Commit 09be27eb authored by Kirk Webb's avatar Kirk Webb
Browse files

* Software Setup section of the emulab FAQ updated:

Section mentioning non-default daemons started by testbed scripts updated
with slothd info, and corrections were made to the healthd info.
parent eba8fff8
......@@ -683,7 +683,6 @@
href="tutorial/tutorial.php3#CustomOS">Custom OS</a> section of
the <a href = "tutorial/tutorial.php3">Emulab Tutorial.</a>
</p>
</ul>
<li><a NAME="SWS-8"></a>
<h3>What if I need more disk space on my nodes?</h3>
......@@ -698,26 +697,53 @@
mount /dev/hda4 /mnt;
</code>
</blockquote>
This partition is only 6 Gigs, the size of the leftover space on our
smallest drives. If you need more space than this, it would be possible
to enlarge this partition on some machines (for example, our pc850s
have 40 GB disks,) but that is outside the scope of this FAQ.
</p>
</ul>
<li><a NAME="SWS-9"></a>
<h3>Are there testbed-specific daemons that could interfere with my
experiment?</h3>
<p>
The only daemon that the testbed adds to the OS's standard set is
<code>healthd</code>. This daemon periodically polls things such as
motherboard/CPU temperature and fan speed, and reports them back to
our <code>boss</code> node. We have picked a large polling interval
(five minutes) to minimize the impact on nodes. If you are worried
about side effects, however, you can safely kill it off.
<code>healthd</code> is stared by <code>/etc/testbed/rc.healthd</code> .
By default, the testbed startup scripts currently start two daemons in
addition to the OS's standard set:
</p>
<p>
<blockquote>
<li><code>healthd</code> - A low overhead hardware health monitor.
</blockquote>
</p>
<p>
This deamon periodically polls the machine's health monitoring
hardware and sends this information back to our <code>boss</code>
node for analysis. The hardware is polled once per second, and a
status datagram is sent out once every five minutes.
<code>Healthd</code>'s overhead is quite low, but it can be safely
killed and disabled from startup if you're worried about possible
side effects. It is started by <code>/etc/testbed/rc.healthd</code>.
</p>
<p>
<blockquote>
<li><code>slothd</code> - A low overhead usage analysis tool.
</blockquote>
</p>
<p>
<code>Slothd</code> is important to efficient testbed utilization
and should run on every node whenever possible. Its overhead is
almost negligible (essentially like running 'ls -l /dev' once per
hour), and should not interfere with your work. However if your
experiment is exceptionally sensitive, then you may arrange with
us to disable <code>slothd</code>. Please note that we will
restart this daemon if it is not running unless prior arrangements
have been made.
</p>
<p>
If you have requested automatic routing on your nodes with the
<code>tb-set-ip-routing</code> command in your NS file, this will
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment