:-- See Note 4 CMB Any Any 10:-- See Note 5 Parity /DMA Arbitration Any Any 14:-- See Note 6 Disk/Controller Problem 7.2 3000* 0:0 20CFC Change to DTR Flow Control Problem 7.2 3000* 0:16 1C4E8 Upgrade 7.5B*22 Wrting Integer to Short 7.3A 2000 0:16 1CCDC Upgrade 7.5B*22 UPD Free 7.3A 3000* 0:16 1E81A Upgrade 7.5B*22 UPD Free 7.3A 2000* 0:11 35D28 Upgrade 7.5B*22 UPD Free 7.4A 2000 N/A 1D2AA Upgrade 7.5B*22 Stty Write 7.4A 2000 N/A 1E436 Upgrade 7.5B*22 Stty Write 7.4A 3000* N/A 205DA Upgrade 7.5B*22 Stty Write 7.4A 3000* N/A 20642 Upgrade 7.5B*22 Stty Write 7.4A 3000* 0:16 1A894 Upgrade 7.5B*22 Memory Compaction 7.4A 3000* 0:16 1E234 Upgrade 7.5B*22 System Stack Overflow 7.4A 2000 0:16 1BF92 Upgrade 7.5B*22 System Stack Overflow 7.4B 3000* 0:16 1E2B8 Upgrade 7.5B*22 Excessive Interrupts 7.4B 2000 0:16 1C020 Upgrade 7.5B*22 Excessive Interrupts 7.4B 3000* 0:16 7B7A Upgrade 7.5B*22 Excessive Interrupts 7.4B 3000* 0:16 182B8 Upgrade 7.5B*22 Excessive Interrupts 7.5B 3000* 0:16 1E5DO Upgrade 7.5B*22 Excessive Interrupts 7.5B 3000* 0:16 1E608 Upgrade 7.5B*22 Excessive Interrupts HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB026 Pg002 Note 1: To determine which memory board had the parity error, use the following chart to find the address range (in hex) in which the 'T' value is located. BOARD refers to 256 KB increments of memory (i.e. 1st addressable board, 2nd addressable, ...). BOARD ADDRESS RANGE BOARD ADDRESS RANGE ============================================================ 1 00000 -> 3FFFF 4 C0000 -> FFFFF 2 40000 -> 7FFFF 5 100000 -> 13FFFF 3 30000 -> BFFFF 6 140000 -> 17FFFF Note 2: To determine which memory board had the parity error, use the following chart to find the address range (in hex) in which the 'T' value is located. MB refers to the MegaByte of memory in which the parity error occurred. If the system uses 4 MB boards and the error occurred in the third (3rd) MB of memory, the board to replace would be the first (1st) 4 MB board. BOARD ADDRESS RANGE BOARD ADDRESS RANGE =============================================================== 1 00000 -> FFFFF 9 800000 -> 8FFFFF 2 100000 -> 1FFFFF 10 900000 -> 9FFFFF 3 200000 -> 2FFFFF 11 A00000 -> AFFFFF 4 300000 -> 3FFFFF 12 B00000 -> BFFFFF 5 400000 -> 4FFFFF 13 C00000 -> CFFFFF 6 500000 -> 5FFFFF 14 D00000 -> DFFFFF 7 600000 -> 6FFFFF 15 E00000 -> EFFFFF 8 700000 -> 7FFFFF 16 F00000 -> FFFFFF Note 3: The minor number in the error log entry denotes the printer or printer port that is generating errors. To determine the port, use the chart below. MINOR DESCRIPTION MINOR DESCRIPTION ============================================================== 0 CMB Parallel port 3 1st port on 1st 8/way 1 CMB Serial port A 4 2nd port on 1st 8/way 2 CMB Serial port B 5 and so on . Refer to SPx Handbook FIB 24 for range of minor numbers by PCBA. Note 4: The minor number in the error log entry denotes the port on the CMB that is generating the error. To determine the port, use the chart below. MINOR DESCRIPTION MINOR DESCRIPTION =============================================================== 0 CMB Serial port A 2 CMB Parallel port 1 CMB Serial port B Note 5: When entries have a major number of "10", it usually denotes parity, cabling, excessive interrupts and/or framing errors. To determine the actual port/device that is generating the error(s), look at the minor number. The first port on the first 4/way or 8/way is "0", the second is "1" and so on. If the controller is an 8/way, the value of 'A' is 8 in the error log entry. If the value of 'A' is 4, the controller is a 4/way. The table below gives a few examples. HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB026 Pg003 MINOR DESCRIPTION MINOR DESCRIPTION =============================================================== 0 Port 1 on 1st 8/way 8 Port 1 on 2nd 8/way 1 Port 2 on 1st 8/way . . 15 Port 8 on 2nd 8/way 7 Port 8 on 1st 8/way 30 Port 7 on 4th 8/way Refer to SPx Handbook FIB 24 for range of minor numbers by PCBA. Note 6: When entries have a major number of "14", it denotes disk or controller problems. The 'A' value is the block number on the the disk where the error was received. To find the disk drive or partition that contains the bad block, use the chart below. MINOR DRIVE MINOR DRIVE ============================================================== 0 - 31 0 128 - 159 4 32 - 63 1 160 - 191 5 64 - 95 2 192 - 223 6 96 - 127 3 224 - 255 7 *** Before sending in a system dump tape, process the dump tape using this field bulletin. If you are unable to determine why the system dumped by reviewing the known system dumps, contact a National Support Specialist. A Dump Report Form (which can be found in CPU-BASIC FOUR-2500,3000,4000 FIB 61) is ABSOLUTELY REQUIRED to have Tustin analyze a dump, also a system alert must be generated before sending in a dump tape. *** IF THE DUMP REPORT FORM IS NOT COMPLETELY FILLED OUT WITH ALL NECESSARY INFORMATION, THE DUMP TAPE WILL NOT BE PROCESSED AND THE TAPE RETURNED TO THE CUSTOMER. (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB026 Pg004 Example # 1 SYSTEM CRASH: Version: 'EOS4A24, BOSS/IX release 7.4A*24 (Jan 12 1988) Time: Tue Jun 14 1988 14:27:51 System serial number 300082886 SPR#: Comment: Current Operating System Configuration: System name: 'MAI 3000' Startup in secure, multi-user mode (system dump enabled) Root device: 1403 Swap device: '/dev/swap', Size=4096 blocks (2048KB) Override mount error on system boot: disabled Start program: '/etc/start' Erase character: 'backspace' (0X08), Linekill character: 'delete' (0x7F) Instruction cache: disabled Data cache: disabled Number of: directory cache entries: 64 disk buffers: 1468 (734KB) well known sockets: 0 dynamic LAN sockets: 0 LAN buffers: Maximum number of: mounted filesystems (including root): 3 processes: 87 shared text segments: 68 file descriptors: 356 file locks: 180 logical units: 510 printers: 7 eventcalls: 50 ghosts: 0 Size of: tty input buffers: 512 tty type-ahead buffers: 256 ram disk: 0 ----------------------------------------------------------------------------- System Error Log log# errno type dec_entries hex_entries class text errbt major:minor A B C D E T U V rev time 1 985 0 0:16 -32767 92 0 0 0 0 28E38 0 0 Jun 14 14:27:51 2 984 0 10:1 4 8 0 0 0 0 80 7002E 0 Jun 14 14:27:50 3 983 0 8:11 -62 0 0 0 1 0 21 22 0 Jun 14 14:23:02 4 982 0 10:2 1 8 0 0 0 0 10 0 0 Jun 14 13:46:31 5 981 0 8:22 -62 0 0 0 1 0 21 22 0 Jun 14 13:31:55 6 980 0 10:19 1 8 0 0 0 0 10 0 0 Jun 14 12:56:45 7 979 0 10:19 1 8 0 0 0 0 10 0 0 Jun 14 12:56:44 8 978 0 10:7 1 8 0 0 0 0 10 0 0 Jun 14 11:23:07 ------------------------------------------------------------------------------- ORIGINATOR: N. Jones Chuck Zimmerman HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB026 Pg005 FIB 00027 01/12/90 *** Adding Winchester disk drive(s) to the SPx system *** o Basic conventions and restrictions: 1. The 2000 system can have a maximum of two drives and one WDC (Winchester Disk Controller) controller. 2. The 2500 system is limited to one drive and one WDC or DSC (Dual SCSI Controller) controller. 3. The 3000 system can have up to three drives and two WDC or DSC controllers 4. The 4000 system can have a up to seven drives, up to three in the CCA Base unit and up to four in the Expansion unit. The Base and Expansion units are limited to two WDCs or DSCs in each cabinet. 5. On the 3000 and 4000 systems only three drives may be installed in the Base unit. The reason for this is due to cooling and power supply restrictions. (The MCS uses the fourth power supply outlet). 6. On the 4000 system, drives WD0, WD1, WD2, and WD3 must reside in the Base unit and drives WD4, WD5, WD6 and WD7 must reside in the Expansion Unit. If there are both a WDC and a DSC controller contained in the Base unit and there is only one drive on the 1st disk controller, the second controller may have two drives attached. In this configuration, the drives must be addressed WD0, WD2 and WD3. 7. On the 4000 system, no cables can be connected between the BASE unit and the EXPANSION unit except the EIA/EIB Interconnect cables. Therefore the disk controller and power supply for a given disk drive must reside within the same CCA cabinet as the disk they supply. o Physical drive installation information: Both disk drives must reside within the CCA of the 2000 system, drive 0 should be the one nearest the controller stack. If the drive is to be attached to a WDC controller, the drive address must be set as follows: If the drive is to be WD0, WD2, WD4 or WD6 set drive address to 0. If the drive is to be WD1, WD3, WD5 or WD7 set the drive address to 1. If the drive is to be attached to a DSC controller, the drive address must always be set to 5. On the DSC controller, all disk drives must have terminator resistor chips installed. (See SPx Handbook FIB 5 for SCSI drive addressing and terminator information) On the WDC controller, one drive only will have terminator resistor chips installed. If there are two drives on the WDC controller, the terminators must be installed in the drive with the highest WD designation. (See SPx Handbook. FIB 5 for drive terminator information) Use Figure 1 on the next page to locate the physical position of the drives in the 3000 and 4000 systems: (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB027 Pg001 ------------------------- ------------------------- | BASE UNIT | | EXPANSION UNIT | | ________ | | | | | Tape | | | | | -------- | | | | _________ _________ | | _________ _________ | | | | | | | | | | | | | | | Drive | | Drive | | | | Drive | | Drive | | | | 3 | | 2 | | | | 7 | | 6 | | | | | | | | | | | | | | | --------- --------- | | --------- --------- | | _________ _________ | | _________ _________ | | | | | | | | | | | | | | | Drive | | Drive | | | | Drive | | Drive | | | | 1 | | 0 | | | | 5 | | 4 | | | | | | | | | | | | | | | --------- --------- | | --------- --------- | ------------------------- ------------------------- Figure 1 o WDC and DSC controller information: The DSC controller may have only embedded SCSI drives connected. Drives WD0 and WD1 must be connected to a WDC or DSC controller which is addressed as 0 and installed in the BASE UNIT. Drive WD2 and WD3 must be connected to a WDC or DSC controller which is addressed as 1 and installed in the BASE UNIT. Drives WD4 and/or WD5 must be connected to a WDC or DSC controller which is addressed as 0 and installed in the EXPANSION UNIT. Drives WD6 and/or WD7 must be connected to a WDC or DSC controller which is addressed as 1 and installed in the EXPANSION UNIT. See SPx Handbook FIB 18 for WDC controller addressing and jumpering info. See SPx Handbook FIB 28 for DSC Controller Addressing and jumpering info. o After installation the new drive should be formatted. See BOSS/IX Error Log and Diagnostics manual (M6204) for complete formatting information. o BOSS/IX changes required after drive installation and formatting: Boot the system and log on under admin. o In order to access the new drive, two buffered (B) and two unbuffered (U) device files must be created in the /dev node: The buffered device file name for the whole disk is /dev/wd# The unbuffered device file name for the whole disk is /dev/rwd# The buffered device file name for the disk file system is /dev/drive# The unbuffered device file name for the disk file system is /dev/rdrive# ( #=Drive Number in the above device file names ) (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB027 Pg002 Use Table 1, below, to find System Minor Numbers for the wd# and drive# you are adding to the system: ___________________ | Disk | Minor | | Drive | Number | |--------|--------| | wd1 | 32 | | drive1 | 33 | |--------|--------| | wd2 | 64 | | drive2 | 65 | |--------|--------| | wd3 | 96 | | drive3 | 97 | |--------|--------| | wd4 | 128 | | drive4 | 129 | |--------|--------| | wd5 | 160 | | drive5 | 161 | |--------|--------| | wd6 | 192 | | drive6 | 193 | |--------|--------| | wd7 | 224 | | drive7 | 225 | ------------------- Table 1 o To create the four required files, at the ADMIN> prompt enter: makedev /dev/wd# B 14 YY [ #=new wd number and YY=Minor number ] then enter: makedev /dev/rwd# U 14 YY [ #=new wd no. and YY=Minor number ] then enter: makedev /dev/drive# B 14 YY [ #=new drive no. and YY=Minor number ] then enter: makedev /dev/rdrive# U 14 YY [ #=new drive no. and YY=Minor number ] o The following will initialize the superblock on the new drive. enter: usb /dev/rwd# -get desc=drive# [ #=new drive number ] now enter: ved /etc/diskdesc/drive# Press n to move the cursor to the blank line below the line in the file like that shown below: **partition # : starting block : length in blocks (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB027 Pg003 now type: 1:0:-1 now type c to exit the ved and select the save edits option. At the ADMIN> prompt enter: usb /dev/rwd# -save -set type=drive# [ #= new drive number] o Once the device has been defined, a filesystem is created using the makefs command. Enter: makefs /dev/rdrive# [ #= new drive number ] o Verify that the drive has been correctly defined and partitioned by using the "space" command. The number of blocks returned should reflect the size of the hard disk installed. Enter: space /dev/rdrive# [ #=new drive number ] o To access the newly installed hard disk, the device must be mounted to a directory on drive 0. It is recommended that the /mnt directory not be used for this purpose. Create a new directory to be used as a mount point for the new drive. Enter: makedir /mntwd# [ #=new drive number ] o So that the customer will have access to the new drive each time they load their system, ved the /etc/sinit file and add the following line at the end of the file: mount /dev/drive# /mntwd# [ #=new drive number] Save the edited file. Each time the system is loaded the file is executed and drive# will be mounted to /mntwd#. Again #=new drive number. o When the customer wishes to use the new drive they need to prefix /mntwd# [ #=new drive number ] to their path name. ORIGINATOR: H. Mitchell HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB027 Pg004 FIB 00028 03/16/90 *** Dual SCSI Disk Controller PCBA *** The following information applies to this PCBA: MM535080 - (DSC) Dual SCSI Controller - 903667-001 -----------------SWITCH SETTING LEGEND----------------- : : : C = Closed or On O = Open or Off X = Not Used : ------------------------------------------------------- o The DSC may be used to connect only disk drives with the embedded SCSI controller. Like the WDC controller, there are two channels. Each channel can have one disk drive connected. o The DSC may be used within the same cabinet with a WDC controller. The addressing scheme is the same on both. There can still be only two total disk controller PCBAs within a cabinet. o Switch Settings There is one 12 position dip switch (S1) on the DSC. The functions selectable via this switch are as follows: __________________________________________________ | Switch S1 | | | Postion(s) | Function | |------------|-----------------------------------| | 1,2,3 & 4 | SCSI Channel A Arbitration Level | |------------|-----------------------------------| | 5,6,7 & 8 | SCSI Channel B Arbitration level | |------------|-----------------------------------| | 9 & 10 | Controller Address | [ CC0000 or CD0000 ] |------------|-----------------------------------| | 11 | Primary Interrupt Level | [ 2 ] |------------|-----------------------------------| | 12 | Not used | |____________|___________________________________| Set the S1 switch using the following chart: _____________________________________________________________________________ | DSC # | S1- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | |---------------|----|----|----|----|----|----|----|----|----|----|----|----| | 0 | O | O | C | C | C | O | C | O | O | O | O | O | |---------------|----|----|----|----|----|----|----|----|----|----|----|----| | 1 | O | O | C | O | C | C | O | O | O | C | O | O | |_______________|____|____|____|____|____|____|____|____|____|____|____|____| o There are five jumpers on this board, JP1 through JP5. For proper operation, all five jumpers must be INSTALLED. o LEDs CR1 and CR2 illuminate to indicate successful self test. o Cable connection J3 is SCSI Bus A to 1st disk drive J4 is SCSI Bus B to 2nd disk drive (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB028 Pg001 The SCSI Bus is 50 pins wide and is the only connection between the DSC and the SCSI disk drive. o Drive addressing and termination All SCSI drives are to be set to address 5. (See SPx Handbook FIB #5) Since the SCSI drives each have a unique cable connecting them to the DSC and they are not "daisyed" together as they are on the WDC, each drive must have the terminators chips installed. o Cable information SCSI Bus cable - P/N 916237-002 Diagnostic loopback cable - P/N 907174-035 MM080678 The Diagnostic loopback cable, when used is connected between SCSI Bus A connector J3 and SCSI Bus B connector J4. * Note * The SCSI Bus cable may be disconnected from the disk drive and that end plugged into the other SCSI Bus connector. This will work in place of the Diagnostic loopback cable. ORIGINATOR: H. Mitchell HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB028 Pg002 FIB 00029 09/17/90 *** 1/2" Tape Controller - Single Board *** The following information applies this PCBA: MM535063 - 1/2" Tape Controller - 903639-001 o This controller does not require the NCR Formatter box which was used with the SCSI adapter. The new controller supports both the 1/2" Cipher streamer and the 1/2" STC GCR tape drives. o When installed in a 4000 system this controller must reside in the BASE cabinet. o Jumper settings are as follows: JMP1 no jumper JMP6 2-3 JMP2 1-2 JMP7 1-2 JMP3 1-2 JMP8 1-2 There is no JMP4 JMP9 1-2 JMP5 1-2 JMP10 1-2 o Tape drive switch setting information: MTC: Switch at location 5W positions 1,2,4,5 on(closed) all others off GCR: IF pcba positions 1,2,4,5 on(closed) all others off CP pcba Switch SW1 positions 1,2,3,4,9,10 on(closed) 5,6,7,8 off CP pcba Switch SW2 positions 3,4,5,7,8,9,10 on(closed) 1,2,,6 off others off. o Cabling information: ______________________________________ | Drive | Controller | Tape Drive | | Type | Connector | Connector | |-------|--------------|-------------| | MTC | J3 | P1 | | | J4 | P2 | |-------|--------------|-------------| | GCR | J3 | J7A | | | J4 | J6A | |_______|______________|_____________| P/N 907065-005 - 25 ft cable P/N 907065-006 - 10 ft cable ORIGINATOR: H. Mitchell HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB029 Pg001 FIB 00030 05/09/94 *** Helpful Hints *** This bulletin contains miscellaneous how-to information, primarily dealing with OS commands, but not limited to just that. If you have suggestions for additions to this FIB, please submit them to Harvey Mitchell via OMS. . .............................................................................. * Freeing a "locked" terminal Whenever a terminal locks or freezes on an SPx system, try the ttymodes command to free it up before re-booting the system. The command below causes the OS to set all the modes of the selected terminal to the initial state. ttymodes -default < /dev/ttyx The x in ttyx should be the tty number assigned to the locked terminal. . .............................................................................. Another method of freeing the locked terminal that has been used with some success is as follows: Determine the "T" number of the port - ie. T12 Log onto another terminal and go to basic mode - ie. ADMIN> basic Type OPEN(1)"TX" The x in "TX" = the actual terminal number. Type READRECORD(1)A$ Type END This method has been especially helpful on terminals attached via modem and/or multiplexer. . .............................................................................. * Setting up a system so it will perform a dump See the CMB FIBs in this handbook for the jumper settings required to cause the system to dump when the RESET/NMI switch is depressed. Also, be aware that "dump enable" must be set in the /boot/etc/conf file. . .............................................................................. * Creating a file which monitors and logs when the system is shutdown and when it is booted. (Contributed by Joe Malski, DSS) This can be helpful if a customer is experiencing file corruption and you suspect the system is not being shutdown properly due to an operator or power problems. To create the log, get to the ADMIN> prompt and: type: echo "" > /etc/boot.log then: shutdown abd re-boot the system To display the log anytime there after, get to the ADMIN> prompt and: type: p /etc/boot.log The system will display the log, showing dates and times. HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB030 Pg001 . .............................................................................. * A macro which will locate a given file in the root file system is as follows: #b locate do 'ls -a -l -r -p | match $1' To use this macro, do the following: 1. Enter the above macro into the system 2. Type cd / 3. Enter locate The system will search the entire system (including any mounted disks) and print the FULL path name of that file(s). . .............................................................................. * The following is designed to easily find the system error description that may correspond to a BASIC error or a system error. Example: Try to open a printer in BASIC and an error 255 occurs, enter "error" and the following is displayed: Basic error code: 255 System error code: -254 Now to find the description of the error -254 the file /etc/comerr.list must be displayed by entering the following: match \-254 /etc/comerr.list |p This command will display all the lines which have "-254" in them. The back-slash (\) character prevents the - sign in front of the error number from being interpreted as an argument. You can also create a macro called 'mtch': Enter: #b mtch do 'match \$1 /etc/comerr.list |p' - or - To make an executible file, ved /bin/mtch and enter: match \$1 /etc/comerr.list |p Now, exit ved and use filemodes to make the file executible: Enter: filemodes +x +x /bin/mtch To use the newly created macro or executable file, enter: mtch . .............................................................................. * On-line disk scan (contributed by Mike Burch) To read a disk while the customer is on-line type: * Any errors will be logged to the error log. ADMIN> debe if=/dev/root of=/dev/null bs=64 -pg if Input file of Output file HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB030 Pg002 bs Number of 512K sectors to get each read. -pg Option to prints dots .... Input file could be: /dev/rdrive1 /dev/rdrive2. ****WARNING:**** Be cautious using debe. Verify the output file is correct. If you typed, of=/dev/kull, the system would try to copy itself to a file " /dev/kull". This would cause the disk to fill and crash the system. . .............................................................................. * Convert Logical Sector Number to Physical Location To convert the logical block number to physical location (Cylinder, Head and Sector) two steps have to be taken: STEP 1. Convert the decimal number obtained from the errorlog to hexidecimal. 1. Boot the diagnostics and load 'dutil' 2. Obtain block number from the errorlog (Decimal number). 3. To convert block number from decimal to hex: Enter after the prompt: dec xxxxxx :where xxxxxx equals the decimal block number The program will print the block number in hex. write it down. STEP 2. Translate the logical block number (in HEX) to physical location: 1. Enter after the prompt: unit X : to select desired disk unit (X= 0-7) block yyyyyy :where yyyyyy equals the HEX block number 2. The program will read the block and print its contents on the screen. The first two lines contain the logical sector number and the physical location in HEX, as follows: SECT: xxxxxx EXP CYL: xx HEAD: xx B.F.I. : xxxx Write this information down to be used during drive format. If there is an error in the ID field of the targeted sector, the program will not print the physical location. It will be necessary to read the blocks before and after the targeted sector to determine the location of the bad sector. Use the following chart to obtain B.F.I.: BFI (BFI) SEC BFI = single-board WD controller HEX (HEX) HEX (BFI) = 2-board WD controller 0096 (0096) 00 02D6 (02D8) 01 0516 (051A) 02 0756 (075C) 03 0996 (099E) 04 0BD6 (0BE0) 05 0E16 (0E22) 06 1056 (1064) 07 HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB030 Pg003 1296 (12A6) 08 14D6 (14E8) 09 1716 (172A) 0A 1956 (196C) 0B 1B96 (1BAE) 0C 1DD6 (1DF0) 0D 2016 (2032) 0E 2256 (2274) 0F 2496 (24B6) 10 ORIGINATOR: H. Mitchell MODIFIED by: Norm Jones HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB030 Pg004 FIB 00031 MAI COMPANY CONFIDENTIAL 02/18/92 *** SPx Data File Integrity Check & File Repair Procedure *** The following step-by-step procedure should be performed anytime a non-standard system shutdown has occurred (power failure, system crash, system hang, etc.) and is based on the system having successfully completed an automatic filesystem repair. This procedure is recommended for use by the system administrator. A qualified system's administrator should have completed MAI's "Managing Your BOSS/IX System", taught at MAI education centers. PLEASE MAKE SURE YOU HAVE A CURRENT COMPLETE BACKUP. If you do not have a current backup, please do one before executing this procedure. You may need to restore your backup if the file repair utility is not able to fix all the data files. NOTE: Use a different set of tapes from your previous backup! 1. Login as "admin" on the system console (T0). 2. Have all users completely log off the system prior to repairing the data files. You can check which users are still logged on the system by using the "who" command at the system prompt. ADMIN> who 3. After all users have logged off the system, shut the system down into single-user mode by entering the following command: ADMIN> shutdown 0 -s It takes several minutes for the system to completely shutdown into single-user mode. Enter the following command: ADMIN> fichk / -f After the command is entered, a filelist of the files that need to be repaired is created. It may take several minutes for the list to be created. After the list is created, the following message is displayed: ______________________________________________________________________ |Filelist '/util/fl/fichk.f' contains filenames that lack integrity. | |______________________________________________________________________| 4. Run the file repair utility by entering: ADMIN> frepair 5. The utility begins with a multiple choice menu containing the parameters for the utility to use during the analyze and repair process. To change the selection of an item, type the item number (without RETURN). An asterisk next to an item, indicates that the item has been selected. Please select the following items: (continued on next page) HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB031 Pg001 ______________________________________________________________________ | File Analysis and Repair Utility | |______________________________________________________________________| | 1. REPAIR MODE 2. FILE SELECTION | | analyze single file | | *analyze & repair *selected files | | retire bad block | | | | 3. REPORT DEVICE | | terminal | | printer | | *file | | | | Type number to change selection, when all are correct: | |______________________________________________________________________| 6. A warning message is displayed on the screen: ______________________________________________________________________ | WARNING: Files or data might be lost, Proceed? (yes/no): yes | |______________________________________________________________________| Type in 'yes' 7. The screen is cleared and you are prompted to enter the report file name. Enter the name in the following format: repair.MMDDYY where: MM=month DD=day YY=year ______________________________________________________________________ | Working directory: / | | | | Type report file name: repair.MMDDYY | |______________________________________________________________________| 8. The screen is cleared and you are prompted to enter the filelist name that contains the file or files to be repaired. ______________________________________________________________________ | Working directory: / | | | | Type a file or filelist name to add, when done: fichk.f | |______________________________________________________________________| 9. Files that need repair will be listed on the screen and a total of file names added appears at the bottom. You are done, so press . The screen is then cleared and you are prompted to edit the list of filenames. At this point, press . ______________________________________________________________________ | Edit list of names | | 1. Add | | 2. Delete | | 3. Display | | 4. Sort | | 5. Change environment | | 6. Save | | Type number of selection, to end: | |______________________________________________________________________| 10. The system will ask if you want to specify template files, answer no. _____________________________________________________________________ | Do you want to specify template files for keyed files? no | |_____________________________________________________________________| HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB031 Pg002 11. When completed, the screen will flash "PROGRAM IS DONE" and take you back to the ADMIN> prompt. Type in 'ls -l /lost*' and check out the lost files. You may need to contact your application software vendor for file identification. ADMIN> ls -l /lost* 12. To bring the system back into multi-user mode, hold down the CTRL key and the letter 'd'. The system will prompt: _____________________________________________________________________ | single, multi, shutdown or powerdown | |_____________________________________________________________________| 13. Type in the entire word: shutdown. Reboot (start up) the system as you would normally. ORIGINATOR: NSC HANDBOOKS-BASIC FOUR-MAI Basic Four SPx Systems-----FIB031 Pg003