SERVICES · LAS VEGAS NV

Charles Trinh — Self-Hosted Drive and Search | Las Vegas, NV · Las Vegas, Nevada · Henderson · Clark County, Nevada · North Las Vegas, Nevada

Charles Trinh — Self-Hosted Drive and Search, Las Vegas NV
Charles Trinh — Self-Hosted Drive and Search

Self-hosted file storage and search for a regulated estate: access models, index security, backup and the vendor-exit argument that makes self-hosting worth the operational cost.

Why self-host at all

The case is control over where data rests, who can read it, and an exit path that does not depend on a supplier's API. The cost is that patching, backup and recovery become yours.

Choose it when residency or exit cost matters more than convenience, and write the reasoning down because the next review will ask.

Access at the storage layer

Model access by group and by document class, and enforce it in the storage layer rather than in the client. A permission enforced in the UI is a permission that survives in the API.

Test it by calling the API directly with a credential that should not have access, rather than by reading the admin console.

The index is a copy of the data

Treat the search index as a second copy of the data and protect it accordingly, including encryption at rest and access control on query.

A rebuilt index that loses permissions is a data exposure, so test restore of both documents and index together.

Backup and verification

Run restore drills quarterly and record time to recovery, because that number is the only one that matters during an incident.

Verify the backup by opening a file from it, not by checking that the job exited zero.

Choosing the software

The choice is between a synchronised file service, a document management system and a web drive, and the three have different permission models and different failure modes. A synchronised service is convenient and gives every client a copy of the data, which is a data-map problem; a document management system gives control and adds workflow. Decide which problem is being solved before comparing features, because the features are similar and the consequences are not.

Check the exit path of the software itself before adopting it, which is the same discipline applied to the vendors you are trying to escape. A self-hosted tool with a proprietary store and no export is a vendor dependency with extra steps, and the irony is expensive.

Permissions that hold under the API

The permission model must be enforced at the storage layer, and the test is a direct API call with a credential that should not have access. An interface that hides a folder is an interface, not a control. Where permissions are cached for performance, the cache needs an invalidation path with a bound, because a stale cache is an access grant that outlives the change.

Group membership should come from the directory rather than from the application, so that leavers lose access at the same moment they lose everything else. Where the application maintains its own groups, the joiners and leavers process has two copies and one of them will be neglected.

The index is a second copy

A search index contains the text of the documents and its own access model, which makes it a second copy of the data with a second set of permissions. Protect it with the same classification, encrypt it at rest, and test that a query as a user without access does not return the document. Rebuilt indexes that lose permissions are a well-known exposure, and the fix is to derive permissions at query time from the storage layer rather than from the index.

Where full-text search requires the content, accept the trade explicitly and record it. The alternative is a search that does not work, and a search that does not work produces the shadow copies the estate was trying to eliminate.

Backup both halves

Back up documents and index together and restore them together, because a restored document store with a missing index looks like data loss to every user even when the files are intact. Record the restore time for both halves separately, since the index often takes longer and the assumption that it will be quick is what makes a recovery plan optimistic.

Verify the restore by opening a file from it and running a search that returns it. Anything less is a check that the job ran rather than that the data is there.

Operating it

Assign an owner and a review date, patch on a schedule, and monitor disk usage against a forecast rather than against a threshold. The failure mode of a self-hosted store is a full disk on a Friday, and the forecast is what prevents it. Keep the runbook with the inventory, because the person who built it will not be the person on call.

The migration question

Moving an existing shared drive into a self-hosted store is a project in permissions rather than in bytes. The old share has permissions that accumulated over a decade and nobody can explain, and the migration is the moment to decide which of them are real. Publish the new model, map groups to it, and run the migration as a copy with a read-only period before the switch, so the exceptions surface with people still able to explain them.

Expect a parallel period where both systems are live, and bound it. An unbounded parallel period produces two sources of truth and the older one always wins because that is where the file was saved last.

Finish with a deletion of the old share, with a date and a record. A share left mounted after a migration is an unmanaged copy of the data the project existed to control.

Monitoring the store

Monitor three things: disk headroom against a growth forecast, the age of the most recent successful backup, and the rate of permission changes. A spike in permission changes is either a migration or a mistake, and both are worth seeing the same day. Everything else is diagnostic depth for the incident.

Alert to the owner with the first action, and record the owner in the component page. A store without a named owner is a store whose failures are discovered by users.

Cost of ownership, honestly

Self-hosting trades a subscription for an operating cost that is easy to underestimate: patching, backup verification, capacity, and the time to answer a user's question about a share. Add those up before adopting, and compare against the vendor cost including the exit premium the self-hosted option avoids. The comparison is usually closer than the ideology suggests, and it should be made in numbers rather than in principles.

Where the numbers favour self-hosting, write down which assumption carries the decision. It is usually the exit cost or the residency requirement, and both change.

Review the store's access log quarterly for grants that were never used and accounts that have not logged in. Both are removal candidates, and both accumulate silently in a store nobody visits daily.

Publish the retention rules for shared drives in the same place as the access model, because a store whose retention is a surprise is a store that will be cited in a records request.

Check the store's search behaviour after any permission change, because an index that has not caught up is an exposure window with a user interface.

Proof

  • ONC (g)(10) self-test PASS, 317 of 317 subtests, April 2026 — public suite URL
  • 67 public hostnames, 51 serving, one AWS account
  • 104 clearinghouse enrolments across 53 jurisdictions
  • USCDI v7 proposal — Patient Digital Identity Credential — listed at isp.healthit.gov
  • SAM.gov UEI MVKQW9DKKGB8 · CAGE 20KM5 · NPI 1194645564
  • AWS Business Associate Addendum active since 18 August 2026

Where this work is delivered

Engagements run from Las Vegas and reach across the valley: clients in Las Vegas, NV; Las Vegas, Nevada; Henderson; Clark County, Nevada; and North Las Vegas, Nevada — plus remote delivery nationwide.

How an engagement runs

1. Trace one real case

Follow a single patient or claim end to end with a timestamp at every hop, before any document is written.

2. Written findings

A findings list naming the defect, the system, the owner and the date.

3. Monthly review

Review the list monthly and check whether the dates held.

Frequently asked

What is this service?

This page describes charles trinh — self-hosted drive and search — the scope, the failure modes it prevents and how the work is measured. It sits in the Cloud and Operations category at trinh.io.

Where is the work delivered?

Engagements run from Las Vegas, Nevada across the valley — Las Vegas, Henderson, North Las Vegas, Spring Valley, Paradise, Winchester, Sunrise Manor, Enterprise, Summerlin, Clark County — and remotely anywhere in the United States. On-site work in the Las Vegas valley, remote delivery elsewhere.

What does it cost?

Standard rate $195 per hour, committed $165 per hour at twenty hours or more per week for three months or more; $1,600 standard day, $1,350 committed day. Scoped in writing before work starts, fixed fee where the deliverable is defined.

Who is Charles Trinh?

A technology architect in Las Vegas, Nevada, working in healthcare identity, consent and interoperability. Not a physician, not a medical practice, and not affiliated with any other Charles Trinh.

How does an engagement start?

With one trace of a real case and a written findings list naming the defect, the system, the owner and the date. The first call is twenty minutes and free: 702-559-3489.

Engagement

Scoped in writing before work starts; fixed fee where the deliverable is defined. First call is 20 minutes and free. 702-559-3489 · charles@trinh.io

Talk to Charles

Record

Profile

Healthcare technology architect working where federal regulation becomes production infrastructure. Built and operates a FHIR R4 clinical platform — SMART App Launch authorization server, patient and provider portals, SQLite — on a hardened single-host AWS estate of 67 public hostnames. The platform returned a self-test PASS on the ONC (g)(10) Standardized API test suite in April 2026, 317 of 317 subtests, publicly verifiable. Designed a consent architecture in which a permissioned ledger records patient grant, refusal and revocation as immutable events while all clinical data stays off-chain and encrypted per patient — no PHI on the ledger by design. Filed a proposed USCDI v7 data element with ONC on verifiable patient identity. Translates the Federal Register, HHS/CMS guidance and HITRUST requirements into architecture, controls and audit evidence.

Selected engineering scale

  • Self-managed estate of 67 public hostnames, 51 serving, on one AWS account.
  • FHIR R4 API with a SMART App Launch authorization server; patient and provider portals; an AI service layer; Hyperledger Fabric peer, orderer and CA; IPFS; a 337-table SQLite clinical store.
  • Two permissioned distributed ledgers in production, one operating a consent registry that records grant, refusal and revocation with no PHI on chain.
  • X12 005010 pipelines across eight transaction sets.
  • Designed, built, deployed and operated end to end by one person.

Credentials and standards work

  • Inferno (g)(10) self-test PASS, 317 of 317 subtests, April 2026 — the ONC Standardized API test suite. Standards validated: US Core 6.1.0, USCDI v3, SMART App Launch 2.0.0, Bulk Data 2.0.0. Public suite URL.
  • USCDI v7 public comment submission, April 2026 — proposed Patient Digital Identity Credential under the Provenance data class, listed publicly on the ONC Interoperability Standards Platform.
  • SAM.gov UEI MVKQW9DKKGB8 · CAGE 20KM5 · NPI 1194645564. AWS Business Associate Addendum active 18 August 2026.

Technical environment

AWS us-east-1 (compute, storage, messaging, SMS, email, voice) · Linux and systemd hardening · nginx · Fabric and EVM ledgers · FHIR R4 · X12 005010 · HL7 · SQLite · self-hosted mail, drive, search and monitoring · Git-based change control and CI release gates.

TRINH — Charles Trinh, technology architect
TRINH

Charles Trinh

Fractional CTO/CIO and Technology Architect — Healthcare Data, Identity & Regulated Infrastructure · Las Vegas, NV

Las Vegas, Nevada · US citizen · Fractional and project engagements

(702) 559-3489 Email Charles Trinh