214: Chapter 214 The Last Record Before Losing Contact on the Far Side of the Moon

The black wall had stood behind the physical warning rack for another ninety-two days.

During these ninety-two days, the temperature difference between the inside and outside of the wall gradually stabilized, the noise floor of the sixteen maintenance optical paths completed three cycles, and the water level in the drainage ditch did not cross the red line Jiang Lin had left on the scale.

There was still a short section of positive pressure behind the wall.

Jiang Lin had the four high-precision inspection platforms retreat to one hundred and twenty meters away, leaving nearby only G-Explorer-C, a low-power engineering platform, and two passive sampling boxes.

After the pressure-equalizing pipe passed through the reserved hole in the wall, the first stream of gas entered the sampling box first, and the second stream was then directed to the isolation analysis unit.

The oxygen content was very low, the proportion of inert gases matched the records in the maintenance volume, no still-spreading corrosive components were detected, and no active biological signals were found.

In the spring of the twenty-third year of this round, the observation index area finally met the conditions for unsealing.

The panel in the center of the wall that had lost its recognition capability was separated from the surroundings by a cold-cutting frame, with the cuts avoiding the internal sealing layer, and the final section was slowly pried open by a mechanical wedge along the original maintenance seam.

The composite board stopped once when it had moved outward to a position sufficient for a probe rod to reach in.

Inside, a long-ago loosened grounding braided strap was snagged on the corner of the board.

G-Explorer-C extended its probe rod into the gap to support the braided strap, waiting for the engineering platform to cut off the failed connection end from the outside before the composite board continued to drop onto the load-bearing frame.

When the portable work light shone inside, it revealed only an arc-shaped maintenance corridor curving to both sides, as well as sixteen low storage doors lined up along the inner wall.

Every door bore the same mark.

[ OBSERVATION INDEX ／ PHYSICAL LEDGER ]

[ PASSIVE MEDIA ／ ONLINE EXECUTION PROHIBITED ]

Jiang Lin first walked a circle along the maintenance corridor.

The sixteen storage partitions shared the same circle of passive heat sinks and structural supports, but the data contents were not archived separately by year or system; instead, they were interspersed and dispersed across all partitions.

Total damage to the media behind any single door would only cause a portion of every time period to be missing simultaneously, rather than causing certain years to vanish entirely from the archives.

This design was very effective against fires, local collapses, and malicious overwriting.

To anyone who later attempted to read it, the attitude was not quite as friendly.

"Security was done very thoroughly, sealing away even the chance to save trouble."

Behind the first storage door, there were only one hundred and forty-four grayish-white media cartridges, jammed into mechanical slots in twelve layers.

The surfaces of the cartridges retained positioning marks, Physics serial numbers, and a transparent readout window for external optical paths to pass through; no processors were visible, nor were there any controllers capable of receiving commands.

The first round of reading was completed right within the original slots.

He first used a portable read head to acquire the surface directory, then had a passive probe sweep along each mechanical slot to verify that the media positions matched the Physics inventory list on the cabinet door.

The directory page recovered only three lines of complete content.

[ VERIFIED FAULT LEDGER ]

[ SIXTEEN PARTITIONS INTERSPERSED STORAGE ]

[ MASTER INDEX NOT IN LOCAL EXECUTION ENVIRONMENT ]

This was followed by a massive margin validation tree; the root of the tree was intact, but the leaf nodes were scattered across more than one thousand media cartridges.

To read out any high-level record among them, one had to first find its corresponding Physics pages from all sixteen partitions, restore their respective time stamps, and then prove that these pages belonged to the same verified snapshot before being sealed.

Jiang Lin left the portable read head in place and returned to the forward base along the newly laid tracks.

Half a month later, the first group of offline readout workstations was transported outside the wall.

Ordinary engineering platforms were responsible for handling, cleaning the casings, and replacing fixtures, while the only things that actually touched the transparent readout windows were six vibration-isolation workbenches, twelve low-power read heads, and storage units independent of the construction network.

All readout circuits had only a one-way channel from the media to the isolated storage, and the optical scanning sequence issued by the workstation was also restricted within the energy range specified by the civilization core maintenance standards.

More than four hundred machines could complete tens of kilometers of underground trunk lines, but they could not magically multiply the number of high-precision read heads.

The reading speed was slow, the media layers were numerous, and a few windows still bore tiny cracks left by thermal shock.

Jiang Lin locked down the expansion permissions of the construction network to the security level, retaining only maintenance and necessary reinforcement for the seventeen construction sites, while prioritizing materials and processing time slots for the read heads, vibration-isolation tables, and external positioning frames for damaged media.

A low-temperature workshop soon appeared outside the observation index area.

After a batch of media cartridges completed their surface scans, they were sent into the workshop by a narrow-gauge cart, returned to their original slots after reading, and both the mechanical locks and physical seals were reset in their original order.

During the first few months, what appeared most frequently on the terminals were still directories and validation blocks.

External observations, the Mars relay, the far side of the moon link, regional arrays, and civilization seed backups were divided into different systems, each using mission time, local time stamps, and internal device counters; even fault severity did not share the same set of definitions.

Arranging the file creation times together would only yield a seemingly neat false timeline.

Starting from the original counts and clock transfer records, Jiang Lin set the multi-source stability confirmation of FDSO-2076A as the common zero point, and then used several independently identifiable astronomical events within the Solar System to verify the long-term drift of each set of time stamps.

The number of the Seventh Canopy Station was fully expanded for the first time during this process.

[ SEVENTH SKYSCREEN NODE ]

[ Ground Phase Benchmark / Deep Space Link Timing Node / 07 ]

Its sixteen partitions, the twelve-degrees-north-by-west reference axis, and the facilities of Northeast Seventy-Three and the far side of the moon used the same set of engineering benchmarks.

Jiang Lin changed the dashed lines left in the mission pages for many years into solid lines.

That sixty-meter-tall tower that had stood on the wasteland for nearly a hundred years finally had a confirmed identity, but what it preserved was still only a ground timing mirror; the main archive address was not in the tower.

After the timing records were supplemented, the fault pages originally scattered across the sixteen partitions began to enter the same timeline.

The first to close was the external observation source.

Jiang Lin arranged the original particle counts of different devices according to heliocentric distance, retaining their respective observation epochs on the vertical axis, and a wide band inclined inward from the outside of the Solar System gradually appeared on the screen.

High-speed, small-particle-size components crossed the outer observation surface first, while the main body with higher density and a broader particle-size distribution fell behind; the projection direction of origin did not experience any shifts in the independent data sets that could change the conclusion.

The time difference of the same structure passing through different observation surfaces was consistent with the multi-source combined result in MIRROR-07.

The Dark Dust String hypothesis, pieced together in earlier years from PMCU-17 fragments, obtained for the first time here a complete observation chain capable of tracing back item by item to the original counts.

Devices that went out of service were archived separately according to direct damage, attitude loss of control, transmission failure, and secondary overload.

Some probes directly lost their payloads after local counts rose, some lost attitude control first, and some were still able to observe but could never find a sufficiently stable transmission window.

More faults came from the secondary burden after link reduction.

The remaining nodes took over timing, caching, and computation; the cooling and energy storage windows were repeatedly squeezed out, and small faults that could have been isolated subsequently entered the regional system.

Jiang Lin divided direct observations, device responses, and engineering inferences into three layers; whenever any causal chain lacked independent records, it stopped at the corresponding layer.

When the readout work entered its third year, the first cross-system snapshot finally lit up completely on the isolation terminal.

[ FDSO Multi-source Stability Confirmation / Common Zero Point ]

[ Main Front Peak Window / Twenty-six to Twenty-seven Years Later ]

[ Star Tide Preparatory Engineering / Initiated ]

The next snapshot came from the eleventh year after the common zero point.

Several overdue-service scientific probes on the outside had successively stopped transmitting back, the available forwarding window of the Mars relay had dropped to two percent of the original, and the dedicated outer defense nodes all remained in the planning state.

Attached to the bottom of the record page was a temporary bandwidth allocation.

Model attachments, instruction documents, and low-priority engineering volumes were postponed, and the remaining windows transmitted original counts, clock verification, and device status first.

That allocation was renewed forty-seven times in total.

During the last few issuances, whenever the Mars relay obtained a short section of stable link, it pushed a batch of data backlogged by more outer devices to the Earth-Moon system.

Only the workstation number remained on the issuance page, the name column was already blank, but the traces of that modification remained completely intact.

The fault descriptions that were originally ranked first were dragged to the end, leaving only the original counts and validation trees permanently at the very front.

Jiang Lin preserved that modification in the recovery records.

It explained why the ground archives could prove that the dust structure truly existed, yet lacked materials on how many people at the time understood it, debated it, and allocated resources.

They first preserved the things that could be re-examined later.

From the twelfth year to the fifteenth year after the common zero point, the intervals between snapshots grew shorter and shorter.

The Mars relay recovered several times and fell back into a low-availability state several times; far-side-of-the-moon observations shifted from continuous transmission to segmented submission, and the Earth-Moon defense engineering began operating relying on local time stamps.

Ground regional arrays like TM-7 took over more cache, verification, and model update tasks.

The once mutually unconnected load migrations in the Northeast Seventy-Three outer ring fault ledger found their origins on this timeline.

For every group of external observation sources that withdrew, the Mars relay had one less available baseline; for every section of Mars forwarding lost, the far-side-of-the-moon and ground arrays had to retain one more copy of data that had not yet arrived.

The phase-change cooling circuit increased its circulation frequency, the bypass takeover window was compressed, and the regional cache entered an append-only state.

These records could explain why TM-7 was overloaded, but they still could not independently prove the complete mechanism of the surface disaster.

Jiang Lin wrote the closed fault chains into the facts column, while social organization, resource decision-making, and surface personnel status remained in the pending verification area.

After the first timeline was closed, the media cartridges continued to enter and exit the low-temperature workshop at their original speed.

The closer it got to the final few snapshots, the more scattered the physical pages were, and the more media versions there were.

A record sometimes needed its main text retrieved from nine media cassettes, then went to another four media cassettes to find margin checksums, while the remaining three partitions only saved its state signature at the time of sealing.

After the twenty-sixth year of the current cycle, the twelve read heads were expanded to twenty-four, yet the real bottlenecks remained the damaged readout windows and the limited high-precision positioning components.

Layers that could not be read were left blank without exception.

Every time the positioning rack was modified for a new round, it would read just one more layer, which was then cross-checked against the redundant pages of the other fifteen partitions.

In the twenty-ninth year of the current cycle, a media cassette from the twelfth district remained in its original slot throughout the transfer.

Its transparent window had delaminated, and if scanning continued at the original angle, the local temperature rise would exceed the safety upper limit given in the maintenance volume.

This media cassette preserved the last set of timing pages of the lunar farside link before it transitioned from segmented submission to disconnection.

Jiang Lin individually built a variable-angle read head for it, reducing the incident energy to a quarter of its original workstation value, and broke down a single complete scan into one hundred and ninety-six interval tasks.

Only a small piece was read each time, and scanning resumed only after the temperature returned to the baseline.

This work took a year and three months.

During this period, the media cassette stayed in its original slot the whole time, the engineering platform changed the external position mark once a day, and the energy records of the read head were written separately into a paper ledger.

The index area was more than forty kilometers away from the forward base, and Jiang Lin later partitioned a lounge next to the low-temperature workshop, staying there for months at a time.

The meal prepared in the morning was packed into a thermal container and came over along with the narrow-gauge vehicle delivering spare parts.

On one occasion, the vehicle undertaking the transport task happened to be T-01.

Its four sets of wheel-side modules had been replaced several rounds ago, and the earliest chassis still retained the shallow dent bumped during the ramp test.

Besides two positioning components, a roll of insulated wire, and the lunch for Jiang Lin, the vehicle also carried a box of old fixtures ready to be taken back to the Outpost for cleaning.

Jiang Lin took away the thermal container and found that the return waiting time arranged for it on the task page was only eight minutes.

"Lunch has been delivered, and you will not even stay to sit for a while."

He placed the thermal container under the workbench and first finished signing off on the handover of the two positioning components for T-01 before the transport vehicle left along the original route.

Halfway through his meal, the read head of the twelfth district happened to finish the seventh scan of the day.

A section of edge encoding that could overlap with the timing page of the third district appeared on the terminal.

Jiang Lin put down his chopsticks, wrote that section into the verification queue, and did not finish the rest of the half-bucket of rice until the afternoon.

In the thirty-first year of the current cycle, the last layer of the damaged media cassette finally passed cross-verification.

The four sets of snapshots prior to the disconnection of the lunar farside link were thereby arranged into their correct positions.

In the first set, the lunar farside facilities were still submitting observational data in segments, and the tasks sent from the ground to the lunar farside had been compressed to clock synchronization, raw data receipts, and the local status of the protection engineering.

In the second set, the available forwarding window of the Mars relay dropped below two percent once again, and the lunar farside stopped waiting for the complete model package, switching to receive only raw records and validation trees.

In the third set, the Civilization Fire Archive began changing its primary custody location.

The multiple active arrays on the ground were no longer suitable as the sole primary archive; the complete observational sources, engineering status, model versions, low-dependency civilization cores, and recovery tools were repackaged, with the target address falling on an independent custody facility on the lunar farside.

This migration was not a temporary upload.

The lunar farside first received the segmented data, returning physical page checks and write confirmations batch by batch; only when the ground confirmed a batch would the next batch be allowed to be demoted from the active archive into a mirror.

Any content that had not obtained a lunar farside receipt continued to be kept in its original position and was not cleared in advance.

The fourth set of snapshots stopped at the end of the migration.

The master catalog had been fully delivered, yet the last root validation tree was still missing one of the sixteen confirmation items.

The missing item happened to fall in the sixth partition, which had not yet been restored.

Jiang Lin still marked the lunar farside primary archive as "Transferred".

The target address on the isolation terminal was already very clear, but the evidence was still short of the last page, making it impossible to rewrite fifteen-sixteenths to complete.

The sixth partition was located at the deepest part of the curved maintenance corridor; the media cabinet looked intact on the outside, but a set of mechanical slots inside had experienced high temperatures before the cataclysm.

The transparent readout window remained intact, but the metal frame fixing the media had undergone slow creep, and the optical layer deviated from the relative positioning mark by 1.6 millimeters.

Directly dismantling the media cassette would cause the frame, which had already lost its preload, to continue deforming.

Jiang Lin left the entire set of mechanical slots inside the wall, established twenty-four measurement points along the outer frame, and then had three positioning platforms refit the position of each layer without touching the media cassette.

This set of pages accounted for only a very small part of the master catalog.

It kept the lights in the low-temperature workshop on for more than another three years.

In the summer of the thirty-fifth year of the current cycle, the last page of the root validation was finally read out from the misaligned optical layer of the sixth partition.

Jiang Lin was already in his fifties, with an additional near-vision lens on the inside of his protective mask. After operating for a long time in front of the low media cabinet, he would drag a folding stool to the wall and sit for a while.

That afternoon, he sat before the isolation terminal and fed the margin checksums of the sixth partition into the master catalog.

The sixteen confirmation items turned green one by one.

The root directory immediately unfolded the receipt pages that had been locked previously.

[ CIVILIZATION FIRE ARCHIVE / Primary Custody Transfer ]

[ Target: LUNAR FARSIDE ARCHIVE / LF-03 ]

[ Root Directory Submission: Completed ]

[ Lunar Farside Media Write: Sixteen Partitions Confirmed ]

[ Ground Receipt Validation: Passed ]

Jiang Lin returned the sixteen receipts to their respective original pages and verified the submission signature once again using the lunar farside public key.

The signature was valid, and the root validation matched the fault ledger saved in the observation index area.

The complete primary archive had indeed been handed over to the lunar farside facilities at that moment before the cataclysm.

"The raw data is on the lunar farside, and the ground left a receipt for it."

He continued to scroll down the subsequent status pages.

After completing the primary custody switch, the lunar farside had returned three routine timing syncs to the ground.

Only the first half of the fourth timing sync came back, and the ground automatically switched to the extended reception window.

The Seventh Canopy Station maintained the phase reference, the Mars relay exited the timing chain, and TM-7 wrote the unfinished receipts and regional status into the local append-only cache.

After the extended window ended, the lunar farside link status turned to disconnected for the first time.

[ LUNAR FARSIDE OBS LINK / NO RETURN ]

[ Earth-Moon Protection Engineering / Switching to Local Autonomy ]

[ Regional Array / Taking Over Ground Index and Fault Preservation ]

This status was maintained for seventeen hours.

Subsequently, the TM-7 outer ring load crossed the last takeover interval, and nine local mirrors received the freeze command simultaneously.

OF-9 fused in turn, the active network and passive mirrors were physically severed, and the black data cassette in MIRROR-07 preserved the last cross-verified ground snapshot.

The fault ledger ended here.

The countdown of the main front window was still retained on the right side of the page.

[ Distance to the lower bound of the main front peak window: Seven years and two months ]

The high-density main body had not yet arrived in the inner Solar System.

The outer observational sources had withdrawn in large numbers, the Mars relay had only sporadic windows left, the lunar farside was out of contact, and the ground regional array had also entered frozen preservation.

Jiang Lin kept these states on the same timeline, leaving the endings that did not exist in the archive blank.

The existing records were sufficient to prove that the star-tide response system had lost its outer observation, cross-layer forwarding, and unified scheduling capabilities before the arrival of the main front.

It could not prove whether the lunar farside facilities were still running later, nor could it alone answer how surface civilization ultimately vanished.

In the primary archive successfully submitted for the last time by the lunar farside facilities, the complete observational sources, engineering volumes, model versions, and low-dependency civilization cores were preserved.

What the ground left behind were merely the master catalog receipts, the fault ledger, and a disconnected access route.

Attached behind the routing page were also the ground-end requirements of the awakening protocol.

The Seventh Canopy Station was responsible for providing the original phase reference, the new Earth-Moon relay had to enter the designated communication position, and the lunar farside would initiate a read-only challenge using the primary archive root validation.

The ground-end keys were not stored centrally; part of them remained in the passive mirror of the Seventh Canopy Station, and another part was kept separately by the sixteen partitions of the observation index area.

Both the Seventh Canopy Station and the sixteen partitions were still within the ground internal network that Jiang Lin had restored, and the key materials did not leave along with the primary archive.

The access route was broken at the Earth-Moon relay segment; the handshake signal sent from the ground could not reach the lunar farside, and the lunar farside response could not come back either.

Jiang Lin closed the ground primary archive search task that was still in the queue.

Over the past decade or more, the observation index area had completed surface scans of more than two thousand media cassettes, and all root addresses pointing to more complete content ultimately fell on the same location.

Continuing to dig into the ground cache would only yield more receipts and fault fragments.

He wrote the reception proof of the lunar farside primary archive into three independent media, leaving the parts for which evidence had not yet been obtained in the blank columns.

On the task page, the identity of the Seventh Canopy Station was changed from candidate to confirmed, and the status of the lunar farside primary archive had only two lines.

[ Last confirmation before cataclysm: Reception completed ]

[ Current status: Unknown ]

The twenty-four read heads were still working, and the narrow-gauge vehicle still sent the media cassettes back to the sixteen partitions according to the original queue.

Jiang Lin took off his protective mask, wiped the near-vision lens clean, and snapped it back into the inside.

The access route on the main screen pointed all the way to the surface from behind the black wall, crossed the already failed relay position, and stopped at the lunar farside.

At the very bottom of the page was still that status without a return.

[ EARTH － MOON RELAY / NO LINK ]