This is a list of tests to perform on a slave node after kickstarting it.
Please indicate passed tests with *PASS*, failed
tests with *FAIL* and provide additional information
whenever possible. Also, please alter the above _ to the currnet
kickstart/clone version you are testing.
Note: the slave node must have the correct
bios settings before kickstarting and testing.
Does sensors report the memory, voltages and temperature correctly? No alarms shown?
(Run sensors to find out.)
*FAIL* No sensors found!
Are there any error or warning messages...
On boot up (run dmesg):
During and after latest boot in /var/log/messages:
Do the /data and /datc directories have reduced # of inodes and reserved space for root?
(Check that the total space for /data is approximately 75360860 and approximately 117237964 space for /datc using
df. If the machine has /localscratch2 check for 155972708.) *PASS*
Do /boot, /lib/modules not contain vestiges of old kernels, etc?
Does X run properly using startx as root? This should work on a generic monitor such as a Viewtronics 771.
*PASS* TWM starts
Do gcc and ddd work correctly?
Does networking run properly? Is it running full duplex 100baseT (you can use mii-tool to test)?
*PASS* eth0: negotiated 100baseTx-FD flow-control, link ok
Does it correctly gets its identity (eg, s012) using dhcp from the master on bootup?
Does the system shut down and power off with shutdown -h now?
After power-down, if AC power is cycled, does system remain off?
Does the system reboot with shutdown -r now?
If a running system is unplugged, then plugged in, does it remain off?
If system is cleanly shutdown can that system then be powered on from another machine using etherwake or loopWake?
Does it fsck and boot up correctly?
Does the machine act properly as a UPS master and UPS slave? If a UPS
master, does it properly issue commands to shut down other machines under its
control when the battery is low on the UPS? If a UPS slave, does it
understand commands given to it by a UPS master? If power is restored to the
UPS, does the slave remain turned off?
*PASS* s308 logs into s001 as its master
Are there no files with dates in the future? (One way to check this is
to do the following: cd / ; touch temp.data ; find . -cnewer temp.data
-print -xdev ; rm -f temp.data If any files are listed, they have
dates set later than the moment you executed the above command.)
Is NTP running correctly? Is the hardware clock synchronized with the
software clock after the software clock has had its time synched with an ntp
server? (Look in /var/log/messages for messages from xntpd. If ntp is
working correctly, the /etc/ntp/drift file will contain a non-zero entry.
Set the hardware clock in the bios to something incorrect (like last week).
Boot up. Check the BIOS clock again. It has been set correctly to GMT. This
might not work if you are off the UWM subnet. It depends how tight they are
on securing the ntp servers.)
*PASS*But no /etc/ntp/drift
Is vgablank running correctly? (It should call vga_screenoff ten minutes
after keyboard input stops, and calls vga_screenon when input resumes.)
Does running drag 1.2 show > 290 Mflops (at n=220) provided the screen is blanked
Does hdparm -tT /dev/hda report good speeds (~120 and ~28 MB/s)?
Does hdparm -tT /dev/hdc report good speeds (~120 and ~28 MB/s)?
Does hdparm -tT /dev/hdd show good results with a cd in the drive?
Is automount on the slave configured so that cd /mnt/cdrom
works correctly if a cd is present?
Are big files (2 GB and higher) properly supported? If the following commands do not dump
core, they should be supported:
cp /etc/termcap a; cat a a a a a > b; cat b b b b b > c; cat c c c c c > d;
cat d d d d d > e; cat e e e e e e > f
Does automount work?
On the master, does cd /net/s012 work properly (with whatever the correct slave node # is)?
Does cd /net/m001 on the slave properly mount the master?
Does cp file /net/s012/ work correctly?
Can root on the master log into a node with just rsh s012 (no password needed)?
*PASS* With odd behavior
Does rsh work correctly, e.g. does rsh s012 uptime correctly return the uptime
on s012 from m001?
Do the above automount and rsh tests work correctly between two slave nodes?
*FAIL*You can get in, but you can't get out.
Is the file /root/cloned_date present and contain the correct information?
Does recloning over the network preserve the /data and /datc partitions and remake
the / and /boot partitions?
Does /proc/fans contain 'files' that correctly report the fan RPMS?
*FAIL* Fans RPM not installed
Does mail correctly relay to master001.medusa.phys.uwm.edu?
Is sensorscheck operational and reporting errors if they occur?