web5 reboot

[Jeff]

Hi,

web5 is currently undergoing an emergency reboot. More information will be available later today as it becomes available. We apologize for any inconvenience this may cause.

[Jeff]

web5 is back online. Again we apologize for any inconvenience this may have caused.

[Jeff]

Below is an excerpt of the logs the Operating System's kernel wrote to the system log file earlier,
indicating the issue:


web5 kernel: Uhhuh. NMI received for unknown reason 31 on CPU 0.
web5 kernel: Dazed and confused, but trying to continue
web5 kernel: Do you have a strange power saving mode enabled?



Basically what happened is that there was a power issue due to some cards in the server needing to be reseated, which is believed to have resolved the issue.

For the techies - the NMI (Non-Maskable Interrupt) logs above were followed by a few LAPIC_NMI and ACPI messages, indicating a power issue. A quick trip to the datacenter (we are local to the Equinix facility in Ashburn, VA, thus we don't ever need to rely on the datacenters' "remote hands"), the reseating of a few cards, and everything is believed to be back to normal.

[cvos]

i have to know - Dazed and confused - is this a technical term, or born from kernal programmers who went through their formative years in the 1960's

[Jeff]

I couldn't tell you, but this guy could:


# head -8 traps.c
/*
* linux/arch/i386/traps.c
*
* Copyright (C) 1991, 1992 Linus Torvalds
*
* Pentium III FXSR, SSE support
* Gareth Hughes <gareth@valinux.com>, May 2000
*/






558 static void unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
559 {
560 #ifdef CONFIG_MCA
561 /* Might actually be able to figure out what the guilty party
562 * is. */
563 if( MCA_bus ) {
564 mca_handle_nmi();
565 return;
566 }
567 #endif
568 printk("Uhhuh. NMI received for unknown reason %02x on CPU %d.\n",
569 reason, smp_processor_id());
570 printk("Dazed and confused, but trying to continue\n");
571 printk("Do you have a strange power saving mode enabled?\n");
572 }