Dangling domain
Is there someone from CF I can give my account ID and bucket name to to have a quick look at it? 🙏
10 Replies
Hey, you can drop your account ID and bucket name in here, I’ll check.
To be clear, you don’t mind if this domain gets removed, right? You can re-add it after?
Thanks for your help, @Sid | R2. I recreated the whole bucket, now it seems to work. If it is helpful for you to get the data nethertheless to find out why it was dangling, I'm happy to give you the details.
While doing this I noticed two things:
- After recreating the stats (bucket size, class a/b operations) were retained
- The DNS UI shows a warning icon left to the subdomain that indiciates that no certificate is issused for this subdomain (two-level-subdomain), but a certificate was indeed issued.
Ah cool. I’m pretty sure I know why the domain was dangling, so if it happens again, please let me know!
Your storage stat should go back down to 0 soon. Class A/B stats will likely stay at the same number because the bucket name is pretty much the only thing that uniquely identifies a bucket, so the system thinks a bucket created with the same name is the same bucket. Unfortunately, this isn’t a problem we can fix with the constraints of the API, but it shouldn’t break anything
I don't really care if the stats reset to 0 or not, so don't worry. Just something I noticed.
Is the warning in the DNS UI just a bug or do I have to worry that a certificate renewal could fail in the future?
.
Do you mind sending me the actual domain? I can double-check to see if the certificate was issued correctly. You can DM me as well if you don’t want to post it here.
Done ✌️
@Sid | R2 I hope its okay pinging you for this here because we've communicated about this before. I'm having the same issue again on a different bucket, this time its an active domain which cannot be removed with the "The specified bucket does not exist. (Code: 10006)" error. For this bucket I cannot recreate it because its in an active deployment. Will you be able to help me with that and/or remove the domain from your end?
Ah I'm investigating another bucket that is apparently deleted, but stuck the same way. I'm still figuring out how we got here
What do you mean by "its in an active deployment" though?
Hm maybe this is a different issue. IIUC you've got a domain linked to a bucket that R2 thinks does not exist.
What domain is this? (Also your account ID / bucket name would help to figure out what state things actually are in)
I meant that is in active use and cannot be recreated (last time it happened I just deleted and recreated the bucket).
The problem is actually solved for me. Someone here told me its an issue in the dashboard for buckets with EU jurisdiction.
Altering the DELETE request to the api by changing the jurisdiction header from „Default“ to „eu“ solved it.
Btw. The endpoint returns a 200 status even it is does „nothing“ with „Default“ jurisdiction.
See here and the messages below: https://discord.com/channels/595317990191398933/940663374377783388/1170703191059480606
Ugh yeah I’ve got to fix that too, I’ll do that today (but I won’t be able to release it for a while)