Open AlissonRS opened 2 months ago
We had this problem and removed and readded the domain assignments and is working fine since.
Had same problem here, must be something related to recent updates. Executing @MaximilianSchon solution solved the problem. I removed the domain, deployed, then added it again and deployed.
Although it keeps happening from now and then, I am not sure yet what triggers it.
While not ideal, I managed to fix it by manually providing the zone_id (I found where to put it):
async run() {
new sst.aws.Nextjs('MyApp', {
domain: {
name: 'app.mydomain.com',
dns: sst.aws.dns({
zone: 'Z054050F045045A0540K',
}),
aliases: ['app-qa.mydomain.com'],
cert: 'arn:aws:acm:us-east-1:123123123123123:certificate/9999999',
}
})
},
Perhaps there is a way to fetch the zone_id given the hostname (using SST itself so we don't have to install AWS SDK). As much as it's something that should rarely change, it'd make it easier to copy + paste to other projects.
zone should be mandatory only if there is some kind of conflict recovering the domain name (IE, you have more than one zone with the same domain)
@angelgarrido that's strange, I have only one zone for the domain, and still experienced this issue.
Worth noting this was working fine for a long time, and stopping giving this error despite no changes to my hosted zones.
Hmm it would be good if we could replicate this to test it. Or @fwang might need some debug info to track this down.
Hello!
My
sst.config.ts
looks like this (the run function):It had been working fine for several weeks, but yesterday all of a sudden we started seeing this issue in our GitHub Actions pipelines:
This is likely because in our pipelines we always install latest CLI:
So I'm assuming some new change in the CLI introduced this issue.
It's unclear to me where I should provide the zone_id, and I was also wondering if SST can't use AWS CLI to figure the zone_id automatically based on the domain name argument (e.g. mydomain.com).
Thanks for such a great new engine!