The Help Desk implementation continues this week amid ice storms, power outages, and windstorms. Jay Rollins and his team look at Nimsoft and IPMonitor solutions.
This week had a few challenges. Aside from ice storms, power outages, and wind storms, we had more work to do. The focus for this week was reviewing some of the basic functionality of the help desk system, adding a couple of other capabilities, and reviewing monitoring technologies.
Our data center service provider covers all of our equipment today with IPMonitor. The challenge with the monitoring tool is that it doesn't provide indepth proactive monitoring of our network infrastructure. It can look at all the SNMP traps on the hardware (CPU, memory, etc.) and do ICMP tests but there is nothing at the TCP/IP level that can help us see and proactively address network latency, hung processes and user experience monitoring and alerting.
We then went to our network services provider. They just launched a new service built around Nimsoft's Indicative suite. This tool is sweet. Expensive, but sweet. It has all of the proactive monitoring we were looking for, plus all of the ICMP and SNMP tools. How the system handles alerts and drilldowns, issues, tests, changes, etc., is very intuitive. Also, server management can leverage various other tools such as Microsoft Perfmon, and generate a real-time rendering of perfmon logs in the console for troubleshooting and setting alerts off of perfmon thresholds.
The OpsManager tool from ManageEngine uses the Windows WMI services that SMS uses. Unfortunately, the bugginess of WMI is pretty well documented so we took a pass on OpsManager.
With our monitoring options laid out, we started down the business case/funding part of the process. We're in favor of the Nimsoft solution for network and IPMonitor for servers for now. Why not use just one? One reason is price and the second is being able to absorb the the girth of what the Nimsoft product offers. If the network management piece gets integrated without issue and the various alerts are refined, we may start looking at the tool for other services.
One of the challenges we ran into this week was building the process for Problem Management. The template works fine. It's actually building the process for transition from incident or request to a problem that's a challenge. We can associate incidents with a problem, but we can't seem to find a way to convert an incident into a problem. Still working through this process.
Next, we ran into an issue regarding sites or locations. There is an area where you can create sites and group those sites into regions. However, when I ran a report on incidents by site, all of the tickets had "Not Assigned." When we looked into it, we found that we couldn't make the site selection mandatory. This has to do with how the system handles the "requester" object. The requester is imported from your active directory and it takes "Site" from the "Office" field in AD. You have the option within the ticket screen to select a site, but you can't make it a mandatory field. So, the plan is to update our AD "Site" information with all of the correct information and then re-import the AD users into the system.
We added an additional status to the interface as well. The existing status entries included Open, Closed, or OnHold. The OnHold status stopped the clock while the technican was working on the ticket. This was like waiting for a response from the user or the results of a test. Unfortunately, it was also being used to stop the clock while we ordered parts or new equipment. So we created an "Order Parts/Equipment" status for those so we can differentiate between them.
Next, we looked at the purchasing workflow tool. Basically, technicians can create purchase orders and submit them, and they get routed to an approving authority. Unfortunately, that's about all you can do. There's no capability for cascading approvals. We wanted to set dollar thresholds for approvals. Anything under $1,000 could be approved by IT management. Everything over needed to be approved by VP of IT. If it wasn't over $1,000, the VP wouldn't need to deal with them. But, no can do. ManageEngine support told us that it was a feature that has been requested by many of their customers and it's being planned for a future release, but no information on how soon we would see that release.
That's about it for this week.