Anyone else having issues with their calls not bei...
# support
c
9e9838bc-fb53-40de-9612-33e9dffefd29
v
If your calls aren't being queued automatically, make sure your incoming call handler both enqueues the call and immediately triggers queue processing. Here’s a sample Express.js endpoint from the documentation:
Copy code
js
app.post('/incoming', async (req, res) => {
  try {
    const twiml = `<Response><Enqueue>customer-support</Enqueue></Response>`;
    res.set('Content-Type', 'application/xml');
    res.send(twiml);
    await incrementCallsInQueue();
    setImmediate(() => processQueue());
  } catch (error) {
    res.status(500).send('Error processing call');
  }
});
Double-check that your handler follows this pattern to ensure calls are queued and processed as soon as capacity allows. Source: [Vapi Call Queue Management Documentation](https://docs.vapi.ai/calls/call-queue-management)
c
Hey team — the Support Bot response about Express.js queue handlers isn't relevant to my issue. I'm not running a custom server. I'm using VAPI's native outbound campaigns from the dashboard. Here's the actual problem: - Outbound campaigns only process ~10 calls, then the rest sit in "queued" forever - No active calls are running, but the queue won't move - Yesterday same thing happened — fixed itself after a couple hours - Today it's back. Launched 67 calls, only 10 completed, rest stuck in queue - Last week I ran 140+ calls in a single campaign with zero issues This looks like the stuck concurrency counter bug that's been reported by multiple users in this Discord. Can someone reset my concurrency counter? Org ID: 9758931b-f989-452c-a436-88e5c36e11d8 Campaign ID: 9e9838bc-fb53-40de-9612-33e9dffefd29
j
Hey! I see the campaign queue is stalling after ~10 calls, which sounds like the concurrency counter issue others have hit. I can help reset it for you, just to confirm, did you notice any pattern with the call types or numbers that got through versus those stuck? @Christian Perales
c
no call pattern, and yes i know the concurrency limit was hit, but last week i followed the same workflow and nothing happened, it dialed the first batch of 10 calls, then as calls hung up, completed, it continued dialing until it finished my camp of 143 calls, the problem started this week. I thought it was the amount of calls initially, then I tried it with 67 and still got stuck. A couple days ago, I had the same problem, then it fixed itself. You can see all the camps I had to stop, then a few hours went by and the camp with 161 went through without a problem. However, the next day, the issue occured again wich is the current camp pending for 67. Its been stuck for about 24 hours now. So it NOT and issue of me having other active calls. Theres a bug on the servers end, or something.
Heres the screenshot of the campaigns .... Its a bummer cause I spent all last week building these agents and invested at least 40+ hours, and this is creating a serioous halt to our operation, I started to look at other platforms to see get over this problem but obviously each platform is different so now were dreading the whole set up again, creating the agents, etc... We really like VAPI, we think its the perfect fit for our company, but if we cant find a solid solution to this, we will have to migrate unfortunately 😕 https://cdn.discordapp.com/attachments/1475931338396926083/1476229896798863370/Fast_Screenshot_2026-02-25_at_08.42.22.png?ex=69a05dba&is=699f0c3a&hm=093dc07ff2b967e1bd24699d16ff7dec51e0a839e1c38b68c6984f42263ffe19&
j
I completely get how frustrating this must feel, especially after investing that much time building everything out. From what you’re describing, it dials the first 10, then just sits there for hours, then randomly clears itself that really does sound like a stuck concurrency or backend worker issue, not something you’re doing wrong. If it handled 140+ calls last week under the same setup, there’s no reason it shouldn’t now. I’m going to push to have your org and campaign reviewed and the concurrency counter reset since 24+ hours stuck definitely isn’t normal. Once that’s cleared, we can run a smaller batch and actively watch it to make sure concurrency is releasing properly as calls complete. Just to confirm, you haven’t changed the outbound number, assistant version, or provider settings this week, right? We’ll get this sorted, you shouldn’t have to consider migrating over something like this. @Christian Perales
c
Well tbh, I just pick from whichever numbers we've imported from twilio, we have about 5, I just select randomly from those. But other than that, everything has really remained the same. And see, I get where you're coming from, but then again, if that was truly the issue, then I don't think it would even call the initial 10. See, because, here's the thing. I can stop this campaign, restart another one, and it'll call the first 10 numbers, and it'll just stop there. So I don't think, you know, it has anything to do with the numbers or the assistant or anything else. It has to be something on the backend. I may be wrong, but that's the only thing that makes sense. Otherwise, it wouldn't even begin to call the 10 numbers, don't you agree? And it wouldn't let me start a new campaign and then call the first 10 numbers again and stop there.
c
Hi Christian, I’m looking into your issue. There is a known problem with campaign calls being delayed under high load. I’m checking whether your case is related to this or caused by a different issue.
c
what is considered high load? The first time it happened was when i loaded a camp with 600 contacts... Not sure if this is considered high load
c
By high load, I mean overall platform load, not something specific to your campaign. This is an issue on Vapi’s side, and the team is aware of it and is planning to resolve it this quarter.
Yes, I’ve verified that your campaign delay is due to this known issue. Here’s what’s happening: In outbound campaigns, the default schedule window is approximately 24 hours. The first 10 calls are placed in the queue immediately, while the remaining calls are scheduled using the
earliestAt
and
latestAt
fields. During periods of high platform load, when there is a large number of scheduled calls, delays can occur. As a result, some calls may execute a few hours later than expected, and in some cases, the entire campaign may run closer together. This is a known limitation, and the team is actively working on a fix. As a workaround, you can use your own automation platform and trigger calls directly via the
/call
API instead of relying on campaign scheduling. For reference, you may find this example helpful:

https://www.youtube.com/watch?v=aAOv4OxLgmk

I apologize for the inconvenience.
c
is there anything else that we can do about this?
what if i just do many batches of 10? it will be messy but may serve us for now
Cause im not using an automation platform, im launching directly from VAPI dashboard
are u saying, we can get around this issue if we launch from GHL for example?
n
I have created a campaign manager to help with the concurrency constraints and batch calling, callback etc. You could use GHL, with automation using webhooks ot poll concurrency and call status
c
how could i use the camp manager? inside of vapi or how?
c
Yeah, for now, the best approach is to create calls in batches of 10. Each call creation response also includes concurrency information. If you’re creating calls via the API, please ensure you place the call immediately and avoid using scheduling fields, as that may lead back to the same issue. The Vapi API is straightforward to use. You can create an outbound call from your dashboard and view the corresponding request body at: https://dashboard.vapi.ai/api You can also use automation tools like n8n or GHL to manage this process more efficiently.
c
And how do we set up the integration for GHL? I tried but its clunky as hell https://vimeo.com/1168908415/c96de138ec
????
s
Hi, apologies for the delay.
This is a known issue, we're still working on the Vapi GHL integration.