Salvaging IT hardware after water or smoke damage means stabilizing the room, documenting it for the insurer, and triaging equipment by component class before corrosion makes it unrecoverable. Recovery plans describe failing over to another site but rarely the equipment left behind, which is both a significant asset and a source of ongoing liability.
The decisions about it are time-sensitive in a way that is easy to miss during an event where everyone's attention is on restoring service. Corrosion from smoke residue and from water contamination is progressive. Hardware that could have been recovered in the first two days becomes unrecoverable in the first two weeks, and the difference is what happens before anyone walks into the room.
How Do Water and Smoke Damage Hardware Differently?
Water damage is mostly about what the water carried and what it left behind. Clean water that reaches a powered-down device and is removed quickly often does very little. The same water on an energized system causes immediate short-circuit damage, and the distinction between the two is whether power was cut before or after the water arrived.
Contamination determines the rest. Water from a sprinkler discharge is relatively clean. Water from a burst chilled water line carries glycol and corrosion inhibitors. Groundwater and flood water carry silt, sewage, and dissolved salts, and salt contamination is the worst case: it is hygroscopic, it keeps drawing moisture, and corrosion continues long after the equipment appears dry.
Smoke damage is chemical. Combustion products from burning PVC cable insulation include hydrogen chloride, which combines with ambient humidity to form hydrochloric acid on every exposed surface. This attacks connectors, contacts, and circuit traces continuously, and it does so on equipment that was never touched by fire or water. A room adjacent to a fire can contain hardware that looks untouched and is actively corroding.
Soot is also conductive and abrasive. It penetrates enclosures through every fan intake, and the hardware that was running during the event drew it directly across the boards.
What Should Happen in the First 48 Hours?
Stabilize the environment: cut power, dry the air, and document. That matters more than any individual recovery action.
Power stays off until equipment has been evaluated. Energizing wet hardware to see whether it works is the single most damaging thing that gets done during these events, and it converts recoverable equipment into scrap.
Humidity control is the intervention with the highest return. Corrosion is a function of moisture, which is why ASHRAE TC 9.9 guidance on IT equipment environments pairs humidity limits with contamination control, and getting relative humidity down quickly slows every damage mechanism at once. Commercial restoration firms bring desiccant dehumidification for exactly this, and the response time on that call is the variable that most determines what survives.
For smoke events, the corrosive residue should be neutralized rather than allowed to sit. Restoration specialists have established processes for this, and they are effective when applied early and much less so later.
Document everything before anything moves. Photographs of the room, of equipment in place, of water lines on walls and racks, and an inventory of what was where. This record serves the insurance claim, and the insurance claim is frequently a larger number than the salvage.
Notify the insurer immediately. Most policies require prompt notice, many require the insurer's participation in salvage decisions, and disposing of damaged equipment before an adjuster has seen it can compromise a claim.
Which Components Are Worth Recovering?
Drives and storage media first, then powered-off servers and network gear exposed only to clean water; power infrastructure and contaminated cabling are usually replaced. Recovery likelihood varies enormously across a rack, and triage should reflect that rather than treating the rack as a unit.
Drives and storage media are the priority, and they are a separate question from whether the array works. A drive that has been submerged should not be powered on. Professional data recovery services handle contaminated media routinely and their success rates are high when the media has not been dried out with contamination still on the platters. The data is usually worth more than the hardware, and the handling that preserves the data is not the handling that preserves the drive.
Servers and network equipment that were powered off, exposed to clean water, and dried promptly have a reasonable chance. Anything energized during the exposure should be presumed damaged. Anything exposed to salt or sewage should be presumed damaged regardless.
Power infrastructure — UPS units, PDUs, transfer switches — should not be returned to service on a visual inspection. The failure mode is delayed and the consequence is another outage or a fire. Manufacturer evaluation or replacement is the defensible path.
Cabling is cheap relative to the labor of tracing an intermittent fault six months later. Contaminated copper and the connectors on it are usually replaced rather than assessed.
Recovered Is Not the Same as Reliable
Equipment that passes testing after a water or smoke event carries elevated long-term failure risk. Corrosion continues at a microscopic level on contacts and traces that look clean, and failures appear months later as intermittent faults that are difficult to diagnose because nobody connects them to an incident from the previous year.
Manufacturer warranties typically exclude damage of this kind, and equipment restored by a third party is generally out of support afterward. That is a real cost to weigh against the salvage value.
The reasonable position is to use recovered equipment where its failure is tolerable. Lab environments, development systems, non-critical workloads, and cold spares are all sensible destinations. Returning it to production for a tier-one service saves capital and accepts a risk that is hard to quantify and easy to forget about.
Whatever is recovered should be tracked as recovered (tagged, recorded, and visible in the asset inventory) so that the elevated failure rate two years later is interpretable rather than mysterious. And whatever is scrapped still contains data-bearing media, which means the disposal follows the same NIST SP 800-88 sanitization (performed by a licensed contractor) or destruction, and the same certificate process, as any other decommissioning, wet or not.