The real test is only beginning when the WiFi network system of accommodations is on-line. The construction process has been watched by the manufacturer, and the school itself is now the day after the day. We talk to the Internet center teacher most about what we should look at not how, but usually when we see something. The billing system does not seem like a normal network device, it has money and student experience, it causes problems, it's financials and students' meetings.
First-line: reconciliation of payments and bills
This is one thing that has to be seen every day. Pulling a bill for the current and billing of payments on that day, compared with the exact amount received and recorded, and whether the difference is acceptable. We saw the reconciliations being done half a month later, and we found that one day the large area of the return was lost, that the accounts were all artificially filled up, and that the finances were almost crazy. The reconciliations are not complicated, so the system is daily reports, and the wiring points open the differences, but they have to be done daily.
Second stakeout: buildings are online and traffic is variable
Daily carriers are used to the online numbers and flow curves of buildings. A building suddenly crashes, possibly with an access controller hanging or connected; a building has an unusually high traffic at some point, perhaps someone is brushing down or having its equipment stung. We suggest several thresholds for warning, such as single-building lines falling below 30% per person average, or single-facilities 10 times more than per person, and automatic alarms. The skyscanning curves are much faster than waiting for students to complain.
Third-line: ratio of unpaid and disconnected nets
The default rates rise suddenly, often not because students have no money, but rather because the puncture channels are wrong, such as paying for a faulty interface and having a self-help page open. We usually let carriers look at the difference between the amount of the payment due and the number of breakouts every day, and check the system first if there is an abnormally large number of breakout nets.
Fourth stake: Certification success rate
Certification is the entrance, and the entry card is a total collapse. We suggest that the success rate be considered as a core health indicator, below a threshold. The success rate may be lost because there are insufficient resources for certification services, perhaps slow databases or network shaking. We have to keep this indicator as tight as possible in the first month, and then stabilize every day. The first word of students not even online is always broken, but 80% of them are certified cards, so that when you look at it, you can see where you're going.
Five-point: Equipment and account number is abnormal
The system should run an unusual log-in scan every day: the same account number is short, multiple building logs are bad, single equipment is in a high volume of sudden traffic and long hibernation accounts are out. We've seen those who have pulled through the whole net with this and we've found that graduation numbers are still being billed. Not necessarily bad, but this screen, which turns into a hole. Half of daily transport is reflected in this scan.
Leave an enforceable contingency plan
Finally, staring at the real problem will come up one day, so there's a plan. The central database is on how to downgrade to the building, pay for interruptions in how to get back to the building, and the certification service is full of how to limit the core of the fluid security system. It's really not a problem. We suggest that we do an exercise every semester with a real failure scenario, so don't let the pre-case lie in the folder.
Make a regular volume disk
The daily watch is now, but the capacity increases with the number of people and equipment. We suggest that each term be re-recorded to see if a single copy of the package, flow, storage is fast approaching the design ceiling and plans for expansion ahead. Many schools wait until the system starts to start to build up their cards, so neither purchase nor deployment can catch up.
The knowledge base is going to sink.
The pits that were stepped on, the parameters adjusted and the malfunctions processed in the transportary are to be deposited into the school's own knowledge base, not only in the brain of a teacher. We have seen the new teachers move away from the system, with the same pit being blacked out, and then we step on it again. The knowledge base does not require more formal documents, but rather an updated document, which records phenomena, causes, treatment, prevention. It accumulates for several years, the most valuable operating asset of the school and the most tangible thing at hand.
Create feedback channels with students
The system is connected to the student’s experience, which is not enough to keep track of itself. We suggest that a portal for self-help pages be left open, with students having problems with Internet access, bills and one key submission. These feedback aggregates to fill invisible issues in the system’s monitoring, such as the area’s coverage of dead ends and the time-deeping experience. Students are the most frontline sensors, opening up the feeder channel, wide viewing the carrier, and many problems are recovered before complaints break out.