Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations derfloh on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
PDT
- Thread starter Roomer
- Start date
Performance Diagnostic Tool (PDT) is used to determine potential performance problems on systems. It is an optional installable component of Base Operating System (BOS) and the fileset is called bos.perf.diag_tool. PDT takes about 1MB of disk space.
PDT application consists of 3 components :-
1. Collection :- programs that run from cron or by the user to collect and record the data.
2. Retention :- The data collected is reviewed and discards the data that is obsolete.
3. Reporting :- A report is produced from the current set of data collected.
PDT attempts to identify :-
1. Unbalanced resources i.e. disks , paging space , spread of disks
2. Usage trends , How many users use the system
3. Processes take up a lot of CPU, inappropriate system parameters
4. Hardware / Software problems that may affect system performance
NB:- After the use of this tool it still requires further system admin as to find the root cause of the problem .
To enable PDT execute the folowing script as root :- /usr/sbin/perf/diag_tool/pdt_config
A menu will be displayed , select option 4 , this will add entries to the cron entry adm uder /usr/spool/cron/crontabs
Sample report would be something like this
I/O CONFIGURATION
- Note: volume hdisk7 has 6848 MB available for allocation
while volume hdisk5 has 0 MB available
PAGING CONFIGURATION
- Physical Volume hdisk1 (type: SCSI) has no paging space defined
- All paging spaces have been defined on one Physical volume (hdisk1)
- Physical Volume hdisk7 has 800 MB of paging space,
while Physical Volume hdisk0 has only 256 MB
- Paging space hd6 on volume group rootvg is fragmented
I/O BALANCE
- Phys. volume hdisk10 is not busy
volume hdisk10, mean util. = 0.00 %
- Phys. volume hdisk6 is not busy
volume hdisk6, mean util. = 0.00 %
FILE SYSTEMS
- File system hd2 (/usr) is nearly full at 92 %
VIRTUAL MEMORY
- VMM has identified 9 bad frames
you can set the severity level depending on what you want to monitor.
Severity 1 Problems
 Journaled File System (JFS) becomes unavailable
 JFS nearly full
 Physical volume not allocated to a volume group
 All paging spaces defined on one physical volume
 System appears to have too little memory for current workload
 Page space nearly full
 Possible problems in the settings of load control parameters
 VMM-detected bad memory frames
 Any host in .nodes becomes unreachable
Severity 2 Problems
 Imbalance in the I/O configuration (for example, disks per adapter)
 Imbalance in allocation of paging space on physical volumes with paging space
 Fragmentation of a paging space in a volume group
 Significant imbalance in measured I/O load to physical volumes
 New process is identified as a heavy memory or CPU consumer
 A file in .files exhibits systematic growth (or decline) in size
 A file system or page space exhibits systematic growth (or decline) in space utilization
 A host in .nodes exhibits degradation in ping delays or packet loss percentage
 A getty process consumes too much CPU time
 A process with high CPU or memory consumption exhibits systematic growth (or decline) in resource use
 maxuproc indicated as being possibly too low for a particular user ID
 A WORKLOAD TRACKING indicator shows an upward trend
Severity 3 Messages
Severity 3 messages provide additional detail about problems identified at severity levels 1 and 2. This includes the data-collection characteristics, such as number of samples, for severity 1 and 2 messages.
There are threshold files which you can set up for monitoring specific tasks and if they are reached , allow you to send an email or an alert .
The tool is o.k to start off getting an overview of system , but i found that you can write your own , and tailor it according to your own needs.
HTH a little
laters
PDT application consists of 3 components :-
1. Collection :- programs that run from cron or by the user to collect and record the data.
2. Retention :- The data collected is reviewed and discards the data that is obsolete.
3. Reporting :- A report is produced from the current set of data collected.
PDT attempts to identify :-
1. Unbalanced resources i.e. disks , paging space , spread of disks
2. Usage trends , How many users use the system
3. Processes take up a lot of CPU, inappropriate system parameters
4. Hardware / Software problems that may affect system performance
NB:- After the use of this tool it still requires further system admin as to find the root cause of the problem .
To enable PDT execute the folowing script as root :- /usr/sbin/perf/diag_tool/pdt_config
A menu will be displayed , select option 4 , this will add entries to the cron entry adm uder /usr/spool/cron/crontabs
Sample report would be something like this
I/O CONFIGURATION
- Note: volume hdisk7 has 6848 MB available for allocation
while volume hdisk5 has 0 MB available
PAGING CONFIGURATION
- Physical Volume hdisk1 (type: SCSI) has no paging space defined
- All paging spaces have been defined on one Physical volume (hdisk1)
- Physical Volume hdisk7 has 800 MB of paging space,
while Physical Volume hdisk0 has only 256 MB
- Paging space hd6 on volume group rootvg is fragmented
I/O BALANCE
- Phys. volume hdisk10 is not busy
volume hdisk10, mean util. = 0.00 %
- Phys. volume hdisk6 is not busy
volume hdisk6, mean util. = 0.00 %
FILE SYSTEMS
- File system hd2 (/usr) is nearly full at 92 %
VIRTUAL MEMORY
- VMM has identified 9 bad frames
you can set the severity level depending on what you want to monitor.
Severity 1 Problems
 Journaled File System (JFS) becomes unavailable
 JFS nearly full
 Physical volume not allocated to a volume group
 All paging spaces defined on one physical volume
 System appears to have too little memory for current workload
 Page space nearly full
 Possible problems in the settings of load control parameters
 VMM-detected bad memory frames
 Any host in .nodes becomes unreachable
Severity 2 Problems
 Imbalance in the I/O configuration (for example, disks per adapter)
 Imbalance in allocation of paging space on physical volumes with paging space
 Fragmentation of a paging space in a volume group
 Significant imbalance in measured I/O load to physical volumes
 New process is identified as a heavy memory or CPU consumer
 A file in .files exhibits systematic growth (or decline) in size
 A file system or page space exhibits systematic growth (or decline) in space utilization
 A host in .nodes exhibits degradation in ping delays or packet loss percentage
 A getty process consumes too much CPU time
 A process with high CPU or memory consumption exhibits systematic growth (or decline) in resource use
 maxuproc indicated as being possibly too low for a particular user ID
 A WORKLOAD TRACKING indicator shows an upward trend
Severity 3 Messages
Severity 3 messages provide additional detail about problems identified at severity levels 1 and 2. This includes the data-collection characteristics, such as number of samples, for severity 1 and 2 messages.
There are threshold files which you can set up for monitoring specific tasks and if they are reached , allow you to send an email or an alert .
The tool is o.k to start off getting an overview of system , but i found that you can write your own , and tailor it according to your own needs.
HTH a little
laters
- Status