# Static IP for IP Allowlisting: One Address for Remote Teams | ProxyJam

A static IP for IP allowlisting: why home and mobile addresses break an allowlist, how a remote team shares one fixed address, and why 2FA still matters.

- Canonical: https://proxyjam.com/blog/static-ip-for-ip-allowlisting/
- Updated: 2026-11-02
- Language: en

An IP allowlist admits a connection by the address it comes from. Home broadband addresses can change, and mobile addresses are shared, so a team without an office needs one fixed address of its own. A static ISP proxy is one: every member connects with its username and password, and the systems you allowlist see a single address. Keep it next to per-person sign-in with two-factor authentication.


## Check these before the allowlist goes live

1. The address on [What Is My IP](/what-is-my-ip/), opened through the proxy, is the one you are adding to the list.
2. At least two administrators can reach the system's settings, and one of them has a way in that does not depend on the proxy.
3. Every person who needs the system has their own account on it, with two-factor authentication turned on.
4. Auto-extend is on for the address in My Proxies, or someone owns the renewal date.
5. The proxy's username and password sit in your team's password manager, shared only with the people who use the allowlisted systems.

## Give the team one address

1. **Order one address** Pick the country your team works in on the [ISP proxies page](/isp-proxies/) and order one address for the whole team.
2. **Copy the connection details** Take the host, port, username and password from My Proxies in the dashboard. The order starts on HTTP, which carries HTTPS, and that is the protocol to keep for browsers.
3. **Set up each browser** Install [ProxyJam Connector](/browser-extension/), add the proxy under Manage profiles, choose Auto and write a rule that sends the allowlisted system's hostnames through it.
4. **Set up scripts and tools** Give the same four values to code that calls the system. The [integrations guides](/integrations/) show where requests, Selenium and Playwright take them.
5. **Confirm the address** Open [What Is My IP](/what-is-my-ip/) through the proxy and copy the address it shows.
6. **Add it to the allowlist** Enter it as a single address, which AWS writes with a /32 suffix, then test from a member's browser before you switch enforcement on.

## Office line, VPN or static proxy

|  | Office line | VPN with a fixed exit | Static ISP proxy |
|---|---|---|---|
| The fixed address comes from | A business internet plan at the office | The VPN server, or a business VPN plan with a dedicated address | The proxy, held for the period you pay for |
| Traffic that uses it | Everything on the office network | The whole device, or the apps a split tunnel names | Only what you route through it: a browser, a script, a tool with a proxy setting |
| What each member installs | Nothing, while at the office | A VPN client on every device | A browser extension, or a proxy setting in the tool |
| What someone has to run | The office network | The VPN server and its accounts | Nothing beyond the order and its renewal |
| Fits a team that | Works from one building | Needs desktop apps and SSH on the allowlisted path | Reaches the allowlisted systems from a browser or code |

Each row describes how the setup works, from the vendor documentation listed under Sources and our own product pages. We have not tested any VPN product or office connection, and none of the three is rated better overall: the right one depends on which traffic has to reach the allowlisted systems.


## Quick answers

### Can several people use one static IP at the same time?

Yes. Each proxy carries up to 100 concurrent sessions, whatever the traffic (Terms of Service, section 2.4). A browser opens several connections at once, so route only the allowlisted systems through the proxy and leave the rest of the team's browsing on its own connection.

### Should the proxy itself check our addresses, or a password?

A password, for people. An order can also admit clients by an IP allowlist of its own, set through the API with PUT /orders/ORDER_ID/whitelist, and while that list is set the username and password stop working. Members' home addresses change, so that list would lock them out of the proxy. It suits a server or a test runner with a fixed address of its own.

### Does the address change when we renew?

No. The address stays yours for the period you paid for, and renewing keeps the same address. A different address takes a new order, and its allowlist entry has to be added before the old one is removed.

### Does an allowlist need a residential address?

No. An allowlist compares numbers, and a datacenter address passes it the same way. A static ISP address helps when the same address also reaches sites that treat hosting networks differently, and it spares you running a server.

### Can we allowlist a phone's hotspot address?

Not safely. Mobile carriers commonly put many subscribers behind one public address through carrier-grade NAT, so an entry for it can admit people who are not on your team, and the address moves when the carrier reassigns it.

### Does the extension send all of a member's traffic through the proxy?

No. It sets the browser's proxy only, and in Auto mode only for the sites your rules name. Desktop apps and command-line tools use their own proxy settings, or none, which is where a VPN fits better.


## Sources

- [GitHub Docs: Managing allowed IP addresses for your organization](https://docs.github.com/en/enterprise-cloud@latest/organizations/keeping-your-organization-secure/managing-security-settings-for-your-organization/managing-allowed-ip-addresses-for-your-organization), what an IP allow list restricts, single addresses and CIDR ranges, the lockout warning and the advice on backup access
- [Google Workspace Admin Help: Protect your business with Context-Aware Access](https://knowledge.workspace.google.com/admin/security/protect-your-business-with-context-aware-access?hl=en), access levels by IP address range and by geographic origin
- [AWS: Security group rules](https://docs.aws.amazon.com/vpc/latest/userguide/security-group-rules.html), a single IPv4 address written with the /32 prefix length as the source of an inbound rule
- [RFC 2131: Dynamic Host Configuration Protocol](https://www.rfc-editor.org/rfc/rfc2131), dynamic allocation of an address for a limited period of time
- [RFC 6888: Common Requirements for Carrier-Grade NATs (CGNs)](https://www.rfc-editor.org/rfc/rfc6888), one public address shared by many subscribers
- [RFC 6269: Issues with IP Address Sharing](https://www.rfc-editor.org/rfc/rfc6269), section 13.2, why address-based access control lists fail when an address is shared
- [NIST SP 800-63B-4: Authentication and Authenticator Management (2025)](https://pages.nist.gov/800-63-4/sp800-63b.html), section 2, IP address and geolocation as fraud indicators that do not substitute for an authentication factor




An IP allowlist lets a system accept connections only from the addresses you
list. It is easy to set up when everyone works from one office and awkward when
nobody does. This guide is for a distributed team that has to reach systems it
runs or is authorised to use, such as a code host, an admin panel or a cloud
server, from one fixed address. ProxyJam sells [ISP proxies](/isp-proxies/),
one way to get that address, and the guide also covers the two other ways and
where each one fits. Our
[editorial policy](/editorial-policy/) explains how articles like this are
sourced and corrected.

## What does an IP allowlist check?

The source address of the connection that reaches it, and nothing about the
person behind it. Each of the three common places a team meets one works the
same way.

- **A code host.** GitHub's
  [IP allow list](https://docs.github.com/en/enterprise-cloud@latest/organizations/keeping-your-organization-secure/managing-security-settings-for-your-organization/managing-allowed-ip-addresses-for-your-organization)
  decides "whether users or apps can access protected resources through the
  web UI, APIs, or Git". It takes a single address or a range in CIDR notation,
  and it "applies to users with any role or access, including enterprise and
  organization owners". It is available to organisations on GitHub Enterprise
  Cloud.
- **Office software.** Google Workspace's
  [Context-Aware Access](https://knowledge.workspace.google.com/admin/security/protect-your-business-with-context-aware-access?hl=en)
  can build an access level from an IP condition, which "specifies an IP
  address range from which a user can connect to an app", and from the
  countries a user connects from.
- **A cloud server.** An AWS
  [security group rule](https://docs.aws.amazon.com/vpc/latest/userguide/security-group-rules.html)
  names the source it admits, and a single IPv4 address is written with the
  /32 prefix length, as in 203.0.113.1/32.

When a connection goes through a proxy, the system sees the proxy's address.
One address can therefore stand for a whole team, and the allowlist cannot
tell the members apart.

## Why do home and mobile connections fail an allowlist?

They do not hold one address that belongs to one subscriber. Home broadband
often gets its address by what the DHCP standard,
[RFC 2131](https://www.rfc-editor.org/rfc/rfc2131), calls dynamic allocation:
an address assigned "for a limited period of time". When it changes is up to
the provider. An entry that matched a member's address on Monday can stop
matching on Thursday, and the member is locked out until someone with admin
rights updates the list.

The workaround GitHub's documentation offers for this is a CIDR range wide
enough to "include dynamically assigned addresses". That keeps the member in,
and it also lets in every other subscriber of that provider whose address
falls in the range.

Mobile connections have the opposite problem. Carriers commonly put many
subscribers behind one public address through carrier-grade NAT
([RFC 6888](https://www.rfc-editor.org/rfc/rfc6888)), which our
[comparison of mobile and residential proxies](/blog/mobile-vs-residential-proxies/)
walks through. The IETF's review of address sharing,
[RFC 6269](https://www.rfc-editor.org/rfc/rfc6269), states in section 13.2: "Simple address-based identification mechanisms that are used to
populate access control lists will fail when an IP address is no longer
sufficient to identify a particular subscriber." An allowlist entry for a
phone's hotspot can admit strangers. Hotel and co-working networks add a third
case, with an address nobody on the team controls.

A team without an office therefore needs an address with two properties: it
stays the same, and nobody outside the team sends traffic from it.

## Office IP, VPN or a static ISP proxy: which fits a team without an office?

Any of the three can give a team one fixed address. They differ in what goes
through it and in what someone has to keep running, as the table Office line,
VPN or static proxy sets out.

### An office line

A business internet plan usually comes with a static address, and GitHub's
documentation uses exactly this case as its example: allowing access
"exclusively from the IP address of your office network". It works for the
people in the building. Remote members then need a way onto the office network,
which in practice is a VPN into it.

### A VPN with a fixed exit

A VPN sends a device's traffic, or the part a split tunnel selects, out through
a server, and the allowlist sees that server's address. You can run the server
yourself on a cloud machine with a fixed public address, or buy a business VPN
plan that includes a dedicated one. It is the choice when desktop apps, SSH
sessions or anything without a proxy setting must reach the allowlisted
systems. The cost is a client on every device and someone who runs the server
and its accounts.

### A static ISP proxy

A proxy carries only the traffic you send to it: a browser through an
extension, a script or a tool through its proxy setting. For a team whose
allowlisted systems are web apps, admin panels and APIs, that is usually
enough, and there is no server to run. Our addresses are on a named internet
provider's network, served from data-centre hardware, as the
[definition of ISP proxies](/blog/what-are-isp-proxies/) explains, and each one
stays with you for the period you pay for.

For the allowlist itself, the network type makes no difference. A cloud
server's datacenter address passes an allowlist exactly as an ISP address does,
and [residential against datacenter proxies](/blog/residential-vs-datacenter-proxies/)
explains where the difference does count. If a VPN server on a cloud machine is
something your team already runs, it may be all you need.

## How do you set up one address for the team?

Order one address, give every member the same connection details, route the
allowlisted systems through it, and only then add it to the allowlist. The
steps are listed under Give the team one address. Three choices in them are
worth explaining.

**Authenticate with the username and password.** Every address comes with its
own host, port, username and password. An order can instead admit clients by
an allowlist of their IP addresses, set through the API, and while that list is
set the password stops working
([integrations](/integrations/)). For people working from home that is the same
problem moved one step back: their changing addresses would lock them out of
the proxy. Keep the password for people, and use the order's allowlist only
for a server or a test runner that has a fixed address of its own.

**Route only the allowlisted systems.** In the browser,
[ProxyJam Connector](/browser-extension/) is a free extension for Chrome, Edge
and Firefox. Its Auto mode lets a rule send just the allowlisted system's
hostnames through the proxy while everything else leaves from the member's own
connection, the approach covered in
[using a proxy only for certain sites](/blog/chrome-proxy-for-specific-sites/).
This also keeps the proxy within its session cap. Each address carries up to
100 concurrent sessions, whatever the traffic
([Terms of Service](/terms-of-service/), section 2.4), and a
browser opens several connections at a time.

**Stay on HTTP in Chrome.** The order starts on HTTP, which carries HTTPS as a
tunnel. Chrome cannot send a username and password to a SOCKS5 proxy, a limit
explained in [the proxy sign-in prompt](/blog/chrome-proxy-password-prompt/),
so a team on Chrome keeps the proxy on HTTP. Switching the protocol also gives
the order a new port and new credentials, so settle it before you hand the
details out.

Scripts, CI jobs and other code that call an allowlisted API take the same
four values. The [integrations guides](/integrations/) show where Python
requests, Selenium and Playwright expect them.

Before you switch enforcement on, keep a way back in. GitHub warns that losing
access to the addresses on the list "could result in unintended consequences,
such as getting locked out of your enterprise or organization", and recommends
more than one owner and a static network kept on the list for backup access.
The list under Check these before the allowlist goes live covers the rest.

## What breaks an allowlist, and how do you keep the address?

An allowlist breaks when the address it names stops being the one your team
sends from. With a proxy, that happens in three ways.

- **The period ends.** The address is yours for the period you paid for, and
  renewing keeps the same address, as our [ISP proxies page](/isp-proxies/)
  states. Turn on Auto-extend for it in My Proxies, which renews it from your
  balance, or put the renewal date in someone's calendar. An address that is
  allowed to lapse takes the allowlist entry with it.
- **You move to a different address.** The address is not replaced during
  the period you paid for; a different one takes a new order and is assigned
  at random ([Terms of Service](/terms-of-service/), section 2.5). Add the new
  address before you remove the old one.
- **A location rule disagrees.** Workspace's access levels can also check the
  country a user connects from. That country comes from a geolocation database,
  and databases can place one address differently. If yours moves the address
  out of its country, the location rule can refuse it while the IP entry still
  matches.

Changes on the system's side take time too. GitHub notes that, because of
caching, adding or removing an address "can take a few minutes to fully take
effect", so do the swap while someone is watching.

Addresses are priced per IP per period, on the [pricing page](/pricing/), and a
longer period costs the same per month. What it buys is a renewal date further
away.

## Why must an IP allowlist never be the only control?

An allowlist says where a connection came from, and nothing about who made it.
Anyone holding the proxy's username and password sends from the team's
address. NIST's digital identity guideline,
[SP 800-63B-4](https://pages.nist.gov/800-63-4/sp800-63b.html), allows signals
such as "authentication from an unexpected geolocation or IP address block" to
lower the risk of a wrong sign-in, and states in section 2 that using them
"does not impact or change the AAL of a transaction or substitute for an
authentication factor".

In practice that means three habits.

- **One account per person.** Every member signs in to each allowlisted system
  as themselves, with two-factor authentication turned on. GitHub, Workspace
  and AWS all check the address on top of the sign-in.
- **Removal goes through accounts.** When someone leaves, removing their
  account is what cuts their access. The address stays the team's, and it would
  still admit a former member who kept the credentials.
- **The proxy credentials are a shared secret.** Keep them in the team's
  password manager, give them only to people who use the allowlisted systems,
  and leave them out of chat messages and code repositories.

## When is this the wrong setup?

The address is for systems your organisation owns or is authorised to use.
Section 1 of our [Acceptable Use Policy](/acceptable-use-policy/) permits
controlled network access for legitimate business purposes as long as it is
not used to "act against a system's access rules", and section 5 prohibits
accessing services in breach of their regional availability or platform
access rules. Getting onto another organisation's allowlist without its
agreement is outside both.
Three other cases call for something else.

- **Regulated systems.** If the allowlisted system belongs to financial
  services, healthcare or government, section 2 of the policy asks for
  verification before we open access to it.
- **People outside your organisation.** Section 6 forbids sharing or
  redistributing access to the service without our written permission, so a
  client or partner company needs its own arrangement.
- **Traffic outside the browser and your code.** Desktop apps, SSH and anything
  without a proxy setting need a VPN, or an office line.

For a team whose allowlisted systems live in a browser or behind an API, one
static address is where it starts. The [ISP proxies page](/isp-proxies/) lists
the catalogue, 76 countries with a named carrier behind every
address, and the [pricing page](/pricing/) shows what one address costs for
each period.

