i am pretty sure that in my case it is not a hardware.

If you're having hard locks, there's 3 thinks to check right off the bat. Memory, PSU, and cooling. Those can be checked via software like lm-sensors, but software is never reliable for find out for sure. Memory testers tend to be false negative. Both software and hardware based ones. The best way to check memory if you've got more than one stick is run with a single stick for awhile and see if things seem more stable, then try running with just the other one for awhile and see if there is any difference. You may also wanna try experiment with setting your memory timings more conservatively. If you find that when you make the timing more conservative than what the memory is rated at and stability gets better, then that is a good indication that you have some really poor grade memory in there. Personally, I don't touch OEM memory with a ten-foot pole. Have ran into way too many systems that ended up having all sorts of random problems (most often is random freezes, and sometimes data corruption over time) from it turning out to be bad memory that was some OEM flavor of the month (and pretty much every time the offending memory checked out ok at any test I threw at it. Even passed dedicated memory testing devices. Yet when I put a good quality one in, problems instantly gone and system solid as a rock from that day on). I will only buy memory from a reputable brand that comes in sealed retail packaging these days.

The best way to test the PSU is with a voltage meter. You could use a program like lm-sensors (if your motherboard supports it) for a fairly good indication (this sort of things tends not always be that accurate with some motherboards, but it's still a handy tool to help diagnose.). If your PSU is an el-cheapo, then I would and get one of decent quality anyway. Sub-quality PSUs are responsible for all sorts of problems. Everything from lack of system stability to eventually or suddenly destroying hardware like a motherboard, CPU, and/or video cards, and the list can go on. It pays off in the long to get a good quality one. It very well could be that the capacitor issues like described earlier in this thread, are the likely result from very sub-par voltage regulation of a crappy PSU. It would actually make alot of sense. lm-sensor can also be handy for checking your temperature as well.

I got a question for the guy whose motherboard capacitors were going. Were you by any chance running with a Q-tec PSU?

Why? Because it has less problems with XFCE? I don't use XFCE but it's been the general concensus that XFCE is less of a memory hog than gnome. If, in your case it turned out to be bad memory, then it might be that since gnome is more memory intensive that it is more likely to bring the issues of bad memory more to the surface.

Linux, by nature, doesn't tend to lockup with good hardware. Linux by it's very nature is quite robust as OSes go. However, it tends to make appearant any faulty hardware much more than Windows does.

Now if you were having issues where programs were acting flaky but the operating system keeps on going on strong, then I would suspect it being software related. For instance, say the system starts acting weird and you kills some processes and it fine after that, then that would indicate that something software related was flaking out.

One other things, make sure you're not running out of memory. However I've noticed that when memory starts getting tight that Linux tends to be pretty good about kill off processes and still going on strong. Like I used to keep too small of a swap partition in the past and gnome would sometimes die on me, but the system as a whole didn't go down at all.

Of course anything is always possible, but your problems seems to be indicative of a possible hardware issue. Lockups and crashes shouldn't happen that often with Linux unless things are really setup in a very very bad way or there is some sort of data corruption going on.

Another thing, is you might want to check your logs if you haven't already for more possible clues to help diagnose the problem.

I've also heard that compiling is a good way to test out one's hardware, especially memory, since it tends to put it to the test.

Yes, please check the hardware as I told you. Software tests like compiling are a good way to do this. Also since you said that it wasn't a specific app that froze your system I'm convinced it's hardware.

The only thing that has ever caused my systems to lock up hard is software. Some Flash movies and javascript in browsers, and certain gl actions including screensavers. If you're convinced that it's not hardware, I would try upgrading to the newest kernel, and the newest xorg. Also your video card driver is a consideration.

I an sorry ,I have the same problem
debian etch netinstall cd
kernel 2.6.16-2-686
it always hungs.
each time is different,most of the time ,no any logs,
here is the latest one,

continue,log continued

and it always reboot with no portent,
the latest time is when
apt-get install gcc,
it reboots again,
why?

processor : 0
vendor_id : GenuineIntel
cpu family : 15
model : 2
model name : Intel(R) Celeron(R) CPU 2.60GHz
stepping : 9
cpu MHz : 2626.250
cache size : 128 KB
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 2
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe cid xtpr
bogomips : 5260.89

i do not have less problems. i have NO problem with xfce and icewm. i tried many simulataneous applications like cd player, firefox, thunderbird, oo applications, process monitor. no problems whatsoever.

and for the same matter there are no freezes with windows xp (dual-boot) and there were none with windows me (previous os) while running memory consuming applications.
i tried checking my harware and i cannot see any issues.