Server maintenance and annual contracts

Q: Our servers are still under warranty. Do we need maintenance?

The original warranty is 'free replacement of broken hardware', but regardless of inspection, backup, security, response speed (original T+1 or even slower), not to mention RAID degradation this 'not bad but almost broken' state. Our maintenance and original warranty do not conflict, many customers are 'original warranty hardware + we guarantee operation' combination.

Q: What's the difference between you and IT outsourcing?

Maintenance focuses on 'equipment health'. IT outsourcing also includes daily desktop support (employee computer problems, software installation, network adjustment). Both can be signed separately or packaged. It is usually more cost-effective for enterprises with less than 50 people to package. See the IT outsourcing operation and maintenance page for details.

We provide server maintenance by platform/year for SMEs: regular inspection + fault response + spare parts first, turning "fire protection" into "fire prevention."

calculate an account What is included in the maintenance contract (clearly stated)What exactly does a standard inspection check (real case price reference FAQ

Calculate an account: maintenance for one year vs downtime for one day

Many companies feel that repair fees are "white flowers" until something happens. Calculate the actual account for you:

the result in downtime: A server running ERP/financial system downtime, 20 people of the company per hour output value is calculated by RMB 50, stop a day is RMB 8000; if it is e-commerce/order receiving system, the loss is calculated according to the order, far more than this number.
Single maintenance cost: Over-protection server motherboard failure, board-level maintenance RMB 800–2500;RAID crash data recovery RMB 3000–15000. A major failure once a year, more than maintenance premium.
hidden risks: No regular inspection, yellow light on hard drive, RAID degradation, UPS battery failure These problems are "silent", and often have been superimposed into disasters by the time they are discovered. At least half of the failures we take over can be stopped in advance inspection.

The essence of annual maintenance:With a fixed small amount of money, lock in uncertain large losses.

What is included in the maintenance contract (clear price, no word games)

Our maintenance is divided into three levels, and all what we dos are written into the annex of the contract, to what extent it is clear at a glance:

Basic file (inspection type): On-site inspection once a quarter (hardware status, RAID health, log, temperature, UPS battery)+ inspection report + telephone/remote support + fault response within 8 hours. Suitable for companies with few equipment and part-time IT.
Standard (responsive): Monthly inspection + remote monitoring access + failure 4 hours response, arrival the next day + common spare parts library advance payment (hard drive/power supply/fan, repair and settlement)+ data backup scheme configuration and quarterly recovery drill. Suitable for 10-50 scale enterprises The main choice.
Full file protection (escrow type)All standard files + 7×24 telephone response + key spare parts on-site standby + system-level maintenance (patches, security reinforcement)+ annual IT asset inventory and budget recommendations. Suitable for enterprises with more than 3 servers and zero downtime tolerance.

The cost of replacing damaged parts after the hardware itself has been overinsured (according to the actual occurrence, we only finish work) is not included in the third grade. Many companies are vague about this, and we write it in the contract.

What exactly does a standard inspection check (19 items list)

Inspection is not "look at the machine is still on the light". Our inspection list has 19 items, each time taking photos to keep files:

Server hardware: front panel alarm light, hard drive indicator light block by block confirmation (yellow light = warning, red light = fault)
RAID card status: enter the management interface to check the Media Error count, Predictive Failure flag and rebuild records of each disk
System logs: hardware events (WHEA/MCE), abnormal restart logs, disk bad sectors alerts
Temperature and fan: inlet air temperature, fan speed, room air conditioning settings
UPS: battery health (internal resistance/capacity test), load rate, last power failure record
Backup verification:A random backup file is actually restored.--backup only, no verification equals no backup
Disk Space and Growth Trends: Predicting How Soon It Will Be Full and Scaling Up Earlier
Security items: system patch level, open port, weak password spot check, antivirus software status
Network equipment: switch port error count, optical module optical failure, wireless AP online rate
Physical environment: cable label integrity, dust, rat marks, water leakage marks

Written report after inspection: score the status quo, grade the problems found according to red, yellow and green, and suggest the time limit for treatment. Red items shall be treated at the same time, and yellow items shall be included in the plan.

⚠️ Before signing the maintenance contract, recognize two items:First, the response time limit and the arrival time limit must be written into the contract ("as soon as possible response" is equal to not written); second, ask clearly "who bears the replacement cost", some companies quote low maintenance, three times the price when replacing parts. These two clauses of our contract are clearly stated.

real case

This is how we help our clients solve problems.

Patrol found RAID5 degraded for 47 days, no one knew, change disk to keep 8TB data

In May 2026, we took over the maintenance of a logistics company in Dongchong (3 servers). The first inspection found that a hard drive in RAID5 had been offline for 47 days-the system had been streaking in a single disk redundancy degradation state, and during this period, another bad disk would be 8TB of data. The reason: the hard drive started reporting Media Error three months ago, but no one read the log; the original "maintenance company" had not been on-site for half a year. We adjusted the same specification enterprise disk (Seagate Exos 8TB) to complete the reconstruction on the same day. It took 11 hours to rebuild, and during the period we stared at it without breaking it. The so-called maintenance before this customer was introduced by acquaintances as "two thousand RMB a year", but in fact it was broken and called. Later, we renewed the standard file. RAID degradation is not handled, which is equivalent to driving blindfolded on the edge of the cliff.

There are 12 restaurants in the restaurant chain, and all UPS batteries fail once.

In June 2026, a restaurant chain in Panyu (12 stores, each with a POS terminal server + network rack) signed the basic file maintenance. The first round of inspection ran to 12 stores, and found FAQs: UPS batteries in stores failed after an average of 4 years of use (power off), 3 store switches were placed next door to the kitchen, and 2 POS terminal servers had SMART alarms. Batch processing: 12 UPS battery replacement (package price, single unit is 40% lower than retail price), problem switch cleaning shift, early warning hard drive planned replacement (pick low business peak store by store, zero data loss). This customer has not had a checkout shutdown throughout the year-for restaurants, the turnover and reputation lost by one hour of checkout is much higher than the maintenance premium.

price reference

Test first, quote later, repair later

Item	Reference Price	Notes
Basic file (inspection type)	RMB 2400/set/year	Quarterly inspection +8 hours response
Standard (responsive)	RMB 4800/set/year	Monthly inspection +4 hours response + spare parts first
Full file protection (escrow type)	Quote on request	7×24 response + on-site standby
Single patrol (not contracted)	RMB 300/set	Includes 19 exams + reports
UPS Battery Test Replacement	by model	Internal resistance test
Data Recovery Walkthrough	RMB 500/time	Real restore verifies backup availability

* The above is the reference price, the specific price after inspection/survey shall prevail; batch/enterprise customers can negotiate discounts. Final pricing is subject to the actual on-site inspection.

FAQ

About server maintenance and annual contracts, people often ask

Our servers are still under warranty. Do we need maintenance?

The original warranty is "free replacement of broken hardware", but regardless of inspection, backup, security, response speed (original T+1 or even slower), not to mention RAID degradation this "not bad but almost broken" state. Our maintenance and original warranty do not conflict, many customers are "original warranty hardware + we guarantee operation" combination.

Are you responsible for losing data during maintenance?

It is clearly stated in the contract: We are responsible for performing backups and recovery drills at the agreed frequency (records are available for inspection). If the backup obligations are fulfilled in accordance with the contract, the data will still be lost due to the complete destruction of the hardware and new data generated after the backup point, which belongs to the backup window. Period losses, which are not covered under any maintenance contract. What we can do is to keep the backup window to the shortest possible length (such as hourly increments). We will explain this clearly before signing the contract and will not play word games.

Is it worth signing up for just two servers?

Worth it, and the basic file is designed for this scale. 2 servers RMB 4800 a year, spread to RMB 13 a day, for a quarterly comprehensive physical examination + failure 8 hours response. Many big failures are caused by small problems.

What's the difference between you and IT outsourcing?

Maintenance focuses on "equipment health". IT outsourcing also includes daily desktop support (employee computer problems, software installation, network adjustment). Both can be signed separately or packaged. It is usually more cost-effective for enterprises with less than 50 people to package. See the IT outsourcing operation and maintenance page for details.