Jump to content

MicroVAX II Troubleshooting Guide

From RetroTechCollection
KA630-AA CPU module; its four red LEDs show the power-up test number in binary (Phiarc, CC BY-SA 4.0)

A MicroVAX II reports its own progress at power-up: the KA630 boot ROM counts down through its tests from F to 0 on the 7-segment display of the rear CPU insert, on four red LEDs on the CPU module and, from 7 down, on the console terminal. If a test fails, the count stops on the failing test. DEC's MicroVAX II Maintenance Guide lists the probable failing units for each number, and the owner's manual covers the simpler power and terminal problems.[1][2]

Power and console problems

[edit | edit source]
First checks (630QB Owner's Manual, Table 3-2)
Problem Possible cause Action
Nothing when switched on, switch not lit No mains, or the system circuit breaker has tripped Plug in; try another outlet; reset the breaker at the rear. If it trips again, the fault needs investigating
Power on, no console display Terminal off or off-line, cable not seated, baud rates mismatched, mode switch on the CPU insert at T Check the terminal and cable; match the baud rate; set the mode switch for normal running
Power on, no display, terminal good Console interface on the CPU See the LED codes below; codes 9 and 8 cover the console

[2] The control panel DC OK light should glow green, the RUN light should come on, and a fixed disk's Ready light should glow green within 30 seconds of power-on.[3] A green DC OK LED on the CPU module itself shows that the module sees DC OK.[1]

Power-up test codes

[edit | edit source]
KA630 power-up codes and probable failing units (MicroVAX II Maintenance Guide, Table 7-1)
Code Test Probable failing units, in DEC's order
F Waiting for DC OK KA630 not seeing DC OK; power supply negating DC OK; a Q22-bus device negating DC OK; backplane with DC OK shorted; power supply cable
E Waiting for POK KA630 not seeing POK; power supply; a Q22-bus device negating POK; backplane with POK shorted
D Checksum of the CPU ROM KA630
C Searching for the RAM the ROM programs need KA630; MS630; KA630 to MS630 ribbon cable short or open
B Reading the KA630 IPCR register (Q22-bus access) KA630; a Q22-bus device or the backplane preventing the CPU from getting the bus
A Testing the VCB01 video console, if fitted VCB01 keyboard, display, module or insert; KA630; a Q22-bus device or backplane
9 Identifying the console terminal KA630 (after 6 seconds without a response the CPU moves on)
8 Language inquiry or CPU halted Normal while waiting for a language (times out after 2 to 6 minutes). If no heading appeared on the terminal: console interface on the KA630, console cables, baud rate, terminal off or broken, CPU insert broken or wrongly set
7 Data tests on RAM KA630 RAM; MS630; backplane CD interconnect; ribbon cable
6 Address tests on RAM MS630; backplane CD interconnect; ribbon cable
5 Tests using the Q22-bus map to reach local memory KA630; a Q22-bus device or backplane
4 CPU instruction and register tests KA630
3 Interrupt tests KA630; a Q22-bus device requesting interrupts wrongly; backplane with a BR4–BR7 line shorted
2 Searching for a bootstrap device RQDX controller, RD5x or RX50 drive or cable; TQK50, TK50 or cable; MRV11-D; DEQNA, its cables, fuse or transceiver; KA630
1 Bootstrap device found The Q22-bus boot device, its signal or power cable, or the KA630
0 Testing completed –

[1] The owner's manual describes code 1 as "loading secondary start", and advises that a machine stopping at 1 most likely has a boot device without bootable software.[2] Test 7 builds a map of good and bad memory pages and fails only if it finds less than 64 KB of good memory; the service version of MMS can read the map to locate bad pages.[1]

Subtests

[edit | edit source]

For tests 7 to 3 the console prints the failing subtest after the test number. DEC's subtests include:[4]

  • Test 7: 64 KB of good memory not found; unexpected interrupt; error in a known good page.
  • Test 6: error in memory on the KA630, on the first MS630 or on the second MS630; error in a known good page; write wrong parity logic error; non-existent memory time-out.
  • Test 5: IPCR could not be cleared; bits do not match the mask; unexpected interrupt.
  • Test 4: byte mask error; 64 KB of good memory not found; memory management unit error; floating-point unit error.
  • Test 3: interval timer error; console error; IPCR and MSER subtest failures; could not drive data into memory over the Q22-bus; failure to create a non-existent memory time-out.

Console loopback test

[edit | edit source]

Setting the power mode switch on the CPU insert to T before power-up runs a continuous loopback test of the console port. The insert's display should alternate between 9 and F. Anything else means the insert, its cables or the CPU has failed. Repeating the test with a KA630CNF configuration board (H3263-00) in place of the insert tells them apart: if it then fails (judged by the LEDs on the CPU module), the CPU is bad; if it passes, the insert or cabling is.[1]

Console error messages

[edit | edit source]

The console prints a hexadecimal code and mnemonic when something goes wrong in console mode or during booting. Some that point at hardware:[5]

Selected KA630 console error codes (MicroVAX II Maintenance Guide, appendix B)
Code Mnemonic Meaning
05 DBL ERR A second machine check occurred while reporting the first to the operating system
0C SCB RD ERR Hard memory error while reading an exception or interrupt vector
21 FND ERR A FIND command could not find the restart parameter block or 64 KB of good memory
23 MEM ERR Parity error detected
40 NOSUCHDEV No bootable devices found
4B CTRLERR Boot device I/O error
4C DEVINACT Failed to initialise the boot device
4D DEVOFFLINE Device is off line
4E MEMERR Memory initialisation error
53 INSFMAPREG Invalid memory configuration
54 RETRY No devices bootable, retrying

[5]

Disk problems

[edit | edit source]
  • Check signal and power cables before condemning a drive or controller, then look at the power-up LEDs on the devices themselves.[1]
  • RD53 (Micropolis 1325) drives often spin up and then spin down. Jörg Hoppe had two drives with this symptom: one never reached speed (a worn spindle bearing or motor), the other reached speed but its heads stayed against the inner track (actuator fault). He reports others recommending cooling the drive or freeing heads stuck to their rest position.[6] See Hard Drive Maintenance and Repair.
[edit | edit source]

References

[edit | edit source]
  1. ↑ 1.0 1.1 1.2 1.3 1.4 1.5 Digital Equipment Corporation, MicroVAX II Maintenance Guide, AZ-GM3AA-MN, first edition October 1985, section 7.2.1, Table 7-1 and Figure 7-1, pp. 7-1 to 7-7 (File:MicroVAX II Maintenance Guide (AZ-GM3AA-MN).pdf).
  2. ↑ 2.0 2.1 2.2 DEC, AZ-GLCAA-MN, chapter 3, power-on messages, Table 3-1 and Table 3-2, pp. 3-2 to 3-6.
  3. ↑ Digital Equipment Corporation, MicroVAX II 630QB Owner's Manual, AZ-GLCAA-MN, Tables 2-1 and 2-2, pp. 2-4 to 2-6 (File:MicroVAX II 630QB Owner's Manual (AZ-GLCAA-MN).pdf).
  4. ↑ DEC, AZ-GM3AA-MN, Figure 7-2 and Table 7-2, subtests for tests 7 to 3.
  5. ↑ 5.0 5.1 DEC, AZ-GM3AA-MN, appendix B, console error messages and explanations, pp. B-1 to B-4.
  6. ↑ Jörg Hoppe, "Defective DEC RD53 disk drives", retrocmp.com, retrocmp.com.