Secure network: certificates, identity and PasspointAuthentication
Migrating from Aruba ClearPass to cloud RADIUS: policy export to cutover
You can plan a ClearPass to cloud RADIUS migration in five phases, moving policy before traffic so every wave is a reversible server change. You will be able to export and rebuild ClearPass services, run both RADIUS servers side by side, apply go or no-go criteria per wave, and retire the appliances with an...
16 min read

Watch the explainer
You can migrate from ClearPass to cloud RADIUS in five phases without a flag day when policy moves before traffic does. Export and rebuild the policy first. Then run both RADIUS servers side by side and move sites in small waves. With IEEE 802.1X settings and VLAN assignments unchanged, every wave is a reversible server change.
What changes when you move off ClearPass, and what stays the same?
The end state is short to describe. Your access points, SSIDs, VLANs and firewall zones stay where they are. The RADIUS server, its certificate and the policy engine change. Identity moves from ClearPass authentication sources to your cloud identity provider, and identity sets the VLAN.
Two terms first. IEEE 802.1X is the standard that holds a device off the network until an authentication server approves it. RADIUS (RFC 2865) is the protocol your access points use to ask that server. The answer can include the VLAN, the network segment the device joins.
| Element | On ClearPass today | After the move | Changes at cutover? |
|---|---|---|---|
| Access points and controllers | Cisco Meraki, HPE Aruba, Ruckus, Juniper Mist or Ubiquiti UniFi | The same hardware | No |
| SSIDs and VLANs | Existing names and VLAN IDs | Unchanged | No |
| EAP method | PEAP or EAP-TLS | The same method on cutover night | No |
| RADIUS server and shared secret | ClearPass appliances | Purple cloud RADIUS | Yes |
| Server certificate | Issued to ClearPass | Issued to the new RADIUS service | Yes, and this is the risk |
| Policy | ClearPass services and enforcement profiles | Rules keyed on identity provider groups | Yes, rebuilt in phase 2 |
The single risk that bites is the server certificate. During every EAP exchange, the device checks the certificate the RADIUS server presents. Managed laptops and cell phones usually trust ClearPass by issuing CA, by server name, or both. Point them at a server with a different certificate and they refuse or prompt. On a Monday morning, that looks exactly like an outage.
{"tone":"warning","title":"Push certificate trust before the first wave","text":"Devices that trust the ClearPass server certificate by name or issuing CA will reject or prompt on the new RADIUS certificate. Deploy the new trust profile through your MDM in phase 2, never on cutover night."}
What has to be true before phase 1 starts?
Do not book a cutover weekend until each of these is true. A gap in any one of them is enough to stall a wave.
- A complete device list. Export every network access device (NAD) from ClearPass: each controller, access point group and switch that sends RADIUS. Anything missing fails silently on the night.
- Identity in the cloud. The groups that drive ClearPass roles exist in Microsoft Entra ID, Okta or Google Workspace, with the same membership.
- A certificate plan. You know who issues device certificates for EAP-TLS, the certificate-based method defined in RFC 5216. You also know how your MDM, such as Microsoft Intune or Jamf, deploys trust for the new server.
- Outbound RADIUS from every site. Firewalls allow authentication and accounting traffic to the cloud service. The standard ports are UDP 1812 and 1813, with accounting defined in RFC 2866.
- A baseline. At least two weeks of ClearPass authentication counts, success and failure, per service and per site. Without it, you cannot prove the new system behaves the same.
- Dates and owners. Your ClearPass license and support end date, plus a change window and a named owner for each wave.
What are the five phases, in order?
The plan has five phases. Phases 1 and 2 touch no live traffic. Phase 3 runs both systems side by side at pilot sites. Phase 4 moves the rest of the estate in waves, and phase 5 retires ClearPass. Every phase has an exit test, and the next phase waits until it passes.
{"kind":"timeline","direction":"lr","caption":"A five-phase ClearPass migration for a 40-site estate. Set the length of each phase to fit your own estate.","alt":"Timeline of five phases from policy export through parallel running and wave cutover to ClearPass decommission, with a dashed rollback path from cutover back to parallel running.","nodes":[{"id":"p1","label":"Phase 1: export and map","sub":"Services, roles, enforcement profiles and NADs mapped to a target","tone":"neutral","when":"Weeks 1-2"},{"id":"p2","label":"Phase 2: rebuild and lab test","sub":"IdP groups, VLAN rules, certificates and trust profiles","tone":"neutral","when":"Weeks 3-4"},{"id":"p3","label":"Phase 3: parallel pilot","sub":"Pilot sites on cloud RADIUS, ClearPass kept as secondary","tone":"brand","when":"Weeks 5-6"},{"id":"p4","label":"Phase 4: wave cutover","sub":"One site group per weekend change window","tone":"brand","when":"Weeks 7-10"},{"id":"p5","label":"Phase 5: decommission","sub":"Zero ClearPass requests for 30 days, then retire","tone":"success","when":"Weeks 11-14"}],"edges":[{"from":"p1","to":"p2"},{"from":"p2","to":"p3"},{"from":"p3","to":"p4"},{"from":"p4","to":"p5"},{"from":"p4","to":"p3","label":"Roll back a site group","style":"dashed"}]}
{"caption":"ClearPass migration phase plan","asOf":"2026-06-10","columns":[{"key":"phase","label":"Phase"},{"key":"scope","label":"Scope"},{"key":"owner","label":"Owner"},{"key":"exit","label":"Exit test"}],"rows":[["1. Export and map","Every ClearPass service, role mapping, enforcement profile and NAD","Network architect","Every service marked rebuild, replace or retire, with a target network and VLAN"],["2. Rebuild and lab test","Identity provider groups, VLAN rules, certificates and MDM trust profiles","Identity and access lead","Each role authenticates on a lab SSID per vendor and lands on its VLAN with no prompt"],["3. Parallel pilot","Two or three pilot sites with cloud RADIUS primary and ClearPass secondary","IT manager","Pilot success rate matches the ClearPass baseline for the full window"],["4. Wave cutover","One site group per weekend change window","IT manager","Go criteria met before the wave; baseline held for one business day after"],["5. Decommission","ClearPass appliances, licenses, NAD entries and old trust profiles","Network architect","Zero ClearPass requests for 30 days and an evidence pack filed"]],"footnote":"Owners are typical roles. Assign named people before phase 1.","download":{"filename":"clearpass-migration-phase-plan.csv"}}
Phases 1 and 2: how do you export ClearPass policy and rebuild it?
Phase 1: export and map every service
In ClearPass Policy Manager, policy lives in services. Each service matches incoming requests and checks them against authentication sources. It then maps the result to a role and applies an enforcement policy. Enforcement profiles return the RADIUS attributes that place the device, usually a VLAN or an Aruba-User-Role.
Export the services, role mappings, enforcement policies, enforcement profiles and NAD list using ClearPass's own configuration export. Then build a mapping sheet with one row per service. Record what the service matches, which identity source it checks, which role it assigns and which VLAN results.
Give every row one of three decisions: rebuild, replace or retire. Expect retirements. Older estates often carry services for SSIDs that no longer broadcast. Each retired service is one less rule to test.
The table below shows where each ClearPass function lands across the three networks Purple runs on your access points: open, secure and xPSK.
{"title":"Where each ClearPass function lands","tabs":[{"label":"Staff 802.1X services","body":"Rebuild on the **secure network**. Each ClearPass role becomes a rule keyed on an identity provider group that returns a VLAN.","bullets":["Match on Microsoft Entra ID, Okta or Google Workspace groups","Return the VLAN with the RFC 3580 tunnel attributes","Keep the EAP method unchanged for cutover night"]},{"label":"Guest access and captive portal","body":"Replace with the **open network**. A captive portal is the web page a visitor sees before getting online.","bullets":["Self-registration and sponsored access move to the open network","Guest marketing runs in Purple Engage","BYOD onboarding starts on the open network"]},{"label":"MAC auth and headless devices","body":"Replace with the **xPSK network**. iPSK and PPSK give each device or resident its own key, and the key sets the VLAN.","bullets":["Scanners, printers and sensors get a key each","Revoke one key without touching the rest","No MAC address lists to maintain"]},{"label":"Onboard, posture and device admin","body":"Decide per feature. These are the services teams most often forget.","bullets":["ClearPass Onboard certificate issuance moves to your MDM or PKI","OnGuard posture checks move to MDM compliance policy","TACACS+ device administration and wired 802.1X need their own plan"]}]}
Phase 2: rebuild identity and policy, then prove it in a lab
Connect your identity provider first, then rebuild each "rebuild" row as a rule. Return VLANs with the standard attributes in RFC 3580: Tunnel-Type, Tunnel-Medium-Type and Tunnel-Private-Group-ID. The same rule then works on every vendor in your estate.
If HPE Aruba controllers enforce Aruba-User-Role today, decide now how that role survives. Either keep role enforcement on the controller, or replace it with VLAN by identity. Confirm the attribute behavior with Purple before lab testing starts.
Check your EAP method too. PEAP (Protected EAP) wraps a password check inside a TLS tunnel. If ClearPass checks PEAP passwords against on-premises Active Directory, treat the move to EAP-TLS as its own piece of work. Cloud identity providers such as Microsoft Entra ID do not offer the MSCHAPv2 password check that PEAP depends on. Deploy certificates and trust profiles in this phase, so cutover night changes the server and nothing else.
The exit test is a lab SSID on one access point per vendor you run. For each role on the mapping sheet, a test device authenticates, lands on the expected VLAN and shows no prompt.
How do ClearPass and cloud RADIUS run side by side through cutover?
Phase 3: parallel running at pilot sites
Pick two or three pilot sites with a forgiving population, such as a corporate headquarters floor or an IT-heavy site. Make Purple cloud RADIUS the primary server on the production SSID. Keep ClearPass as the secondary.
Know what the secondary does and does not give you. Access points fail over to the secondary when the primary stops answering. They do not fail over when the primary rejects a request. A secondary ClearPass protects you against an outage, not against a policy gap.
The parallel window is therefore a comparison exercise. Measure pilot authentication success against the ClearPass baseline, per role and per site. Confirm each role lands on its VLAN. Watch helpdesk tickets for certificate prompts.
Phase 4: cutover in waves
Each wave is a server-order change per site group, made in the vendor console inside a change window. RADIUS migration is usually a weekend exercise, so plan one site group per weekend. Keep the first wave small, then size later waves to what your team can test in one window.
{"title":"Cut one site group over to cloud RADIUS","steps":[{"title":"Confirm the go criteria","body":"Run the go or no-go check for this site group and note the ClearPass baseline for these sites.","check":"The wave owner signs off every go criterion."},{"title":"Add cloud RADIUS to the SSID","body":"Add Purple cloud RADIUS as an authentication and accounting server on the production SSID, using the server details and shared secret from the Purple console. Keep ClearPass in the list.","check":"The console shows both servers on the SSID, with ClearPass still listed.","variants":[{"label":"Cisco Meraki","body":"In the Meraki Dashboard, the SSID uses **Enterprise with my RADIUS server**. Add the Purple servers to the RADIUS and accounting server lists."},{"label":"HPE Aruba","body":"In Aruba Central or on the Mobility Conductor, add the Purple servers to the authentication server group and accounting list the WLAN uses."},{"label":"Ruckus","body":"In SmartZone or RUCKUS One, create authentication and accounting services for Purple and attach them to the WLAN."},{"label":"Juniper Mist","body":"In the Mist portal, add the Purple servers to the RADIUS authentication and accounting lists on the WLAN with **Enterprise (802.1X)** security."},{"label":"Ubiquiti UniFi","body":"In the UniFi Network application, create a RADIUS profile for Purple and select it on the WPA Enterprise network."}]},{"title":"Test before you reorder","body":"From a test laptop on site, run eapol_test against the Purple server for each role on the mapping sheet. Then join one real device per role.","code":{"lang":"bash","text":"eapol_test -c staff-role.conf -a <purple-radius-address> -s <shared-secret>"},"check":"eapol_test reports SUCCESS and the returned VLAN matches the mapping sheet."},{"title":"Make Purple primary","body":"Reorder the servers so Purple answers first and ClearPass sits second. Change one site group at a time.","check":"New sessions appear in the Purple console and ClearPass requests from these sites stop."},{"title":"Hold for one business day","body":"Compare authentication counts, VLAN placement and helpdesk tickets with the baseline. Leave ClearPass as secondary until phase 5.","check":"Counts match the baseline and no role lands on an unexpected VLAN."}]}
Run this check before every wave, not just the first.
{"title":"Go or no-go for the next cutover wave","start":"q1","nodes":[{"id":"q1","question":"Do pilot sites match the ClearPass baseline for authentication success?","help":"Compare per site and per role across the whole parallel window.","options":[{"label":"Yes","to":"q2"},{"label":"No","to":"o1"}]},{"id":"q2","question":"Does every role land on its expected VLAN?","help":"Check one test device per role against the mapping sheet.","options":[{"label":"Yes","to":"q3"},{"label":"No","to":"o2"}]},{"id":"q3","question":"Do managed devices trust the new server certificate without a prompt?","help":"Test one device per operating system and MDM profile.","options":[{"label":"Yes","to":"q4"},{"label":"No","to":"o3"}]},{"id":"q4","question":"Is ClearPass intact as secondary, with NAD entries and shared secrets unchanged?","help":"Rollback depends on it.","options":[{"label":"Yes","to":"o4"},{"label":"No","to":"o5"}]},{"id":"o1","outcome":"No-go: find the policy gap","detail":"Compare failing requests against the mapping sheet. A missed service is the usual cause."},{"id":"o2","outcome":"No-go: fix the VLAN attributes","detail":"Check the RFC 3580 tunnel attributes on the rule and any controller role mapping."},{"id":"o3","outcome":"No-go: push trust first","detail":"Deploy the new trust profile through your MDM and retest before the window."},{"id":"o4","outcome":"Go: cut the wave","detail":"Make Purple primary for this site group and hold for one business day.","link":"[Compare with a Cisco ISE cutover](guide:migrate-from-cisco-ise)"},{"id":"o5","outcome":"No-go: restore the fallback","detail":"Reinstate ClearPass as secondary before any site moves."}]}
Worked scenario: a 40-store retail estate on Cisco Meraki
Situation. Forty stores on Cisco Meraki run staff WiFi through two ClearPass virtual appliances. Staff laptops use PEAP against on-premises Active Directory. Handheld scanners join through MAC authentication. The ClearPass renewal falls in five months.
What was done. Phases 1 and 2 take four weeks. Scanners move to the xPSK network store by store, one key per device. Microsoft Intune issues EAP-TLS certificates to staff laptops, with the new trust profile alongside. Two pilot stores run for two weeks, then three waves of 12, 12 and 14 stores follow.
Outcome. All 40 stores reach cloud RADIUS across four change weekends. One store rolls back on the night because its firewall blocks outbound RADIUS, then cuts over the following weekend. Both ClearPass appliances retire before the renewal date.
How do you roll back a wave?
Rollback is the cutover in reverse: put ClearPass back as primary for the affected site group. That only works while ClearPass is fully intact. Until phase 5 closes, hold these four lines:
- Keep ClearPass licenses and support live.
- Leave every NAD entry and shared secret on ClearPass unchanged.
- Keep ClearPass listed as the secondary server on every SSID.
- Keep the old trust profile on devices alongside the new one, so they trust both servers.
Agree on the rollback triggers before the window opens. Typical triggers are a role landing on the wrong VLAN, failures rising above your baseline, or a device class prompting for certificate trust. Roll back the affected site group, not the whole wave, wherever the console allows it.
Worked scenario: a 25-hotel group on HPE Aruba
Situation. A 25-hotel group runs staff WiFi on HPE Aruba with ClearPass. Guest access sits on a separate open network with its own captive portal. Wave two covers six hotels in one weekend.
What was done. At one hotel, back-office PCs land on the general staff VLAN instead of the finance VLAN. ClearPass had placed them with an Aruba-User-Role, and the rebuilt rule had no equivalent. The team reorders servers for that hotel's access point group only, inside the same change window. The other five hotels stay on cloud RADIUS.
Outcome. The fix is a VLAN rule keyed on the finance identity group. The hotel cuts over the next weekend. The group completes all 25 hotels in five weekends, with one site-level rollback and no change to guest access.
How do you know the move is finished?
The move is finished when ClearPass can be switched off and nobody notices. Phase 5 proves that with evidence, not a feeling. Close it when all of these hold:
- ClearPass logs show zero authentication requests for 30 consecutive days.
- Every SSID lists Purple only, with ClearPass removed as secondary.
- Every role on the mapping sheet is verified on its VLAN at one site per vendor.
- Authentication success per site sits at or above the ClearPass baseline.
- WiFi helpdesk tickets are back to pre-migration levels.
- An evidence pack is filed: the ClearPass export, the rebuilt rule set, change records and authentication logs.
- ClearPass appliances are wiped, licenses are ended and the old trust profile is removed from devices.
The evidence pack matters beyond the project. It is evidence for the access control and logging requirements in PCI DSS and ISO 27001 audits, where an assessor asks who changed network access and when.
{"title":"ClearPass migration control sheet","groups":[{"heading":"Before phase 1","items":[{"text":"NAD list exported from ClearPass, every controller and access point group included"},{"text":"Identity provider groups match ClearPass roles","why":"Rules key on these groups in phase 2"},{"text":"Certificate issuer and MDM trust deployment agreed"},{"text":"Outbound RADIUS allowed from every site firewall"},{"text":"Two weeks of ClearPass baseline captured per service and site"}]},{"heading":"Policy and identity","items":[{"text":"Every service marked rebuild, replace or retire"},{"text":"VLAN rules return RFC 3580 tunnel attributes"},{"text":"Aruba-User-Role dependencies resolved"},{"text":"PEAP to EAP-TLS move completed where Active Directory is retired"},{"text":"New server trust profile live on managed devices"}]},{"heading":"Parallel running and cutover","items":[{"text":"Pilot sites match the baseline for the full window"},{"text":"Rollback triggers agreed and written into each change record"},{"text":"ClearPass kept as secondary on every SSID"},{"text":"eapol_test passed per role before each reorder"},{"text":"Each wave held for one business day before the next"}]},{"heading":"Decommission and proof","items":[{"text":"Zero ClearPass requests for 30 consecutive days"},{"text":"ClearPass removed from every SSID server list"},{"text":"Evidence pack filed for audit"},{"text":"Old trust profile removed from devices"},{"text":"ClearPass appliances wiped and licenses ended"}]}],"download":{"filename":"clearpass-migration-control-sheet.csv"}}
Which failure modes show up in the field?
Certificate prompts on unmanaged devices
BYOD cell phones outside your MDM never receive the new trust profile. Route them through onboarding on the open network before their site cuts over, or expect calls on Monday.
Failover that hides nothing
Teams assume ClearPass as secondary catches every problem. It only catches timeouts. A reject from the new server reaches the device as a failure, which is why the parallel window compares counts rather than waiting for complaints.
The forgotten NAD
One controller pair or access point group sends RADIUS from an address nobody listed. Its sites fail while every other site works. The fix is the phase 1 export, cross-checked against each vendor console.
Accounting left behind
Authentication moves but accounting still points at ClearPass. Sessions work, yet logs for the audit trail go missing. Move both together in the same step.
Scope creep from wired and device admin
ClearPass often also runs wired 802.1X and TACACS+ for switch administration. Leave them out of a WiFi wave and give them their own dated plan, so they do not hold up phase 5.
{"tone":"note","title":"Scope of this guide","text":"This plan covers WiFi authentication on the open, secure and xPSK networks. Wired 802.1X and TACACS+ device administration need a separate plan before ClearPass can retire."}
Where does Purple fit in a ClearPass migration?
Purple Access runs the target of this plan: cloud RADIUS on the secure network, with EAP-TLS, identity provider integration and Passpoint. Passpoint is the Hotspot 2.0 standard that lets a device join automatically and securely, without a captive portal. Purple runs the largest cloud RADIUS in the world.
The platform is hardware-agnostic. It works as a cloud overlay on Cisco Meraki, HPE Aruba, Ruckus, Juniper Mist, Ubiquiti UniFi, Cambium, Extreme and Fortinet. Cutover is a server change in the console you already use, with no new hardware. Identity comes from Microsoft Entra ID, Okta or Google Workspace, so joiners, movers and leavers update network access from one source.
{"items":[{"value":"99.9%","label":"RADIUS uptime SLA","note":"Purple's own SLA"},{"value":"99.999%","label":"platform uptime","note":"Purple's own data"},{"value":"80,000+","label":"live venues","note":"Purple's own data"},{"value":"90","label":"countries","note":"Purple's own data"}]}
{"tone":"proof","title":"Staff WiFi at bank scale","text":"JPMorgan runs staff WiFi on Purple across 5,000 branches. Purple runs the largest cloud RADIUS in the world."}
The open and xPSK networks absorb the rest of the ClearPass footprint. Guest marketing on the open network runs in Purple Engage, and DNS filtering runs in Shield. Purple holds ISO 27001, Cyber Essentials and B Corp certification, and supports CCPA/CPRA obligations.
For the wider picture, read the secure network guide and the enterprise WiFi security guide. If you are still choosing a target, see ClearPass alternatives.
Frequently asked questions
How long does a ClearPass to cloud RADIUS migration take?
The cutover is usually a weekend per wave, while the preparation sets the calendar. RADIUS migration is usually a weekend exercise, so a 40-site estate in four waves needs four change weekends. Most of the elapsed time sits in phases 1 and 2: exporting and mapping services, then deploying certificate trust. Phase 5 adds a 30-day wait for zero ClearPass requests before you retire the appliances.
Do I need new access points to move off ClearPass?
No. Purple is hardware-agnostic and works as a cloud overlay on Cisco Meraki, HPE Aruba, Ruckus, Juniper Mist, Ubiquiti UniFi, Cambium, Extreme and Fortinet. The change on each SSID is the RADIUS server list and shared secret, made in the console you already use. Your SSIDs, VLANs and EAP method stay the same on cutover night.
Will staff have to re-enroll their devices?
Not if you deploy the new server trust through your MDM before cutover. Devices already on EAP-TLS keep working where your PKI continues to issue their certificates. Devices on PEAP against on-premises Active Directory need EAP-TLS certificates first, because cloud identity providers do not offer the MSCHAPv2 check. Unmanaged BYOD devices rejoin through onboarding on the open network.
Can we keep ClearPass running while we migrate?
Yes, and you should, because rollback depends on it. Keep ClearPass licensed, with every NAD entry and shared secret unchanged, as the secondary RADIUS server until phase 5. It covers you if the primary stops answering. Remove it only after 30 consecutive days with zero ClearPass requests, then end the license.
What replaces ClearPass Guest and MAC authentication?
The open network replaces ClearPass Guest, and the xPSK network replaces MAC authentication. Guest self-registration and sponsored access move to a captive portal on the open network, with guest marketing in Purple Engage. Headless devices such as scanners and printers each get their own iPSK or PPSK key. The key sets the VLAN and can be revoked alone.
How does the migration support PCI-DSS and ISO 27001 audits?
It produces evidence for both, provided you file the trail. Keep the ClearPass export, the rebuilt rule set, change records for each wave and the authentication logs from cloud RADIUS. Together they show who could reach which VLAN and when that changed. Purple itself holds ISO 27001 certification, which covers the service you are moving to.
How does this compare with migrating from Cisco ISE?
The shape is the same: export, rebuild, parallel pilot, wave cutover and decommission. The differences sit in the export format and the vendor attributes your rules must replace, such as Aruba-User-Role here. If your estate runs both platforms, plan one identity model and two exports. The Cisco ISE migration guide covers the ISE specifics.
What does it cost to run ClearPass and Purple in parallel?
The overlap costs no more than your existing ClearPass license if you plan around its renewal date. Work back from that date: allow for phases 1 to 4, then 30 days of zero requests in phase 5. Starting early enough avoids renewing ClearPass for a term you will barely use.
Frequently asked questions
How long does a ClearPass to cloud RADIUS migration take?
The cutover is usually a weekend per wave, while the preparation sets the calendar. RADIUS migration is usually a weekend exercise, so a 40-site estate in four waves needs four change weekends. Most of the elapsed time sits in phases 1 and 2: exporting and mapping services, then deploying certificate trust. Phase 5 adds a 30-day wait for zero ClearPass requests before you retire the appliances.
Do I need new access points to move off ClearPass?
No. Purple is hardware-agnostic and works as a cloud overlay on Cisco Meraki, HPE Aruba, Ruckus, Juniper Mist, Ubiquiti UniFi, Cambium, Extreme and Fortinet. The change on each SSID is the RADIUS server list and shared secret, made in the console you already use. Your SSIDs, VLANs and EAP method stay the same on cutover night.
Will staff have to re-enroll their devices?
Not if you deploy the new server trust through your MDM before cutover. Devices already on EAP-TLS keep working where your PKI continues to issue their certificates. Devices on PEAP against on-premises Active Directory need EAP-TLS certificates first, because cloud identity providers do not offer the MSCHAPv2 check. Unmanaged BYOD devices rejoin through onboarding on the open network.
Can we keep ClearPass running while we migrate?
Yes, and you should, because rollback depends on it. Keep ClearPass licensed, with every NAD entry and shared secret unchanged, as the secondary RADIUS server until phase 5. It covers you if the primary stops answering. Remove it only after 30 consecutive days with zero ClearPass requests, then end the license.
What replaces ClearPass Guest and MAC authentication?
The open network replaces ClearPass Guest, and the xPSK network replaces MAC authentication. Guest self-registration and sponsored access move to a captive portal on the open network, with guest marketing in Purple Engage. Headless devices such as scanners and printers each get their own iPSK or PPSK key. The key sets the VLAN and can be revoked alone.
How does the migration support PCI DSS and ISO 27001 audits?
It produces evidence for both, provided you file the trail. Keep the ClearPass export, the rebuilt rule set, change records for each wave and the authentication logs from cloud RADIUS. Together they show who could reach which VLAN and when that changed. Purple itself holds ISO 27001 certification, which covers the service you are moving to.
How does this compare with migrating from Cisco ISE?
The shape is the same: export, rebuild, parallel pilot, wave cutover and decommission. The differences sit in the export format and the vendor attributes your rules must replace, such as Aruba-User-Role here. If your estate runs both platforms, plan one identity model and two exports. The [Cisco ISE migration guide](guide:migrate-from-cisco-ise) covers the ISE specifics.
What does it cost to run ClearPass and Purple in parallel?
The overlap costs no more than your existing ClearPass license if you plan around its renewal date. Work back from that date: allow for phases 1 to 4, then 30 days of zero requests in phase 5. Starting early enough avoids renewing ClearPass for a term you will barely use.