Abuse Policy
Reports answered 24/7 · acknowledgement under 30 minutes
Legal · Version 3.0
Abuse Policy — report it, and here is what happens.
This page is for anyone — customer or stranger — whose problem is being caused by a server on our network. It sets out what we treat as abuse, the exact evidence a report must contain before we can act, how fast each severity is answered, which measure we apply, how the affected customer is notified, and how either side can challenge the outcome. WenHost is an independently run hosting business registered in Viet Nam; we are not a court and we do not investigate crimes, but we do stop harm leaving our network, and we do it on a documented clock.
Effective 29 October 2026 · Applies to all WenHost infrastructure
- No account required to report
- Evidence before action
- Both sides heard
- Outcome confirmed in writing
The whole policy in four numbers
Every target below is measured automatically on the abuse queue and published quarterly.
30min
Acknowledgement
A human reads every abuse report and replies with a case number. Critical cases are triaged inside this window, day or night.
4h
Triage complete
Evidence verified against our own logs, severity assigned, and the affected customer notified with what we were shown.
24h
Decision & action
The narrowest effective measure is applied and you receive the outcome category. Critical harm is stopped sooner, not later.
7days
Appeal window
The customer affected may challenge any action, in writing, and roughly one in nine appeals succeeds. We publish that rate.
Two documents, two audiences. The Acceptable Use Policy tells our customers what they may not run. This policy tells everyone else how to report it, and tells both sides what we owe them: evidence, speed, proportionality, and a hearing before anything permanent happens. Where a report concerns prohibited content or network abuse, this policy sets the process and the AUP sets the standard of conduct.
Scope — who may report, and about what
We accept abuse reports about conduct originating from IP addresses, domains or services on WenHost infrastructure. You do not need to be a customer, you do not need an account, and we do not require you to identify yourself to a competitor — but we do require a verifiable contact address, because we cannot act on an anonymous accusation against a paying customer.
This policy covers reports from: individuals and businesses harmed by traffic we deliver; security researchers and Computer Emergency Response Teams; anti-abuse and blocklist operators; brand owners and their agents; rights holders and their representatives; and competent authorities.
It applies to every region we operate and to every layer we control: the instance, its network attachment, its IP allocation, and the reverse DNS we administer. It does not apply to a customer's own internal conduct that never touches our network, nor to services we merely link to from our website.
Definitions
| Term | Meaning in this policy |
|---|---|
| Reporter | The person or organisation submitting a complaint, with a verifiable contact address. |
| Respondent | The WenHost customer whose service is implicated. |
| Abuse | Use of our infrastructure that harms, threatens, defrauds or imposes unagreed cost on a third party, or that breaches the Acceptable Use Policy. |
| Credible report | A dated report naming the offending IP, the exact UTC time, an attached artefact, and a verifiable reporter. |
| Artefact | Primary evidence: full mail headers, packet capture, access log line, or URL with page source and timestamp. |
| Corroboration | Our own telemetry or firewall record confirming the same IP and time window. |
| Measure | Any action we take, from a notice through to termination. |
| Case number | The reference issued on acknowledgement; quote it in all further correspondence. |
What we treat as abuse
These categories are actionable. Each maps to a clause of the Acceptable Use Policy, which is the standard we enforce against.
- Unsolicited bulk email
- Phishing & credential theft
- Malware distribution
- Botnet command & control
- DDoS participation
- Network scanning
- Brute-force attacks
- Exploitation of third-party systems
- IP spoofing
- Fraud & scams
- Illegal content
- Copyright infringement
- Trademark infringement
- Non-consensual imagery
- Harassment & threats
- Open proxies & relays
- Tor exit relays
- Child exploitation
Reports about a server being used against you and reports about a server being itself compromised are both welcome. We treat them differently: the first is a conduct problem, the second is usually a victim, and we would rather clean up a hacked box with its owner than punish them for it.
What we do not handle
✕ Out of scope — and who to contact instead
- Civil disputes between you and a customer.Unpaid invoices, contract arguments, delivery disagreements and refund fights are not abuse. Take them up with the party directly or through your own legal route.
- Content we do not host.If the material sits on another provider's network, send the report to that provider's abuse contact — we cannot act on infrastructure we do not control.
- Domain registration disputes.Ownership and whois accuracy go to the registrar and, where relevant, the registry or UDRP provider.
- Opinions, taste and lawful-but-unpopular speech.We are not a content arbiter. Lawful expression hosted by a consenting adult customer is not abuse.
- Search rankings, SEO and marketplace complaints.Not a hosting matter.
- Requests for customer identity or records without legal process.We will not name a customer to a private complainant (clause 11).
- Legal advice or investigation services.We provide the evidence we hold about our own network; we do not build your case.
How to submit a report
Send it through the contact form and choose the topic Abuse report. That route is monitored continuously and enters the abuse queue directly — it is faster than any forwarded message.
✓ Do this
- One report per incident.Group the evidence for one IP and one time window; do not merge unrelated cases.
- Quote the case number on every follow-up instead of sending a new report.
- Attach the artefact in its original form — raw headers, not a screenshot of a summary.
- State the harm and which clause 3 category it falls under.
- Tell us if it is ongoing and whether someone is in immediate danger.
✕ Avoid this
- Bare IP lists with no artefact.These are returned with a request for evidence, not actioned.
- Mass-reporting to force an outcome.Volume is not evidence and is treated under clause 17.
- Threats of public shaming or of reporting to a payment network as leverage in a dispute.
- Scans of our whole network submitted as "findings" — probing ranges you do not own is itself prohibited.
- Reports about a customer's private conduct that never touches our network.
The evidence every report must contain
This is the clause that decides whether we can help you. A report missing these fields cannot be actioned fairly against a customer, so we ask for the gaps rather than act on a guess.
| Field | Required | Why we need it |
|---|---|---|
| Offending IP or hostname | Yes | Identifies which instance and which customer, if any. |
| Exact UTC timestamp | Yes | Our logs are time-based; "recently" is not searchable. |
| Artefact | Yes | Full mail headers, pcap, access log line, or URL plus page source. |
| Direction & ports | Yes | Source, destination and port separate an attack from a reply. |
| Reporter identity | Yes | A verifiable name and contact address. We do not act anonymously. |
| Your relationship to it | Yes | Victim, rights owner, authorised agent, or researcher. |
| Repeat-incident history | Helpful | Establishes a pattern, which changes the measure applied. |
| Case numbers from before | Helpful | Lets us see whether a fix was claimed and then lapsed. |
| Rights ownership statement | For IP claims | Clause 14 requires it before a takedown is valid. |
What we do with it. We compare your artefact against our own records — firewall logs, connection telemetry, mail-flow and network evidence for the same IP and window. A report is actionable when both sides line up. Where they contradict, we tell you what we found and why, and we do not suspend on your word alone.
Response targets, by stage
| Stage | Critical | High | Medium / Low |
|---|---|---|---|
| Human acknowledgement | < 30 min | < 30 min | < 4 hours |
| Triage & corroboration | < 1 hour | < 4 hours | < 1 business day |
| Respondent notified | Immediate action, notice within 24 h | < 4 hours | With the notice window |
| Measure applied | < 1 hour | < 24 hours | 24 h – 7 days by tier |
| Outcome to reporter | < 24 hours | < 24 hours | < 3 business days |
| Re-review on new evidence | < 4 hours | < 1 business day | < 3 business days |
Targets are working commitments, not warranties, and they are met 89–96% of the time on a twelve-month rolling basis. Where a case needs more — a court order to review, a complex compromise to investigate — we say so inside the window rather than going quiet.
Severity classification
Severity decides the clock and the measure. It is assigned by our team on the evidence, not by how urgently the report is worded, and it is stated to both parties in writing.
| Tier | Typical case | Respondent window | Escalates if |
|---|---|---|---|
| Critical | Active attack source, C2, ransomware, live phishing, child exploitation | None — act first, notify within 24 h | Termination reviewed immediately |
| High | Confirmed spam source, credential brute force, hosting phishing pages | 24 hours to respond and fix | Suspension until evidenced remediation |
| Medium | Scanning a third party, exposed open resolver, repeat blocklist entry | 72 hours | Affected service blocked meanwhile |
| Low | Single complaint, misconfiguration with no confirmed harm | 7 days | Notice with logs and a fix suggestion |
| Informational | Reputation data, third-party list entry, no live traffic observed | 7 days | Advisory only |
Three notices of the same category within 90 days escalate one tier, because the third notice proves the first two did not work. That rule is published in the Acceptable Use Policy and applied identically here.
The measures we apply — narrowest effective action
We choose the smallest action that stops the harm. Taking a whole customer offline for one misbehaving process is not proportionate, and it is not our default.
- Measure 1
Notice
The respondent receives the evidence and a fix window. Nothing is interrupted. Most Low and Medium cases end here.
- Measure 2
Targeted control
Outbound port
25or587blocked, one service disabled, a single URL or path blocked, or a specific destination filtered. The server keeps running. - Measure 3
Rate limiting or temporary null-route
Applied to an IP under attack or saturating a shared uplink, so neighbours keep capacity. Reversed as soon as the event passes, and reported to the customer with start and end times.
- Measure 4
Suspension
Instance powered off, network disabled, data intact and restorable through the client area. Used for High and unremedied Medium cases.
- Measure 5
Termination
Account closed and instance released for Critical or persistently unremedied breaches. Disks destroyed on the schedule in clause 9 of the Terms of Service, with snapshots kept 7 days.
What we will not do: hand over your identity to the respondent beyond what the report itself necessarily reveals, delete lawful content on request, censor a service because a competitor complained, or act as an enforcement arm in a private dispute.
What the affected customer is told
Fairness to the respondent is part of the process, not an afterthought. Within the applicable window the customer receives:
- The category and the clause breached — named, not vague.
- The evidence we relied on: the artefact you sent and our own corroborating logs, with timestamps.
- The measure applied and its start time, plus what will restore service.
- The deadline to respond, and the appeal route with its 7-day window.
- What we did not act on, if part of the report was unsupported.
We do not reveal your identity or contact details to the customer unless disclosure is legally required or you authorise it. Where a report necessarily identifies you — a phishing complaint naming your own brand, for example — we tell you before we forward it, so you can request redaction.
Reporter confidentiality, and customer confidentiality
Two promises, both firm:
- To reporters: we do not publish your name, your contact details or your report. We share the technical facts needed for a customer to fix a problem, not the identity of the person who reported it.
- To customers: we do not disclose who our customers are, what they run, or their identity to a private complainant. Personal data is handled under the Privacy Policy, and disclosure happens only on a valid legal order or with the customer's own instruction.
If you need a customer's identity for a legal claim, the route is a court order, regulator request or lawful discovery process — clause 12 — not a complaint to us. We respond to valid process quickly and we narrow it where we can.
Law enforcement, courts and regulators
- Receipt
Validated for authenticity and jurisdiction
We confirm the request genuinely comes from a competent authority and applies to us, before anything is disclosed.
- Assessment
Narrowed to what is legally required
We disclose the minimum responsive to the order and push back on overbroad requests rather than complying out of convenience.
- Action
Preservation or disclosure as ordered
Preservation requests are honoured promptly so evidence is not lost while the legal process runs.
- Notice
The customer is told — unless barred
We notify the affected customer so they can challenge the order, unless we are legally prohibited or notification would endanger someone.
Emergency requests involving risk to life or to a child are assessed immediately, on a documented standard, and answered the same day where the legal basis exists. Voluntary disclosure beyond that standard is rare, reasoned in writing, and recorded in our transparency report.
Child safety and emergencies — our highest priority
Child sexual abuse material is terminated and reported immediately, without notice to the account and without waiting for any window in this policy. We report to the competent authority in Viet Nam and to the relevant international centre — for example NCMEC where the material or the reporter is in the United States — and we do not condition that report on a request being made.
- Non-consensual intimate imagery is removed on a valid report from the person depicted or their authorised representative, without requiring a court order, and we help with takedown notices to other hosts where we can.
- Imminent threat to life or physical safety is escalated to management and to the authorities on receipt, and we act to stop the service while that assessment runs.
- Any emergency report should say plainly that it is an emergency in the first line, so it is routed immediately rather than queued.
- Reports of this category are handled by a named senior reviewer, and the outcome is recorded and audited.
Copyright and trademark notices
Intellectual-property claims are abuse reports with a stricter evidence standard, because they are made under penalty of perjury in most jurisdictions. A valid notice must contain all five elements:
- Identification of the work claimed to be infringed.
- The exact URL or IP and a timestamp showing the material as served from our infrastructure.
- Contact details of the rights owner or their authorised agent.
- A good-faith statement that the use is not authorised by the owner, its agent, or the law.
- A statement, under penalty of perjury, that the information is accurate and that the sender is authorised to act.
On receipt: logged and acknowledged the same day, forwarded to the customer with the actual evidence, 24 hours to remove or respond, and a valid counter-notice is passed back to the complainant. If the complainant does not pursue it, we restore the content and confirm in writing. Repeat infringement on the same account is terminated after a documented pattern — never on a first allegation, and never on an automated mass notice that names no evidence. Trademark claims follow the same route, and counterfeiting with consumer harm is treated as fraud under clause 15.
Phishing, fraud and brand abuse
These are the fastest takedowns we perform, because every hour a page stays up steals from someone who is not our customer.
✓ Send us
- The full URL and the exact time you observed it, in UTC.
- Page source or a screenshot showing the impersonated brand.
- The lure — the email or message that directed a victim there, with headers.
- Proof you are the brand owner or authorised agent of the impersonated mark.
- Whether credentials were submitted, which raises severity to Critical.
✕ What we will not accept
- Complaints about lawful criticism or parody framed as "brand abuse".
- Trademark disputes between two businesses with no consumer deception.
- Requests to remove a page because it mentions a brand negatively.
- Reports from an unidentified party claiming to act for a brand.
Verified phishing collecting live credentials is suspended within the hour under clause 9 measure 4, and the customer receives the evidence with the notice. Where the instance was compromised rather than operated by the customer, we say so, and remediation replaces punishment.
Spam reports and blocklist removal
- Always send full message headers. A summary or a screenshot cannot be traced; raw headers can. Received lines give us the sending host, the time and the path.
- We corroborate before acting. The header must line up with our mail-flow and connection records for that IP and window. Where it does, the customer is notified with both artefacts.
- Port
25and587are blocked outbound by default on every instance, so most spam originates from a service the customer opened deliberately — which is why the fix is usually a configuration change, not a shutdown. - Blocklist delisting: once the source is confirmed stopped, we assist with removal requests to major lists and provide the evidence they ask for. We cannot promise any list's decision, and we do not appeal a listing caused by the customer's own repeat conduct.
- Abuse of a clean IP on arrival: if an address you were assigned is already listed, tell us within 7 days of delivery and we replace it free, per clause 8 of the Delivery Policy.
Bad-faith, false and automated reporting
The reporting channel is a safety mechanism, and using it as a weapon is itself abuse.
- What we detect: fabricated artefacts, doctored headers, reports that recite our own evidence back after a compromise, mass submissions from one sender, and coordinated flooding of a target account after a competitor's complaint.
- What happens: the report is rejected with reasons, the sender is blocked from the abuse queue, and where a false statement was made under penalty of perjury we preserve the record and may refer it.
- Threats are not leverage: telling us you will report to a payment network, post publicly, or file a chargeback unless we act does not change the evidence standard. It is noted in the file.
- Scanning our network is not research: probing address ranges you do not own breaches the Acceptable Use Policy, even when the findings are then sent to us.
- Good-faith mistakes are treated differently. A well-intentioned report with weak evidence gets a request for more, not a sanction. We only act against senders who repeatedly fabricate or flood.
Appeals — for the respondent and the reporter
If you are the customer: every suspension and termination is appealable within 7 days. Reply to the notice or use the contact form with the case number and your evidence. A reviewer who did not make the original decision answers within 72 hours. Where an appeal succeeds, service is restored, downtime is credited under the SLA, and we never require an admission of fault as a condition of reversing our own mistake. Roughly one in nine appeals succeeds and we publish that figure deliberately.
If you are the reporter: you may ask us to reconsider a "no breach found" outcome by supplying new evidence — a different time window, a fuller artefact, or a repeat incident. We re-review within the clause 7 targets. We cannot disclose the customer's response or personal data to you, but we will tell you which outcome category applies: resolved, no breach found, customer unresponsive, or terminated.
Transparency reporting and record retention
Accountability needs numbers. Each quarter we publish, without identifying any customer or reporter: reports received by category, corroborated by our own logs, measures applied at each tier, time-to-action medians, appeals filed and appeals that succeeded, legal requests received and complied with in full or in part, and emergency reports referred to authorities.
| Record | Kept for | Why |
|---|---|---|
| Abuse reports & evidence | 24 months | Pattern detection and defending a claim |
| Security & access logs | 90 days rolling | Corroborating a report while it is live |
| Network telemetry | 90 days rolling | SLA measurement and attack forensics |
| Case outcomes | 24 months | Transparency statistics and repeat tracking |
| Legal requests | As law requires | Compliance record and transparency report |
| Blocked senders (clause 17) | 24 months | Preventing renewed bad-faith reporting |
Retention periods match the Privacy Policy exactly. Preserve your own evidence too: our logs roll at 90 days, so a report about an event older than that may be unverifiable through no fault of yours.
Changes to this policy, precedence and governing law
The version and effective date at the top of this page show the text in force; superseded versions are available on request. Material changes — a new category, a shorter window, a broader immediate-action list — are announced on the status page and by email to customers at least 14 days before taking effect, matching the standard in the Terms of Service, Privacy Policy and Delivery Policy. Editorial clarifications take effect immediately with the date updated.
Order of precedence, stated identically across WenHost documents: on prohibited content and network abuse the Acceptable Use Policy sets the standard and this policy sets the process; on refund eligibility the Refund Policy governs; on delivery timing the Delivery Policy; on personal data the Privacy Policy; otherwise the Terms of Service. Where two documents still conflict, the reading most favourable to the customer applies.
Governing law: Viet Nam, without regard to conflict-of-law rules. Nothing in this policy limits a non-waivable consumer right you hold where you live, and no clause here obliges us to break the law of a jurisdiction in which we operate.
Seven questions reporters ask us
Short answers, each pointing at the clause that governs it.
Do I need a WenHost account to report abuse?
No. Anyone may report, and you do not need to be a customer. You do need a verifiable contact address, because we will not act against a paying customer on an anonymous accusation (clauses 1 and 6).
I only have a screenshot. Is that enough?
Usually not. We need the raw artefact — full mail headers, a pcap, an access-log line, or the URL with page source and a UTC timestamp — so we can corroborate it against our own logs. A screenshot cannot be traced to a session (clause 6).
How fast will you act?
A human acknowledges inside 30 minutes, triage completes within 4 hours, and a measure is applied within 24 hours for anything above Medium. Active attacks, live phishing and child exploitation are stopped within the hour (clauses 7, 8 and 13).
Will you tell me who the customer is?
No. We confirm the outcome category — resolved, no breach found, customer unresponsive, terminated — but customer identity is disclosed only on a valid legal order (clauses 11, 12 and 18).
You said "no breach found". Can I push back?
Yes. Send new evidence — a different time window, a fuller artefact, or a repeat incident — and we re-review within the clause 7 targets. Volume without new evidence does not change the answer (clauses 17 and 18).
My event happened four months ago. Too late?
Likely unverifiable on our side: security and network logs roll at 90 days (clause 19). Send it anyway with your artefact — a pattern across several reports can still establish a case, and repeat offenders are tracked for 24 months.
Is this the same as your Acceptable Use Policy?
No, and they are written for different readers. The AUP tells our customers what they may not run; this policy tells everyone how to report it and what process both sides are owed. The AUP sets the standard of conduct; this policy sets the enforcement clock.
Something on our network is hurting you
Report it with the IP, the exact UTC time and the raw evidence, and choose the topic Abuse report. A human acknowledges inside 30 minutes, day or night, and tells you which outcome category applies. If someone is in immediate danger, say so in the first line and the report is escalated at once.
Version 3.0 · Effective 29 October 2026. This policy sets out WenHost abuse-handling process and is not legal advice. Where these terms and the Refund Policy differ on refund eligibility, the Refund Policy applies.
