On Mon, 2005-08-01 at 02:30 -0400, Peter Teuben wrote: > > Hi list, > > Once again my FC4 box (Duron 1200mHz, 512 ram) has crawled to a halt. > > I quickly top'ed and found a process prelink that varied between > > 74%-98% CPU usage. > > > > PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND > > 5202 root 39 19 11724 9280 536 R 89.9 1.9 0:23.59 prelink > > > > So I googled for prelink and now know that it helps preload libraries > > so that progreams can start faster. Which is senseable because the > > machine started slowing down when I clicked 'manage bookmarks' in > > firefox, which opens another program. My question is, what can I do to > > prevent this from happening again? Do I have to ust cross my fingers > > every time I open a new program? > > > > Could this be related to the fact that I have no swap partition? I > > know that I should probably add one, I guess that I've just been too > > lazy to learn how. Will adding a swap partition make the machine less > > sluggish? For this 512 ram machine, I was thinking of adding a 2 gig > > swap partition. > > I had reported on a similar problem, and finally paid attention to the > output of "df" which showed 0 usage of swap. In fact, total was 0 too, > i.e. i had no swap despite that it was installed with swap. Turned > out the /etc/fstab file had a rather curious line in it: > > LABEL= swap swap defaults 0 0 > > where the = sign was followed by 15 0xAA characters (that dind't print > in the cut&paste above)!! So, no wonder. I suspect it is something in > the installed that is broken, since i install redhat systems fairly > regularly, or i overlooked a new(?) requirement that partitions be > labeled. > > Anyways, after hardcoding > /dev/hda5 swap swap defaults 0 0 > and running 'swapon' i'm back in business. e2label and making a nice label > of course also would do the job. Careful, not e2label.... To re-create the label on the swap device it appears that we have to actually recreate it with "mkswap -L labelname /dev/partitionname" But, I suspect you might be right about an anaconda anomaly..... --Rob
