Primary interrupt controller detected a lost interrupt.
Secondary interrupt controller detected a lost interrupt.
Spurious hardware interrupt xx detected

LAN receive buffer limit reached.
Device / Drive deactivated due to drive failure.

Without trying to over simplify the process, hardware interrupts are processed as follows:

A peripheral device, such as a Network Interface Card (NIC) or Disk Controller, issues an Interrupt Request (IRQ) to the Programmable Interrupt Controller (PIC). Depending upon whether it is IRQ 0 - 7 or IRQ 8 - 15, the IR is handled by the Primary or Secondary PIC, respectively.


The appropriate PIC asserts the Interrupt Request Line (INTR) to the CPU indicating that an interrupt requires processing.


The CPU will respond with an Interrupt Acknowledge (INTA) pulse which causes the PIC to send an IRQ pointer corresponding to the Interrupt Service Routine (ISR) vector (also referred to as the Interrupt Vector).


However, if the IRQ signal is released by the time the INTA response is received, the interrupt is deemed lost. In such cases, the PIC will send the IRQ pointer for the IRQ 7 vector (if lost by the Primary PIC) or the IRQ 15 vector (if lost by the Secondary PIC).


If the ISR is called and the hardware associated with the IRQ does not require servicing, a Spurious interrupt occurs.

Lost Interrupts or Spurious Interrupts are often the result of NIC or Disk controllers being configured to use IRQ 15. Use of IRQ 15 should be avoided, if possible. Reconfiguring the peripherals to make sure none use IRQ 15 will usually resolve the issue.

If none of the peripherals are configured to use IRQ 15, it may be that one of the drivers associated with a peripheral is not fully compatible with the Intel 80486 architecture. Therefore, insure that the latest driver release available for all peripherals is properly installed.

Spurious interrupts are also known to be reported if the Server has a NIC installed which uses an Intel 82586 LAN Co-Processor. In such cases, the best option is to replace the NIC.

If none of these prove to be the cause or provide a solution, it remains entirely possible that the Server may have a defective PIC or that one of the peripherals in encountering an overrun condition (more often seen on NICs than on Disk Controllers). In either case, hardware upgrades may be the best option.

If all else fails, it is possible to disable the alert messages as follows:

SET Display Lost Interrupt Alerts = OFF (default = ON)
SET Display Spurious Interrupt Alerts = OFF (default = ON)