Outbound
Manage the nodes Tunna can use for proxy traffic.
Updated
An outbound is a node Tunna can connect through. It may be a local node you entered yourself or a node delivered by a subscription. The selected outbound is the node used whenever Routing sends traffic to Proxy, and the outbound list is also where Tunna shows live health signals such as latency, recent checks, and node stats.
When the list is empty
Use the empty-state actions to create a manual node, scan a QR code, paste a provider link, or add a subscription. After a subscription fetches successfully, its nodes appear in Outbound as a group.
What Outbound is for
Add or import
Create a node manually, scan a QR code, paste a provider link, or receive one through Resources.
Select
Tap a node to make it current. Proxy traffic uses that selected node.
Check
Latency checks help sorting, subscription visibility, Top 10 views, and ZAP.
Share
Share a local node as a link or QR code. Treat the result as private credentials.
Using the list
Tap a node row to select it. A selected-node badge marks the node that Proxy routing will use. Tap the latency number or latency dots to open Stats. Local nodes have an info button for viewing and editing the node. Swipe local rows for Delete or Share, and swipe subscription rows for the available Delete, Share, or Update actions. The plus menu offers Node, Scan, and Paste.
Manual node edits stay account-shaped
Manual edit views use the same Xray Core terminology as imported profiles, so they stay concise. Copy provider values exactly, keep imported transport and security fields unless you know the server setup, and use Add or Save after the required address, port, and credential fields are valid.
Rules can use different nodes at the same time
Tunna is not limited to one global proxy choice. Routing rules can target Proxy, which means the currently selected node, or a specific proxy node. Use specific-node rules when one destination should use a different provider node from the rest of your traffic.
Local node or subscription
Local node
A single server profile that you own and maintain inside Tunna.
Good fit You manage one server, received one provider link, or need to adjust one profile carefully.
Not ideal Your provider manages many nodes and changes the list over time.
Subscription
A provider profile that can contain many nodes plus usage, expiry, profile, and support data.
Good fit You want Tunna to refresh a provider-managed node list and choose from a group.
Not ideal You do not trust the profile source or only need one fixed server.
Subscription groups can change after refresh
When a subscription updates, Tunna rebuilds that provider group from the fetched node links. Matching nodes keep stable identity where possible, new nodes are added, and removed provider nodes disappear from the group.
Reading a subscription group
A folded subscription row summarizes the group without showing every node. It can show status dots, a fetching dot while Tunna refreshes, a health donut and count bar for node colors, the selected-node badge when the current node belongs to that group, expiry, and usage when the provider sends quota data. Tap the subscription row to unfold it and choose individual nodes.
List signals at a glance
Health donut
Shows how many nodes in a subscription group are fast, slow, failing, or unknown.
Count bar
Turns the same health mix into a compact bar so large subscriptions can be scanned quickly.
Status and fetch dots
Subscription dots separate normal status from active fetching, so you can tell whether a provider row is stale, failed, or being updated.
Selected badge
The selected-node badge follows the node Tunna will use for Proxy traffic, even when it sits inside a folded subscription group.
A healthy node lifecycle
Latency, loss, data, and stats in plain language
Latency is Tunna's quick health check for a node. Think of it as asking, How fast does this node answer? Smaller numbers usually feel better.
The number
A latency value is a round-trip time: Tunna sends a small check to the node and measures how long the answer takes to come back.
Lower is better
A smaller time usually means the node is quicker to respond.
Average
The main number is a recent average, so one odd check does not tell the whole story.
Plus or minus
The value after the plus-minus sign shows how jumpy the recent checks are. A low, steady node is usually nicer than a low but jumpy one.
The dots
Dots are recent checks. They let you scan whether a node has been healthy, slow, or failing without opening a chart.
Green, yellow, orange, red
Green is fast, yellow is slower, orange is high, and red is a failed check. A failed dot is red immediately; it does not wait for several failures.
Gray placeholder
Gray appears only when Tunna has no recent dots to draw yet.
Probing and Timeout
Probing means Tunna is still checking or has no successful sample yet. Timeout means three recent checks failed in a row.
Opening stats
Tap the latency number or dots on a node row to open that node's Stats view. The bottom picker changes the chart range between Hour, Day, Month, and Year.
Latency chart
Latency is the wait. Lower points mean the node answers quickly; tall spikes mean slow moments.
Loss chart
Loss is missed answers. More loss means Tunna asked the node for a check and did not get a useful answer often enough.
Data Usage
Data Usage is the traffic counter. Sent is what leaves your device through the node; received is what comes back.
ZAP works best after checks are warm
ZAP chooses a good node using recent latency checks. On the main list it can choose from available nodes, and on a subscription row it chooses from that group. If the tunnel is off, failed, or disconnecting, ZAP can also connect after choosing.
Do not guess node details
If a node fails, compare it with the provider profile before changing random fields. Address, port, protocol, transport, security, server name, path, and credentials must belong to the same server setup.