01-06-2014 09:38 AM
I am monitoring application availability & performance using BSM,
BSM collects data from different locations,
I 've two places used by BSM to collect data [ UK & FR ].
if transaction fail in france & succeed in UK .. what does that mean ?
Thanks in advance !!
01-06-2014 11:02 AM
I'm assuming that you use BPM (Business Process Monitor) probes.
If you have two probes and only one of them fail to get data from a specific transaction, probably you have a connectivity issue between this probe and the target of transaction.
If you are using a web transaction I suggest you enable "Generate snapshot on error" in VuGen to clarify what is happening when the transaction fails.
Hope this helps,
01-06-2014 03:17 PM
Thank you pafriere ,
But I've a question, Does that mean that people in France won't be able to access app. when there is issue in France data collector ?!!!!
I heard something like this and i just need to validate, as i believe that there is no relation between data collectors and routes users use to reach BSM.
Could you please advice on this ?
01-07-2014 05:31 AM
I assume that you are using SaaS pops (public pops) . if this is your pops, you need to login to BPM machine to check that you can browse to the site. usually if your BPM health is good (e.g. you can browse to other sites duirng the time that you had a problem) it means that you have problem from this location, and this ISP that host this pop. (which ISP are you using?)
If you are using SaaS / PAW pops, it means that your site is not accesible from France , and the users that are browsing using th esame internet provider may expirience the same problem.
you next step is to check it with the the ISP - is this dns problem (resolving issues)? general timeout problem to specific url? are you using proxy? many parameters important to you debug process.
01-07-2014 05:49 AM
if the transaction is failing from one location and successfully being executed from another location in most cases if should mean that the target application is working as expected from a data center point of view. means the problem obviously is related to the faulty location.
now we still don't know whether or not this is related to a general failure on this site (network problem for example) or if it is only a problem for the local BPM machine. for this and for many other reasons i would strongly recommend to deploy at least two independent BPM instances per site. this will drastically improve the quality of you data and will help you to understand what the cause of such misleading results is. setting up a webtrace can be an additional indicator to help you understand the problem. webtrace will at least give you some information if the connection to the web server can be made (network access, dns resolution, some security restrictions). manually executing the script from the location in question can be another way to narrow down the search for the cause of the problem.
hope this helps a bit
01-07-2014 09:06 AM
You can check probe and even script health in BPM (probe) interface. In there you can see status and execute manually the script and generate a report of execution. To do it access:
In BPM Admin doc attached you can find how to proceed.
01-08-2014 11:07 PM
you can double check by accessing the BPM machine in your faulty site and the run the URL manually to check it's working or not and to indicate as well , if this is a common issue or is related to your BPM application but for all reason you have to go in a manually troubleshooting routine.