Describe the bug
Describe what your problem is.
We had Zscaler installed on our machines, and since then, Multipass has stopped working due to network connectivity issues. When launching a Multipass VM, it starts successfully but has no outbound internet access — pinging external addresses (e.g., 8.8.8.8) results in 100% packet loss. Although the DNS search domains are present, traffic from the VM is not being routed to the internet. This appears to be a routing conflict between Zscaler and the Multipass VM’s network layer.
We rely on Multipass for our local development environment, so this issue is blocking our workflow.
As a workaround, we shut down the VM, change its network adapter to Bridged Mode, and then restart it, but of no use.
Raised the issue with zscaler
This is their response
Issue: ZCC running on macOS using Multipass does not allow the guest VM to connect to the internet.
During the call, we collected ZCC logs to investigate the issue. After reviewing the packet captures, I observed that a RST is being sent from the client on port 9093. In the PCAP where the first RST appears, the packet dump shows the interface as utun4 and the IP as 192.168.2.56. The capture indicates utun4 (the ZCC virtual interface), but we do not see corresponding entries in our tunnel logs.
Specifically:
In packet 9967, a SYN is sent to port 80 of an internet server.
In packet 9968, the same traffic appears to be redirected to port 9093 on the local default interface, after swapping the source and destination IP addresses.
This behavior suggests the traffic is being incorrectly handled by Multipass networking. Since there is no listener on port 9093 on the local interface, the host responds with a RST, which causes the connection to drop.
Could you please check with the Multipass team regarding this networking issue?
To Reproduce
How, and what happened?
- Start a multipass vm with Zscaler installed.
- ssh to VM and ping google or do a apt update
Expected behavior
What did you expect to happen?
Logs
Please provide logs from the daemon, see accessing logs on where to find them on your platform.
Additional info
- OS: [e.g. macOS Tahoe 26.3]
- CPU architecture or model: [Apple M4 Pro]
multipass version
❯ multipass --version
multipass 1.16.1+mac
multipassd 1.16.1+mac
multipass info
multipass get local.driver
❯ multipass get local.driver
qemu
Additional context
Add any other context about the problem here.
Describe the bug
Describe what your problem is.
We had Zscaler installed on our machines, and since then, Multipass has stopped working due to network connectivity issues. When launching a Multipass VM, it starts successfully but has no outbound internet access — pinging external addresses (e.g., 8.8.8.8) results in 100% packet loss. Although the DNS search domains are present, traffic from the VM is not being routed to the internet. This appears to be a routing conflict between Zscaler and the Multipass VM’s network layer.
We rely on Multipass for our local development environment, so this issue is blocking our workflow.
As a workaround, we shut down the VM, change its network adapter to Bridged Mode, and then restart it, but of no use.
Raised the issue with zscaler
This is their response
To Reproduce
How, and what happened?
Expected behavior
What did you expect to happen?
Logs
Please provide logs from the daemon, see accessing logs on where to find them on your platform.
Additional info
multipass versionmultipass infomultipass get local.driverAdditional context
Add any other context about the problem here.