[cobalt-developers] Weirdest Memory problem yet...

[Jeff Lasman <jblists@xxxxxxxxxxxxx>]

We've got lots of RaQ4s that work fine, and one that doesn't.

Here's the cat /proc/meminfo taken a few minutes ago; note the "MemFree"
line:

<snip>
 [admin admin]$ cat /proc/meminfo
         total:    used:    free:  shared: buffers:  cached:
 Mem:  529522688 522850304  6672384 92987392 339124224 104116224
 Swap:        0        0        0
 MemTotal:    517112 kB
 MemFree:       6516 kB
 MemShared:    90808 kB
 Buffers:     331176 kB
 Cached:      101676 kB
 SwapTotal:        0 kB
 SwapFree:         0 kB
</snip>

For comparison, here's a cat /proc/meminfo taken from a "known-good"
system with the same software (the bad machine is a backup of the good
one)...

<snip>
 [admin admin]$ cat /proc/meminfo
         total:    used:    free:  shared: buffers:  cached:
 Mem:  529522688 153366528 376156160 148963328 52600832 36524032
 Swap:        0        0        0
 MemTotal:    517112 kB
 MemFree:     367340 kB
 MemShared:   145472 kB
 Buffers:      51368 kB
 Cached:       35668 kB
 SwapTotal:        0 kB
 SwapFree:         0 kB
</snip>

The "good" machine is running live on the 'net; the "bad" machine is on
the net, but no one's hitting it for anything except that we're running
two copies of "top" so two people can watch it.

Any ideas?

Jeff
-- 
Jeff Lasman <jblists@xxxxxxxxxxxxx>
Linux and Cobalt/Sun/RaQ Consulting
nobaloney.net, P. O. Box 52672, Riverside, CA  92517
voice: +1 909 778-9980  *  fax: +1 909 548-9484