moorthy_kulumani
asked on
Capture Router/switches peformance parameters
We are planning to have a simple script to login to the routers and switches every hour and capture the CPU/memory other performance management parameters by using repective comnnads.
Currently we are using a very famous tools for that, But we would like to discontinue that and out org is not favour of any free tools like MRTG etc, we just want to have a perl script kind of stuff which can login to the switches and get the performance parameter and store in CSV…Later we can pull the values from CSV and manipulate the data the way we want….
Appreciate your help…
Currently we are using a very famous tools for that, But we would like to discontinue that and out org is not favour of any free tools like MRTG etc, we just want to have a perl script kind of stuff which can login to the switches and get the performance parameter and store in CSV…Later we can pull the values from CSV and manipulate the data the way we want….
Appreciate your help…
ASKER
Basically the resistance is due to the support model is not there for Freeware ...That is the only issue....
I will try to use TerraTerm, I have done few stuff with MRTG & Nagios in my previous org may be i will try to propose the same here....
I will try to use TerraTerm, I have done few stuff with MRTG & Nagios in my previous org may be i will try to propose the same here....
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Looks like EE doesn't like the file extension. Rename the file from .txt to .ttl to get terraterm scripting to work properly.
BackupConfigs.txt
BackupConfigs.txt
The problem with only querying every hour is that you will miss transient spikes in usage This might not help with diagnosis or analysis. Admitedly, there is always a trade off between frequency and accuracy, but hourly does seem to be a bit too infrequent for proper analysis of spikes. Also, with the the stats reported, these are the stats at a point in time.
PRTG/MRTG are a LOT more powerful and have some very good inbuilt reports. You can also export the data for manipulation using other tools. I would strongly recommend their use for network monitoring. Is the reluctance due to using SNMP? An understanding of the reason why not to use MRTG/PRTG or similar would help with advice.
I'm not sure how you would do this with PERL.