Premise Keeps Tabs on IBM i Cloud with Custom Dashboard
November 6, 2012 Alex Woodie
The folks at Premise, a managed service provider (MSP) based in Clearwater, Florida, have built a nifty Web dashboard that tells them all they need to know about the IBM i environments they manage for clients. From ASP usage and MIMIX replication status to rack configs and cache battery status, the dashboard gives Premise engineers and customers instant answers to many of the most commonly asked questions. What’s more, the dashboard may soon be available for any IBM i shop to purchase.
Premise is an IBM Premier Business Partner with expertise in hosting IBM i, xSeries, storage, and MIMIX high availability environments for midrange clients across North America. The company currently is managing nearly 80 IBM i images, which exist as partitions on Premise’s IBM i cloud environment; as servers installed in Premise data centers in Tampa, Florida, and Atlanta, Georgia, as part of co-location agreements; and as servers installed at customers’ locations, which Premise manages remotely.
Keeping track of client’s production and backup systems is obviously an important task for any MSP, which is why most employ system monitoring and management tools to automate the task, alleviate eye strain, and cut down on errors. However, the folks at Premise couldn’t find a tool that suited their current and future needs, so they decided to build their own.
The result is the Premise Dashboard. The software, which doesn’t have an official name yet, includes an IBM i agent and a Web browser-based dashboard, which provides several views of different aspects of customers’ IBM i environments.
The main Premise view allows engineers to quickly gauge the general health of their clients’ systems by listing the values of key metrics, like auxiliary storage pools (ASP) usage, QSYSMSG critical errors, error log status, Work with Disk Status (WRKDSKSTS), Work with Problems (WRKPRB) status, and disk cache battery status. If the status is critical, the field gets a red background, while borderline readings get yellow backgrounds.
Other columns on the main screen list general information about the systems, such as serial number, server model number, processor group, OS release level, cumulative PTF level, and even console type and keylock position. These aren’t typically critical pieces of information, but an administrator will appreciate knowing there’s a central place where he can find it.
Administrators will also appreciate the wealth of reports available from the dashboard. Right-clicking on an icon enables admins to drill down and pull up about 20 reports, ranging from firmware status and IPL attributes to rack config and installed software information. There is even a historical performance report available with just a few clicks of the mouse. Reports are displayed instantly in HTML or PDF formats.
Another Premise dashboard view displays only MIMIX-related information, including datagroup status, datagroup errors, audit errors, notify errors, object delay, percentage of ASP used, and other critical information. The screen can be set up to display all the information, or just the errors.
The folks behind the dashboard say they couldn’t find any shrink-wrapped monitoring products that could do all this, and be flexible enough to add support for monitoring other environments. “There are some out there, such as Tivoli, but it doesn’t cover iSeries and doesn’t do non-IBM SANS,” says Premise engineer Ron Venzin.
“We wanted the dashboard to be more than just MIMIX,” Venzin says. “As you can see, they get full system monitoring, plus the MIMIX piece. We’re also adding views for Intel servers and SANs, which we expect to be ready by the end of the year. And we’ll be adding views for System p as well as printers and eventually network equipment.”
The dashboard also provides detailed information about the PTFs loaded onto the system. “They can get it all in one place,” says senior system analyst Scott Crill of the dashboard’s PTF functionality. “They can see where they are, they can see where IBM wants them to be. And they have those URLs either to IBM or IT Jungle.”
(Begin self-congratulatory editor’s note: Crill here is talking about the System i PTF Guide, which provides a weekly update on all IBM i PTFs. A link to the System i PTF Guide is featured in the Premise dashboard, right next to links for IBM’s PTF cover letters. “It’s very simple. We point at you guys, and you do your magic in collecting the information and presenting it,” Crill says. “You guys do a fantastic job of collecting that [PTF] information, because trying to hit IBM’s website at times is challenging.” End self-congratulatory editor’s note.)
The dashboard has generated a lot of interest among Premise prospects and customers, who can use the dashboards to view their own environments, or just rest easier knowing Premise is keeping a close eye on their boxes. “All of a sudden they realize somebody is looking at their system,” Venzin says. “Someone can actually see real time everything about their system, versus knowing it’s running and there’s data being pushed one way or the other.”
The information in the reports is easy enough to get without the dashboard, provided you have a live session to the IBM i environment in question. What’s unique about the Premise dashboard is it provides quick access to a wealth of information from a Web browser. “I go into meetings now and people ask me a question like ‘Do I have this card or this feature,” or ‘What’s my disk space?'” Crill says. “I don’t have to sign on to these boxes. I have all these spool files, these common spool files, and it’s all there, just waiting to be printed or attached to an email.”
The visibility into IBM i disk issues has helped the dashboard pay for itself, both by helping Premise sell additional DASD to customers who were running dangerously high on ASPs, and avoiding system crashes.
“We recently had a customer who, at 3 a.m. had a disk drive error,” Venzin says. “We caught it on the dashboard, and one of our guys sent a note to customer. It [the drive] had not called home yet. The customer called IBM, and by the time the CE was coming out, a second drive started failing. The CE was literally on the doorstep, whereas, if he would have waited for the IBM call home, he most likely would have lost the system.”
If all goes as planned, Premise will add support for the additional monitors, and spruce up the user interface so that it can be displayed on mobile devices including iPads. At that point, the company may choose to sell licenses for the dashboard to the general population.