AOS-241083

AOS-242823

AOS-249257

A few AP-635 and AP-655 access points running AOS-10.5.0.0 or later versions crashed and rebooted unexpectedly. The log files listed the reason for this event as Kernel panic - not syncing: Take care of the TARGET ASSERT first. The crash information showed that it was asserted at ar_wal_tx_de.c:342.

The fix ensures that the APs work as expected.

AOS-241517

Some gateways added NAT flags to the network traffic between the managed device and gateway, instead of sending that traffic over the IPsec tunnel to the VPNC. This caused IPsec connection failure between the managed devices and gateways. This issue occurred because the NAT flags were not cleared when the IPsec sessions were detected.

This issue was resolved by deleting the current session and dropping the packet if the session is IPSec and NAT is enabled.

AOS-242047

AOS-245896

AOS-246137

AOS-247516

A few bytes of TXT record length were truncated in the response sent by the APs running AOS-10.4.0.1 or later versions. This issue occurred because the incoming TXT record length was greater than the supported TXT record limit. As a result, when the message was decoded, the HPE Aruba Networking Central cache message record was dropped. This led to an incorrect TXT record count in the HPE Aruba Networking Central cache table.

This issue was resolved by correcting the TXT record length in the code.

AOS-243275

AOS-245329

The resolvwrap process crashed continuously when a VLAN that is set to dhcp-client failed to get an IP. This issue was observed in gateways running AOS-10.4.1.0 or later versions.

The fix ensures that the resolvwrap process does not crash.

AOS-244218

AOS-252507

AOS-253463

A few AP-635 access points running AOS-10.5.0.1 crashed and rebooted. The log files listed the reason as kernel panic: Take care of the TARGET ASSERT first (ar_wal_tx_send_selfgen.c:4018 ). This issue occurred due to a memory allocation failure for the trigger frame.

The fix ensures that APs work as expected without crashing or rebooting.

AOS-244436

Some APs running AOS-10.4.0.0 or later versions configured with a valid BSSID were incorrectly identified as rogue APs. IDS incorrectly reported AP impersonation events for valid BSSIDs.

The fix ensures no false positives are reported by IDS.

AOS-245014

AOS-249855

The Aruba 7000 Series and 9000 Series gateways running AOS-10.4.0.0 crashed and rebooted unexpectedly. The log files listed the reasons for this event as:

• Reboot Cause: Datapath timeout (Heartbeat Initiated)

• Reboot Cause: Datapath timeout (SOS Assert) (Intent:cause:register 53:86:50:2)

This issue occurred because the Forward Error Correction (FEC) parity buffer became NULL and was accessed by a certainForwarding Processor (FP), which led to the code crash.

The issue is resolved by adding a defensive check to verify whether the parity buffer is NULL before it is used by the FP.

AOS-245327

AOS-247166

Radproxy initialization and binding errors were recorded in the error log at an interval of every 5 seconds. This issue was seen in a few VGWs configured as VPNCs running AOS-10.5.0.0.

The fix ensures that the HPE Aruba Networking-unmanaged VPNCs work as expected.

AOS-245620

AOS-245804

A few AP-635 access points running AOS-10.4.0.0 incorrectly resolved the device connectivity URL to the firewall IP address. This issue occurred because the AP cached the DNS query result into the hosts file.

The issue was resolved by removing the cache entry before the AP connected to HPE Aruba Networking Central.

AOS-245621

APs running AOS-10.4.0.1 did not switch channels to avoid interference, causing performance issues and network drops.

The fix ensures that the APs switch channels when interference is detected in channels other than the main channel.

AOS-245899

In HPE Aruba Networking Central, AI insights incorrectly displayed high CPU usage for a few AP-615 access points running AOS-10.5.0.0 or later versions. The use of smaller data type variables to store large data in the CPU utilization algorithm resulted in data overflow.

This issue was resolved by changing the code to fix the variable type.

AOS-246076

Due to GRE tunnel MTU flapping, MTU update failed which caused crash in the tunnelmgr process. This issue was observed on gateway running AOS-10.4.0.0.

The fix ensures that MTU update is done successfully.

AOS-246229

Some AP-303H access points running AOS-10.4.0.2 were unable to form a stable IPsec tunnel with the primary VPN server. This issue occurred because the static route update was published to L3d for all tunnels including IAP-VPN tunnels.

The fix ensures that static routes are not updated for IAP-VPN maps when L3d request replay all.

AOS-246617

A few AP-515 access points running AOS-10.4.0.0 or later versions crashed and rebooted unexpectedly. The log files listed the reason for the event as Kernal Panic: Ktrace core monitor: cpu3 hung for 45 seconds, hung cpu count:1 Warm-reset.

The issue was resolved by unlocking a spin lock before returning the invalid fragmented entry index.

AOS-246643

Underlay users were not allowed over the port channel. This issue was observed in a AP-gateway topology with gateways running AOS-10.4.1.0. When underlay users hit the gateway as wired users, it resulted in a bridge miss leading to user not getting connected to the network.

The fix ensures that underlay users can connect to the network.

AOS-246735

APs running AOS-10.4.1.0 crashed and rebooted unexpectedly. The log files listed the reason as PC: memcmp+0xd0/0x1c0 (from sos_uquic_analyse_frame+0x30). The fix ensures that the APs work as expected.

AOS-246845

Auth process stopped responding after the gateway was migrated from AOS 8x to AOS-10.4.0.2. Some netdestination configurations were received at the gateway after the auth process had already started. This eventually caused the auth process to go into a loop.

The fix ensures that the netdestination configurations are sent correctly.

AOS-247675

A few AP-515 access points running AOS-10.4.0.0 or later versions experienced intermittent Out-Of-Service (OOS) mode after booting up. This issue occurred because the uplink status was already UP when the AP selected the uplink for the first time. As a result, the OOS was not notified to trigger an event.

This issue was resolved by a code change to ensure the OOS is notified to trigger an event if the status of the uplink is UP.

AOS-247727

Netdestination whitelist was not working on branch gateways running AOS-10.4.0.0 or later versions. This issue occurred because the DNS IP list was not cleared every 24 hours for x86 platforms. This resulted in the DNS IP table getting full and subsequently caused DNS IP allocation failures.

The fix ensures that the DNS IP list is cleared at regular intervals.

AOS-247778

A few APs failed in the re-enrollment process with an EST-configured CCPM server. Re-enrolling failed even when a username and password were configured in the EST profile. The issue was seen on access points running AOS-10.4.1.0 or later versions.

The fix ensures that re-enrollment with EST CPPM servers works as expected.

AOS-248658

When the Meridian app was used on IoT Operations in AOS-10.5.x deployments, most of the Meridian tag reports from AP BLE beacons to the Meridian asset tracking engine were being dropped.

This issue was resolved by increasing the BLE relay limit.

AOS-249449

AOS-249943

The auth process crashed on HPE Aruba Networking9240 gateways running AOS-10.4.0.0 and later versions. This issue caused a cluster split and the gateways dropped off from the cluster, resulting in a network disconnection.

The fix ensures that the gateways work as expected and the client is able to access the network.

AOS-249970

The authentication module crashed repeatedly on a few HPE Aruba Networking9240 gateways running AOS-10.4.0.0 or later versions. This issue occurred due to a coding error with MAC address that was not validated.

The fix ensures that the authentication module in the gateway works as expected.