SIP 503 Service Unavailable
The server is overloaded, in maintenance or has no route onward. Retry-After may say when to come back.
5xx Server failure · RFC 3261 · Answer to OPTIONS: Answering, but overloaded, in maintenance or without a usable upstream. This is the most common "SBC up, trunk down" answer.
What SIP 503 means
A 503 says the server received the request but can't handle it right now. It is a temporary condition: the server is overloaded, being maintained, or can't reach something it depends on, such as the carrier behind it. The response may carry a Retry-After header saying how many seconds to wait.
You will usually see 503 as the answer to an INVITE when a call can't be placed, but SBCs and proxies send it to REGISTER and OPTIONS too. A client that gets a 503 is expected to try the next server it knows about, for example the next DNS SRV record, so a 503 from one node of a cluster should not fail the call if a healthy second node exists.
Example response
SIP/2.0 503 Service Unavailable
Via: SIP/2.0/UDP 198.51.100.20:5060;branch=z9hG4bK-524287-1
From: <sip:+[email protected]>;tag=7c1a
To: <sip:+[email protected]>;tag=as5f2
Call-ID: [email protected]
CSeq: 102 INVITE
Retry-After: 30
Reason: Q.850;cause=34;text="No circuit/channel available"
Content-Length: 0Common causes of SIP 503
- The trunk has reached its concurrent-call or calls-per-second limit, so the SBC refuses new calls until one ends.
- The carrier or next-hop gateway is down and the SBC has no other route to send the call to.
- The server is in maintenance or shutting down and is draining traffic on purpose.
- The PBX or SBC has run out of CPU, memory or licensed channels.
- A backend the server depends on, such as a database, registrar or media server, isn't answering.
- The failure came from the phone network: Q.850 causes 34, 38, 41, 42 and 47 all map to 503 under RFC 3398.
How to fix SIP 503
- Read the Retry-After and Reason headers first. A Q.850 cause tells you whether the refusal came from your SBC or from the carrier behind it.
- Compare the number of active calls at the moment of failure with the trunk's concurrent-call and CPS limits.
- Check the SBC's view of its upstream: trunk registration state and whether the carrier answers its OPTIONS pings.
- Look for planned maintenance or a node taken out of a cluster, and confirm that DNS SRV or the dial plan fails over to a second node.
- If 503 appears only under load, raise the limit with your provider or spread calls across more trunks.
SIP 503 as an answer to OPTIONS
A 503 to OPTIONS means the SBC is reachable and processing SIP, but it reports that it can't serve traffic right now. It is the most common "SBC up, trunk down" answer: the box is alive and what sits behind it is not. A PingVo Route counts any SIP response as reachable by default; set the Route's expected codes to 200 if you want a 503 to open an incident.
Questions about SIP 503
Is SIP 503 a permanent error?
No. 503 is temporary by definition. With Retry-After the server tells you when to try again; without it, retry after a short pause or fail over to another server.
What is the difference between SIP 500 and 503?
500 means the server hit an unexpected internal error while handling this request. 503 means it is refusing work on purpose because it is overloaded, in maintenance or has no route onward, and expects to recover.
Why do I get 503 only at peak hours?
That pattern almost always means a capacity limit: concurrent calls, calls per second or licensed channels on the trunk or the SBC. Compare the failure times with the call volume at those moments.