 |
Hostname Standards
In order to achieve maximum flexibility during normal operations,
maintenance, disaster recovery, and business continuity efforts, it is
important to provide a naming standard for business functions that can
be translated easily into hostnames and/or aliases. The purpose of
using hostnames instead of IP addresses is that they are easier to
remember and use. Hostnames are not necessary, but usually
desirable.
Normal user access to an application or business function will always
be through an alias. Normal users should never access a system using a
hostname. The reason is for portability and availability. It is easy
to redirect an alias to any host, it is significantly more difficult to
change hostnames. By having the users access required services through
aliases rather than hostnames, the users can be redirected quickly to
available services in the event of a failure.
Hostnames
In TriParadigm's environment, a hostname refers to an IP address, the IP
address is associated with one or more network adapters. It is
important to recognize that an IP address is not necessarily tied to a
network adapter, but may float across adapters and machines. The same
is true with the hostnames. A hostname should be viewed as
being independent from any machine or data center. The
hostname shall be an enterprise wide unique value in order to eliminate
conflicts during manual, automated, or disaster recovery failovers.
The hostname shall consist of exactly 10 characters with the
following structure:
LocationCode + OS Type + Environment + ApplicationCode + SequenceID
3 char + 1 char + 1 char + 3 char + 2 char = 10 char
The detailed information for each component of the resource group
name is described below:
HostName
Component
|
Number of
Characters
|
Values
|
|
Location Code
|
3
|
ppp = Van Data Center
mtx = Nashville Data Center
|
|
OS Type
|
1
|
a = AIX
s = Sun
|
|
Environment
|
1
|
a = acceptance
a = pre-production
d = test/development
p = production
t = test
x = disaster recovery
x = pre-production
|
|
Application Code
|
3
|
atl = Atlas
ega = EGATE
nim = NIM
ora = Oracle
tps = Maximo
vio = Virtual I/O
|
|
Sequence ID
|
2
|
0-9,A-Z,a-z
|
Examples of Hostnames (HN):
- pppapega01
-
EGATE Production database on AIX at Van Data Center, first instance
- pppapega01
-
EGATE Production database on AIX at Van Data Center, second instance
- mtxapnim01
-
Production Network Information Manager on AIX at Nashville Data Center, first instance
- pppapnim01
-
Production Network Information Manager on AIX at Van Data Center, first instance
- mtxapvio01
-
Production Virtual I/O Server on AIX at Nashville Data Center, first instance
- mtxapvio02
-
Production Virtual I/O Server on AIX at Nashville Data Center, second instance
- mtxapvio03
-
Production Virtual I/O Server on AIX at Nashville Data Center, third instance
- mtxapvio04
-
Production Virtual I/O Server on AIX at Nashville Data Center, fourth instance
- pppaavio01
-
Acceptance Virtual I/O Server on AIX at Van Data Center, first instance
- pppaavio02
-
Acceptance Virtual I/O Server on AIX at Van Data Center, second instance
- pppaavio03
-
Acceptance Virtual I/O Server on AIX at Van Data Center, third instance
- pppaavio04
-
Acceptance Virtual I/O Server on AIX at Van Data Center, fourth instance
Aliases
The rules for defining alias names are significantly less rigid than
for hostnames. The alias can be any name as long as it is unique within
the domain. This allows the application to be accessed though a name
that makes logical sense to the user. For example, the production EGATE
Application Server at the Van Data Center may have a hostname of
"mtxapega03", however the alias may be "mtxegate". The use of aliases
preserves the structure needed for hostnames and the ease of use desired
by users.
|
 |
