Microsoft’s August cumulative update for Windows 11, KB5063878, is under fire from a growing chorus of users who report that their NVMe SSDs suddenly disappear during large file transfers. The problem, first flagged by Japanese enthusiasts and now corroborated by testers worldwide, appears to be triggered by sustained sequential writes—such as game downloads, bulk backups, or archive extraction—on drives that are more than half full. While Microsoft has yet to acknowledge a storage regression in its official documentation, independent reproductions have demonstrated a consistent and alarming failure mode: the drive drops off the bus entirely, leaving in-flight data corrupted and SMART telemetry unreadable.

Released on August 12, 2025, KB5063878 (OS Build 26100.4946) delivers the monthly security and quality rollup for Windows 11 version 24H2. It also includes a Secure Boot certificate reset to revoke trust for devices whose certificates expire in June 2026. At first glance, nothing in the update suggested it would meddle with storage subsystems. Yet within days, users began posting detailed accounts of SSDs vanishing in the middle of writes—a symptom profile that has since been replicated in controlled tests.

The Update and Its Unintended Consequences

Beyond the Secure Boot reset, KB5063878 was meant to be a routine patch. But it quickly spawned two separate issues. First, an enterprise deployment problem caused WSUS and SCCM installations to fail with error 0x80240069. Microsoft acknowledged that bug and issued a release health advisory with a remediation—an effort that concluded while the storage reports were still trickling in. The second, far more worrying, is the documented NVMe SSD disappearance that has no official fix yet. The dual problems underscore how even well-tested rollups can interact unpredictably with diverse hardware fleets.

Early Warning Signs from Japan

Initial reports flooded Japanese forums and X (formerly Twitter) within days of the update. Enthusiast Necoru_cat provided a precise reproduction recipe: write roughly 50 GB of data sequentially to an SSD that is at least 60 percent full. When the trigger fires, the drive vanishes from File Explorer, Device Manager, and Disk Management. Reboot sometimes restores visibility, but files being written at the moment of failure may be corrupted, and vendor utilities often cannot retrieve SMART diagnostics. The specificity of the trigger—large writes on relatively full drives—explains why the bug can lurk undetected during normal productivity tasks and then strike when a user downloads a game update or extracts a large archive. English-language outlets like Mashable, Wccftech, and Tech Radar soon picked up the story, and reproducibility tests showed the issue is not region-locked.

Which SSDs Are Affected?

Early lists compiled by community testers and niche outlets spotlight drives built around certain Phison controllers. Models such as the Corsair MP600 and various OEM SSDs appear repeatedly in fault reports. However, the problem is not exclusive to Phison; drives using Maxio, Triton, and Marvell controllers also feature in investigator logs. The pattern suggests a host-side storage-stack regression rather than a single-vendor defect. As one forum analyst noted, “The fact that multiple controller families are showing similar symptoms tells us this is likely a host-side regression. Windows may be sending an I/O pattern that pushes certain firmware implementations past their edge cases.” The current list of affected drives remains a working hypothesis, not a definitive inventory, and no SSD vendor has issued a formal compatibility advisory.

Technical Deep Dive: Why SSDs Disappear

Large sequential writes stress an SSD’s SLC cache, metadata management, and NVMe command queues in ways that random I/O does not. If the Windows storage stack—possibly a kernel driver, NVMe port driver, or timeout handler—was altered in the update, it could expose edge cases in controller firmware. Under sustained 50 GB writes, controllers may spike to maximum utilization and then enter an unrecoverable state. The OS interprets the sudden loss of communication as the drive being removed from the bus. A power cycle often reinitializes the controller, which is why many drives reappear after reboot; but the data in flight is gone or corrupt. Recovery varies: some drives return flawlessly, others need vendor tools, and a small number require RMA. The divergence points to the fragility of specific firmware revisions when hit by unexpected host behavior.

Community reproduction efforts consistently hit the same wall: the drive stops responding, SMART telemetry becomes unreadable, and Event Viewer shows NVMe-related errors at the failure timestamp. These symptoms are hallmarks of a controller lock-up, not a file system corruption issue that chkdsk could fix. That is why testers urge users to avoid repeated reboots and instead focus on capturing diagnostic logs.

Microsoft’s Silence and a Parallel WSUS Fix

Microsoft’s KB page for KB5063878 currently states: “We are not currently aware of any issues with this update.” That language has not changed despite the mounting community evidence. The company did, however, address a separate deployment glitch—a WSUS/SCCM install failure producing error 0x80240069—with a release health update and remediation steps. The parallel fix demonstrates that the August rollup has already produced more than one unintended behavior, but it offers no relief to those wrestling with disappearing drives. Until official telemetry confirms a storage regression, users are left relying on community guidance.

What Users Should Do Right Now

If You Have the Update Installed

  • Back up immediately to a different physical device or cloud storage. The failure can corrupt in-flight data, so preserve everything you can now.
  • Pause large sequential writes. Postpone game updates, bulk copies, disk cloning, and archive extraction until a fix is confirmed.
  • Keep Windows Update enabled to receive any future hotfix; disabling updates completely could leave you exposed to other vulnerabilities.
  • Check your SSD vendor’s dashboard for firmware updates. Apply them only after verifying the backup is current and the firmware addresses controller edge cases.

If You Haven’t Installed the Update

  • Consider pausing Windows Updates for up to a week if your workloads involve heavy writes and you use a potentially affected SSD. In Windows Settings → Windows Update, you can defer updates while you monitor community and vendor advisories.
  • If pausing is not an option, ensure backups are fresh before performing any large I/O operations.

If a Drive Becomes Inaccessible

  • Stop all writes immediately. Repeated reboots may overwrite recoverable metadata.
  • Record exact symptoms in Device Manager, Disk Management, and your SSD vendor’s toolbox.
  • Capture Event Viewer logs around the failure time (System log, NVMe events).
  • Attempt a SMART readout using vendor utilities in read-only mode.
  • If the partition is visible but corrupted, create a sector-level image (bit-for-bit clone) to another drive before running any repair tools.
  • Contact Microsoft Feedback Hub and your SSD vendor support with all collected logs.

Enterprise IT: Halt Rollouts and Test

IT administrators should pause mass deployment of KB5063878 immediately. Stage the update in a test ring and execute sustained sequential write workloads on drives that are at least 60% full—the known trigger profile. The earlier WSUS installation error already justified caution; the storage reports raise the risk level significantly. Ensure backup targets are on separate hardware not sharing the same SSD models under investigation. Centralize NVMe error telemetry from endpoint monitoring systems to share with Microsoft and hardware vendors; this data accelerates root cause identification.

What Microsoft and Hardware Vendors Need to Do

Community testers have handed investigators a reproducible lead. Now the responsibility shifts. Microsoft and SSD manufacturers should:
- Issue a transparent advisory if telemetry confirms a storage regression.
- Publish affected firmware revision tables so users can check their drives without guesswork.
- Provide safe diagnostic steps that prioritize read-only imaging over in-place repairs.
- Commit to a timeline for a driver hotfix or firmware update.

Without authoritative guidance, users are left navigating conflicting advice and escalating anxiety.

Risk Assessment

For the typical Windows 11 user, the likelihood of hitting this bug is low to moderate. It requires a specific workload—large sequential writes to a relatively full drive—that many never perform. However, the impact is high: data corruption, possible drive inaccessibility, and in rare cases an RMA. Creative professionals, gamers, and IT operations that routinely move large datasets face the greatest exposure. The geographic clustering of early reports in Japan was a reflection of enthusiast activity, not a region-specific codepath. As the news spreads, reports will likely surface from every market.

Looking Ahead

The next 48–72 hours are critical. SSD vendors may release firmware updates if their products are implicated, and Microsoft could provide an out-of-band fix or at least update its known-issues list. In the meantime, treat community reports as a high-priority early warning. Back up your data, hit pause on that 100 GB game download, and keep one eye on official channels. The Windows 11 storage stack has been mature for years; this regression, once pinpointed, will almost certainly be patched swiftly. Until then, caution is the only sane policy.