This appendix discusses:
Database terminology.
DBspace strategy.
Database server directory structure.
Troubleshooting model.
PeopleSoft uses the following technical naming conventions for databases:
Database |
A set of data tables accessed and managed as a group. Informix manages the database at the system level. |
Informix Database Server |
A cooperating set of host processes and shared memory capable of managing one or more databases—a running online engine. Corresponds to a specific INFORMIXSERVER value and a matching entry in the sqlhosts file. May contain many databases each with its own catalog. |
A collection of database objects. PeopleSoft uses tables, indexes. An object set corresponds to the PeopleSoft owner ID. |
The following strategies can be used for DBspace:
Separate the root dbspace, physical logs, logical logs, and the temporary dbspace from one another and from the application dbspaces.
Place the root dbspace, logical log, and physical log in separate dbspaces on separate disks.
Separate certain high volume application tables to optimize performance.
A minimum configuration for production systems is four physical drives (at least one each for database files, physical logs, logical logs, and temporary area).
Separate the dbspace for data on one drive and dbspace for indexes on a separate drive.
The environment variable $INFORMIXDIR points to the directory where Informix is installed on your machine; normally this is set to /usr/INFORMIX. The standard Informix directory structure is built under INFORMIX directory by the Informix install process.
The standard Informix architecture uses the “Two-Task Model.” In this architecture, when a user connects to the database server a network thread is created to handle the network processing for that user.
Each Informix server instance consists of the following pieces:
At least eight database processes that operate the database.
At least 3 shared memory segments, through which the database processes communicate.
The $INFORMIXDIR/etc/sqlhosts file holds networking parameters for each accessible server instance.
Entries include server name, network protocol, host name and tcp-ip service.
The $INFORMIXDIR/etc/$ONCONFIG file.
This file primarily holds shared memory configuration parameters for the local server instance. These include the number of buffers, number of locks, size of the initial shared memory segment, and so on The “onconfig” file also includes pointers to the root dbspace, the temporary dbspace, and the physical log dbspace, as well as the names of the backup tape devices. By convention, these files are often given a name such as onconfig.inf11, where inf11 is the name of the server. This is helpful when managing multiple server instances on one host.
Database “chunks.”
Hold the data stored in the database. Under UNIX, these may be either “raw” or “cooked” files. In either case they should be stored in a common directory, with links pointing to their physical locations, if necessary.
See Also
Administrator’s Guide for Informix Dynamic Server
This section discusses some steps you can take to diagnose system signon problems. Understanding basic operations and process flow is essential when you are troubleshooting connectivity errors. Use the following model as a reference for this section.
Test terminal connection.
Try using TELNET, or a similar network utility, to get a terminal connection to your database server. If this succeeds, you probably have a problem with the way Informix-Connect or Informix is set up. Check to see if the Informix database server is active. If Step 2 fails, then the problem is within the networking layer.
Consult your networking experts.
The problem has been isolated to something within the network layer. Try to isolate the network problem. Can you log on to other servers? Are other terminals still able to connect? Try lowering level network diagnostics, if they exist