I am attempting to use Visual Cron to monitor our Network Processes and Services.

There are literally dozens of way to configure this but I have some issues with behavior.

If you set up a trigger that monitors a service status change, it will not trigger if the server itself drops off the network. This appears like a mistake to me.

Realistically, I need to have a good setup that will send an email if a process or service goes down (this should include if the server itself reboots or drops off the network)

Does anyone have experience with this or has anyone done this before. I have spent many hours on this and am looking to learn from other's mistakes.

Help please?

Hi,

Question: Is the trigger of the service check going to be inactive when the server is down? If so, you could add this trigger to the job also?

Regards
Erik

---

I'm not sure exactly what you are asking me. The way I tried to set this up was with a service monitor trigger and an email task.

Ideally when the service changes status, it would activate the email task which would send me a notification that there has been a status change.

However, if the service does not change status, but the server itself reboots or loses network connection, the job does not trigger. I believe that the trigger does go inactive when this happens. That is not ideal.

The job should let me know if the service changes status and if the server goes offline. (obviously the service is down if the server reboots) Since this job as I have set it up does not perform as I'd like when configured this way (a target server reboot stymies the job's function) I am looking additional ways to configure this that will meet the need.

Hi,

I thought of the idea of an extra job checking the trigger going inactive. If so, then you can also send an email that the server is down and that you have to re-enable the service trigger.

Regards
Erik

---

We do this for several services on different servers. Instead of using a service trigger, we pole the service. That way, if the server itself is down or unreachable the job will fail and we will get notified of the job failure. We have jobs that run every 5 or 15 minutes depending on criticality of the services they are monitoring. For each service, there are two tasks, each with a condition that allows the task to run if the service is stopped.

The first task sends an email alerting us to the problem.
The second task restarts the service.

Works like a charm. Let me know if you need any further details.

DKOCUR: So if you do not use a service trigger, then how do you poll the service? That is where I am having the stumbling block. I've got the steps I want to take after I've determined there is an issue all laid out, so my main issue here is accurate detection.


ERIK C: Yeah, that could work. I'll have to give that a shot. The fact that the trigger goes inactive and would require human intervention to re-enable it does make this a very robust and automatic monitoring system, however.

I use an interval time trigger. See attached screen shot.

Edited to Add: I've also exported a sample job. See the attached zip file.

We do this for several services on different servers. Instead of using a service trigger, we pole the service. That way, if the server itself is down or unreachable the job will fail and we will get notified of the job failure. We have jobs that run every 5 or 15 minutes depending on criticality of the services they are monitoring. For each service, there are two tasks, each with a condition that allows the task to run if the service is stopped.

The first task sends an email alerting us to the problem.
The second task restarts the service.

Works like a charm. Let me know if you need any further details.

How do you poll services? I can only see options to Start, Stop and Restart Services.

Thanks

Hi,

Sorry. I forgot how obtuse a solution this was. I use a condition of type "Service" to determine if the service is stopped. Download the job (and conditions) that I attached to my last post, and import it into a test copy of VisualCron and I think it'll explain a lot.

I'm headed out for the day, so if you have any further questions I won't be able to answer them until tomorrow.

HTH

Yes, I realize this is not really the best solution for a network monitor, but we are trying to avoid purchasing one of the tools that is designed to do this. Thank you very much DKOCUR for providing this information.

It looks like we will NOT be using Visual Cron for this solution. Our application group is going to write a program to handle this task. Thanks for the information everyone.