Compiled and started running -ac8 last night. The parts I use seem to
work ok, including ppp, which was broken for me for a while (I only
occassionaly need it anymore). I decided to see how the kernel deals
with heavy load.
The system is a P166 notebook, with 144 MB memory and a 68 MB swap
partition.
A "make -j" slowly over the course of 5 minutes drives the load to
about 30. At first the degradation is controlled, with sendmail
refusing service, but at about 160 process visible in top, top quits
updating (set a 8 second updates), showing about 2 MB swap used. At
this point it sounds like the system is thrashing. The pointer in X
quits responding about the same time. The xclock quits updating about
a minute later. The "tail -f syslog" freezes at this point (note that
these displays are under X).
About 5 minutes later it appears (based on 30 second spacing) that the
only thing still working is the system sync. After 5 more minutes,
this seems to fail also (or perhaps it has nothing needing sync-ing),
and finally the 65 seconds setting in hdparm to spin the hard disk
down activates. Five minutes later, there is still silence.
Is this failure process acceptable? I'd think the system should react
differently to the thrashing, killing off the load demanding user
process(es), rather than degrading to a point of freeze.
Regards,
Billy
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Jun 07 2000 - 21:00:21 EST