Welcome to the Application Performance Management Blog, where you can read the perspectives from APM experts. This Blog provides insights into the Application Performance Management solution, as well as technical details about specific IBM products.
One of the most important activities to perform when you are monitoring your IT infrastructure is to identify specifically which resources you are taking under control, as well as which situations you are running and where you are running them. Visibility is the goal. You can accomplish your goal in serveral ways. We suggest you use a couple of IBM Tivoli Monitoring tools: · IBM Tivoli Situation Reporter · ITMSuper A new tool, IBM Tivoli Situation reporter, helps you to identify what you are monitoring in your... [More]
A lot of customers are frequently asking for best practices on how to monitor WebSphere Server environments. On the IBM Service Management Library a white paper is available, giving a bunch of sample situations for ITCAM for AD V7.1. This paper will be updated shortly. Any additional suggestions may be posted within the comment section of the wiki entry under ITM Situation Package .
A poorly written situation can have large impact on CPU usage of IBM Tivoli Monitoring 6.2.x OS Agents main processes (kuxagent, klzagent and kntcma.exe). Using wildcard characters (*) and MISSING function in situations is one of the most common source of high impact on CPU usage related to situations evaluation. There are two main suggestions to follow while rewriting situations in order to reduce ITM OS Agents CPU usage: situation with this simple pattern 'LIKE "*/process"' can be re-written as 'SCAN "/process"' put the... [More]
Imagine you are busy perusing your e-mail when a critical alert occurs on a system within your environment. Wouldn't it be nice if you could receive a message containing a link that would open the Tivoli Enterprise Portal and take you right to the relevant area so you could investigate? This post shows how use Tivoli Monitoring facilities to automatically send a message to do just that. I assume you are familiar with the Tivoli Enterprise Portal Situation and Policy editors and will not go into great detail about them. Here are the steps: ... [More]
On the second installment of the SPUFI series, we look at our first "use case". This blog will show you how to extract key information via the command line from your TEMS relating to events/situations that have fired in your ITM infrastructure. This can be used to health check, sanity check and recheck all key areas of event management. This is a very powerful way to obtain great information from your TEMS If you missed part one on how to use the tool here is a quick link: ---> ... [More]
On the third installment of the SPUFI series, we look at our second "use case". This blog will show you how to extract key information via the command line from your TEMS relating to which situations are started, stopped, deleted or in error in you ITM infrastructure. This is a very powerful way to obtain great information from your TEMS outside the GUI. If you missed part one on how to use the tool here is a quick link: ---> Lin k to blo g on e in... [More]
Always wanted to be able to pull all of the OS and ITM versions for every agent in your enterprise in one single action... On this installment of the SPUFI series, I will show you exactly how to do that. This is another powerful way to obtain great information from your TEMS outside the GUI. If you missed part one on how to use the tool here is a quick link: --- > Li n k t o bl o g o n e i n th e SP U FI s eri e s Overall Goal The goal is to be able... [More]
It is not the first time that I have been asked for some customers about how avoid to miss alarms when a remote TEMS has down and there are true situations in the agents connected to that TEMS . In this document I'm going to try to describe how ITM works to avoid that some important alarms may be missed and how it can be done. From ITM 6.2.2 the agents can work as autonomous when they lose communication with their TEMS. By default, the agents are configured for autonomous... [More]
It all starts with what you name your situations. If you do not have unique names you could potentially cause greater problems for yourself and your environment. Problem description: When creating the name of a situation the ITM6.3 FP02 manual says: ‘The situation name can contain any UTF-8 characters and can be up to 256 bytes long. The number of characters allowed varies depending on the characters used, because some UTF-8 characters require more than one byte.’ (Page 259) Please note when creating... [More]
The normal behavior we are used to see for the historical collections is to have like minimum a warehouse proxy agent installed in our environment and configure via TEP/“history configuration....” the attribute groups to be collected and the agents which have to collect them. Now we have other option, it is to use the “private history”, configuring the attributes to be collected in the agent by a .xml file located in the agent , so the agent starts collecting historical without using TEMS/WPA at all , it works in... [More]
Hello again folks, For this bitesize blog we've got a useful tip for ITM situations, specifically those using the "MISSING" function, which can be a bit thorny. Specifically, we're looking into an issue where a situation using the "MISSING" function does not fire as expected when it's been set up to monitor two items with similar names. The explanation behind this is that ITM framework uses the SQL1 language, which is broadly similar to standard SQL, but not exactly the same which is what... [More]
Hello all, So today we're looking at a slightly puzzling PMR. I had a customer who didn't have a Situation Event Console icon in their TEP client. We tried changing permissions for the users, logging in as different users, trying the desktop client, web client and JWS client, all to no avail. We confirmed that the icons for it existed on the system and all the jars were present but we simply couldn't find this thing anywhere. Frankly finding Carmen... [More]
IBM Tivoli Composite Application Manager for Transactions ~ How to do stuff ~ By Kevin Baldwin - How to..... Generate alerts for RFT subtransactions on RRT / T6 agents Unlike Rational Performance Tester (RPT) playbacks, it is not possible to define response time thresholds for subtransactions in RFT profiles. This blog describes how you can use situations to raise alerts when thresholds are exceeded at the subtransaction level in RFT playbacks. Alerts raised by the situations will be displayed in the TEP in the Enterprise Situation Event... [More]
Here below is an example of possible setting to detect if a program/process like "kt6agent.exe" is not running If the ITCAM for Transactions 'Robotic Response Time' agent is started for example, (as a service (default), or as a process using 'kt6agent.exe test' from a command window) the "kt6agent" process should be running all the time, and playback robotically CLIs, RPT, RFT, LoadRunner, etc... test scripts. A possible way to detect if it's no longer running, and to produce an alert... [More]