HP-UX System Administrator's Guide: Overview HP-UX 11i v3 (B3921-90011, September 2010)

Full Dump versus Selective Dump
You have chosen this section because it is critical to you to capture the specific instruction or
piece of data that caused your system crash. The only way to guarantee that you have it is to
capture everything. This means selecting to do a full dump of memory.
Be aware, however, that this can be costly from both a time and a disk space perspective. From
the time perspective, it can take quite a while to dump the entire contents of memory from an
HP-UX instance using very large amounts of memory. It can take an additional large amount of
time to copy that memory image to the HP-UX file system area during the reboot process.
From the disk space perspective, if you have large amounts of memory (some HP-UX servers
can have terabytes of memory), you will need an amount of dump area at least equal to the
amount of memory in your system; and, depending on a number of factors, you will need
additional disk space in your HP-UX file system area equaling the amount of physical memory
in your system, in the worst case.
Dump Definitions Built into the Kernel
You can configure HP-UX dump devices using one or more of the following methods:
• Preferred Method: At run time using the /sbin/crashconf command
• At boot time (entries defined in the /etc/fstab file)
• During kernel configuration (put the definitions in the /stand/system file). This method
is obsolescent and should no longer be used!
Definitions at each of these places add to or replace any previous definitions from the other
sources. However, consider the following situation:
Example 3-5 Example of a Crash During the Very Early Stages of the Boot Process
Consider a server that has ten gigabytes (10 GB) of physical memory. If you were to define system
dump devices with a total of two gigabytes (2 GB) of space in the kernel file, and then define an
additional nine gigabytes (9 GB) of disk space in the /etc/fstab file, you would have enough
dump space to hold the entire memory image (a full dump) by the time the system was fully up
and running.
But, what if a crash occurs before /etc/fstab is processed? Only the amount of dump space
already configured will be available at the time of the crash; in this example, two gigabytes of
space.
If it is critical to you to capture every byte of memory in all instances, including the early stages
of the boot process, use crashconf with the -s option (which tells crashconf to retain dump
device definitions across reboots) to define enough dump space in advance to account for this.
crashconf is the preferred method for defining dump devices in HP-UX 11i version 3.
Start-up and Shutdown 83