6-STABLE locks solid - current ok, why?

Per olof Ljungmark peo at intersonic.se
Mon Jul 24 20:32:56 UTC 2006


Daniel Bond wrote:
> On 22:22 Mon 24 Jul, Per olof Ljungmark wrote:
>> Hi,
>>
>> Got a testbed Proliant DL380G2, internal SmartArray 5i disabled, have SmartArray 5300 with 6 disks.
>>
>> Installed 6.1-REL last week, rebuilt to -STABLE three times since without a problem. However, when I try to 
>> add various (random) applications the box locks up solid during "configure" or "make". Apps tested are 
>> postfix, jdk15, OpenOffice-2.0 and more. When it hangs the only thing remaining is power cycling, no 
>> keyboard access possible.
>>
>> What puzzles me here is that now I installed -CURRENT from yesterday on same box without any other changes 
>> and it works like charm, so far I'm through building close to 50 ports including kde with dependencies.
>>
>> Next step I guess is to go back to RELENG and check again. Meanwhile, does anyone out there have an idea 
>> where I should look? This box is going into production and I'm not brave enough to run -CURRENT...
> 
> Hi,
> 
> try to disable ACPI. It can be usefull for reading CPU temperature and fan
> speeds, but has no real function on a server. It's mostly usefull for power
> managment, supend and resume support. From the handbook:
> 
> Most system hangs are a result of lost interrupts or an interrupt storm.
> Chipsets have a lot of problems based on how the BIOS configures interrupts
> before boot, correctness of the APIC (MADT) table, and routing of the System
> Control Interrupt (SCI).
> 
> Interrupt storms can be distinguished from lost interrupts by checking the
> output of vmstat -i and looking at the line that has acpi0. If the counter is
> increasing at more than a couple per second, you have an interrupt storm. If
> the system appears hung, try breaking to DDB (CTRL+ALT+ESC on console) and
> type show interrupts.
> 
> Your best hope when dealing with interrupt problems is to try disabling APIC
> support with hint.apic.0.disabled="1" in loader.conf
> 
Sorry, I should have mentioned that I tried both ways with same outcome.


More information about the freebsd-stable mailing list