Also try turning HPET off in the BIOS, as I've had problems w/ RTC when HPET is on...

In my continued obsession to get my workstation running very quiet and very stable (i.e. cool)... I found this good link on temps:

http://www.heatsink-guide.com/content.p ... temp.shtml

So, here is hopefully a little slightly interesting follow up... I just tried 3 scenarios (note just messing with the primary CPU fan, load is at my baseline w/ qmail and apache and X running but no intensive stuff):

1. AMD's fan control:


fermat ~ # sensors
w83697hf-isa-0290
Adapter: ISA adapter ...

Yes, "Don't try this at home," thanks...

I'm not exactly a novice and the box has 3 nexus 120 fans and a thermalright 120 heatsink on the CPU. 2 fans are acting on the heatsink in a push-pull configuration. I could probably remove 1 fan completely and the box would be fine, but run slightly hot ...

Thanks.... I can probably use fancontrol/ pwmconfig to test out what AMD's Cool & Quiet is doing and sort of reverse engineer the table I'm looking for, and it sounds like it will do just what I'm looking for.

Its my understanding that in general most fan controllers "Undervolt" 12V fans (6-7 volts ...

1st, yes I have searched the forums and google'd... while I expected to find this info I didn't.... AMD's site is filled to brim w/ marketing crap!! 8O I wondering if anyone else has seen the specifics from AMD as far as a table that lists Cool & Quiet fan voltages out vs CPU/MB temps? Also, if ...

I ran into this same issue... try downgrading your genkernel version (assuming you didn't roll your own).


emerge unmerge genkernel
emerge =sys-kernel/genkernel-3.3.10


Long term you may want to use /etc/portage/package.mask if this isn't fixed in the next revision.

Check out:

http://forums ...

I don't think the init script is using apache2ctl (I assume you meant apache2ctl), so I think its default location is not relevant. You might see if the init script or conf.d/apache2 (may have paths in it) file doesn't match a fresh one... My apache2 init script has:


reload() {
if ! service ...

Did you try:

revdep-rebuild


You might also try re-emerging apache as well. You haven't recently upgraded apache, have you? I ask because there have been some gentoo specific layout changes that could cause problems. See

http://www.gentoolinux.org/doc/en/apache-upgrading.xml

and

http://www ...

It is certainly ok to run multiple commands in a cron job. I have some questions:

1. Have you tried the various commands that are suspects in the apache crash separately, i.e. at different times and checked apache to see if 1 in particular crashes it?

2. Have you tried the various commands via the ...

Late reply sorry (damn work), but I'm including this in case someone else runs into a simular problem. It turns out the initramfs (initrd) is screwed up. I can use genkernel kernel (as contrasted with 'all') and the system boots fine. I obviously have the boot device driver built into the kernel ...

Thanks... but I seemed to have narrowed it down to genkernel, as I can compile the same kernel manually using the same config with no other changes and it boots nicely. Looking into it...

First time I've ran into this problem and not been able to solve it quickly... someone HELP ;-)

I've done an emerge -up world and upgraded my kernel... when I reboot to the new kernel I get root device not found (prior to this I see that udev has been killed). However, I can enter the busybox shell ...

Thanks much...

I just changed the scheduler to cfq, and added disable_irm=1 and serialize_io=1 to ieee1394 and sbp2. I'm going to try things like that, and then update the kernel/module later. I'm about to go on vacation

I have a server that we use to backup copious amounts of GIS-related data, on the order of 300-400GB at one time. The backup is a simple rsync running off a cron. We are attempting to backup this data to LaCie BigDisk Extreme 500GB firewire drives for offsite storage. What happens is that after some ...

Check out:

http://httpd.apache.org/docs/1.3/misc/perf-tuning.html

or

http://httpd.apache.org/docs/2.0/misc/perf-tuning.html

A couple of things (you may have already, but...):

1) Does it to the same thing with ping? The reason I ask is that ssh in particular can "Stall" if it needs to regenerate a key or if it is trying to do a reverse lookup and the reverse is failing.

2) Have you checked the BIOS as far as attempting ...

Looks fine to me. Check your modules and cross you fingers.

iostat is part of sysstat:
It includes sar as well. I would _guess_ that killing the updatedb process wouldn't be so bad. You might need to remove the db file and regenerate it. It is usually:

slocate's updatedb can drive up the load average because it (obviously) uses a lot of disk while it indexes files. Does the machine have a big disk or array?

Do you have iostat installed? What does it say?

Damn, I hate those sort of reboots! Wait to see if its back to the office or if its time to ...

If you have udev running then you shouldn't have to use mknod. I'd suggest looking at dmesg then looking inside of /sys and creating a custom rule for it (if need be).

I have for example:


paul@fermat $ grep jump /etc/udev/rules.d/10-local.rules
BUS="usb", SYSFS{manufacturer}="LEXAR MEDIA", SYSFS ...

Guys... this guy is just trolling.... don't waste your time.

You don't need to hand type it,just put a floppy or usb key drive in there and (as root otherwise some stuff may not be in your path):


id > /tmp/output
route -n >> /tmp/output
ifconfig >> /tmp/output
lspci >> /tmp/output
lsmod >> /tmp/output
cat /etc/conf.d/net | egrep -v '^#' >> /tmp/output
cp ...

Not much to go on... need more info.

1. What does your routing table look like? Try: route -n (-n because I assume DNS is not working).

2. Please include your ifconfig.

3. What do lscpi and lsmod show? ...to show the hardware you have and what modules are loaded... can check that they match and ...

What does the makefile look like? Have you checked out:

http://www.gentoo.org/proj/en/devrel/handbook/handbook.xml?part=2&chap=1#doc_chap3

... D,S,src_install, install vs einstall

ps: not necessarily bad to install to /usr/local if its a "Local" package (http://www.pathname.com/fhs/). Of course ...