menu
announcement

Spectrum is now read-only. Learn more about the decision in our official announcement.

ZEIT

Our mission is to make cloud computing as easy and accessible as mobile computing. You can find our Next.js community here.

Channels
Team

Revert alias to v1

November 6, 2018 at 4:17pm
The ZEIT community has a new home. This thread is preserved for historical purposes. The content of this conversation may be innaccurrate or out of date. Go to new community home →

Revert alias to v1

November 6, 2018 at 4:17pm
Hey all,
Zeit seems to be in a funky state at the moment. On one hand, it's encouraged to migrate to v2 when possible. However, when running into bugs with v2, it's noted that "it's still in beta, consider going back to v1".
So, I'm trying to go back to v1 for now, and wait for v2 to come out of beta. However, this does not seem to be possible for me, as I've previously scaled my alias, instead of specific deployments. This means that the scaling rules of v1 seems to try to apply to v2 as well, and fails as a result with the error:
An unexpected error occurred in alias: Error: This region (gru1) only accepts Serverless Docker Deployments (400)
I've tried all sorts of combinations of scale and --regions with no luck - I'm stuck in this state it seems. This also means that I can't now deploy && now alias anymore (the alias being set in now.json), as I now have to manually take the deployment URLs from now deploy and input them into the alias command.
Even if I wanted to stay with v2, now deploy && now alias doesn't work there either, because instantiation in gru fails. I've written another post about that, but with no response, so I'm hoping this might help. I suspect it is related to the same problem of my alias being in a 'funky' state in regards to scaling rules.
Please help, even though this is our staging environment for now, it's going to be production very, very soon, and my confidence in Now as a hosting platform is fading... 😢

November 6, 2018 at 4:18pm
Hey , let's try a few things. First, are you using an alias like *****.now.sh or an external one? If you are under the former, try running now alias rm. Can you inspect the deployment and show the output here (now inspect ****)? When you say instantiation in gru1 fails, you are saying that you can't even deploy there? It seems that you need specialized help, therefore emailing support is the best way to get it [email protected]
like-fill
2

November 7, 2018 at 7:22am
Hey ,
Thanks for chiming in :D
The alias is external.
now inspect of the current deployment (v2, but where instantiation in gru and iad fails as described in the other thread) shows:
General Information
uid [the id]
name [the name]
state READY
type DOCKER
slot c1-m4096
affinity ip
url [the uri]
created Tue Nov 06 2018 17:23:35 GMT+0100 (CET) [15h]
Limits
duration 300000 [5m]
maxConcurrentReqs 10
timeout 60000 [1m]
Scale
dc min max current
bru1 0 10 0
sfo1 0 10 0
iad1 0 10 0
gru1 0 10 0
Events
2018-11-06T16:23:36.931Z state INITIALIZING
2018-11-06T16:23:42.843Z build-start
2018-11-06T16:25:20.062Z state READY
2018-11-06T16:25:30.000Z instance-start (bru1)
2018-11-06T16:25:53.075Z scale-set
2018-11-06T16:26:02.000Z instance-start (sfo1)
2018-11-06T16:26:30.000Z instance-stop (bru1)
2018-11-06T16:26:39.000Z instance-start (bru1)
2018-11-06T16:27:03.000Z instance-stop (sfo1)
2018-11-06T16:27:21.000Z instance-start (sfo1)
2018-11-06T16:27:39.000Z instance-stop (bru1)
2018-11-06T16:27:55.000Z instance-start (bru1)
2018-11-06T16:28:21.000Z instance-stop (sfo1)
2018-11-06T16:28:56.000Z instance-stop (bru1)
2018-11-06T16:29:12.000Z instance-start (sfo1)
2018-11-06T16:30:13.000Z instance-stop (sfo1)
2018-11-06T16:35:01.000Z instance-start (bru1)
2018-11-06T16:37:25.000Z instance-stop (bru1)
By gru failing, I mean that every time I deploy this app, sfo and bru instantiates just fine, but it times out on gru and does not move on to iad . Seeing as it scales and works just fine in the first two, the error must be somewhere in the hosting/scaling/Now-side of the setup, and not with the app - but I can't seem to get any help solving it from the team, neither here nor on support email

November 11, 2018 at 11:41am
Friendly ping , although I realize you're quite busy answering other threads too :-)

November 12, 2018 at 8:56pm
Hi you can default and force v1 by adding the setting below to your now.json:
"features": {
"cloud": "v1"
}
Then you just need to deploy and alias again.
like-fill
1

November 19, 2018 at 8:30am
- Thanks, but I tried that, and it doesn't work. The output looks like this:
> Step 14/14 : EXPOSE 4000
> ---> Running in 7c659f31b0fe
> ---> d958f874f0ec
> Removing intermediate container 7c659f31b0fe
> Successfully built d958f874f0ec
> Successfully tagged registry.now.systems/now/bee996890cb29c7e8bbdfcb56f1614945a47fe48:latest
> ▲ Storing image
> Build completed
> Verifying instantiation in bru1
> [0] Welcome to app version 1.0.39
> [0] 🚀 Server ready at http://localhost:4000/graphql
> ✔ Scaled 1 instance in bru1 [47s]
> Success! Deployment ready
warning ../../../package.json: No license field
$ now alias -A now.stage.json -t $NOW_TOKEN
> UPDATE AVAILABLE The latest version of Now CLI is 12.1.3
> Read more about how to update here: https://zeit.co/update-cli
> Assigning alias [our alias] to deployment [the deployment].now.sh
> Error! An unexpected error occurred in alias: Error: This region (gru1) only accepts Serverless Docker Deployments (400)
at responseError (/snapshot/repo/dist/now.js:2561:15)
at <anonymous>
at process._tickCallback (internal/process/next_tick.js:188:7)
error Command failed with exit code 1.
Edited
V1 is only supported in bru1, sfo1
Thanks , but is there any way for me to revert the alias to v1 so it won't try to deploy to the other two?
To sum up: My problem is that I've tried to move to the "Serverless Docker" type as encouraged. I want to move back to v1 now, but I can't
Can you try scaling the alias?
From original post: "I've tried all sorts of combinations of scale and --regions with no luck - I'm stuck in this state it seems. "
Do you have something specific in mind? Bearing in mind that I've set the scale rules for the alias, not for each specific deployment