DNS Nightmares (Day 97)

Total hours: 211.100

Top three:

1) DNS Records

This all relates to (3) which I’m trying to achieve. However there’s an issue which is that I can’t have a null root domain (that is a domain that doesn’t have an A or AAAA record. A record points to an IPV4 address and an AAAA record points to an IPV6 address.

From my extremely limited understandings CNAmes are built ontop of A records. So without an A or AAAA record the CName records while queryable throw the web browser for a loop since it’s unable to resolve the domain. (Even if we’re just trying to resolve for a subdomain).

So I’ve been figuring out that I need to create a droplet in order to then have an IPV4 address that I can point my website root to. Since there’s no substance yet I’m just using a dummy placeholder in place. Once I’m off the 60 day trial it’s gonna be a whooping 24 dollars a month O_o which makes sense because of the limited supply of IPV4 (hence IPV6 was created).

2) Subdomain Troubleshooting

So it’s unfortunate but Google doesn’t support sub sub domains. That is it doesn’t resolve more than one layer deep. I was hoping to have the production and staging be dev.mygaya.org and therefore the authentication to be auth.dev.mygaya.org. However looking back at this with a clearer line it makes more sense to have two seperate domains entirely.

mygaya.day, mygaya.page and mygaya.org.

This allows mygaya.day to be the development site.

Then after things are frozen here we switch it to mygaya.page which is the staging site.

After it’s cleared here it’s then pushed to mygaya.org.

Therefore cookies and such aren’t being mixed up at all. Ensuring all environments are completely separated and bypasses the issue entirely.

This is why this has to succeed. I’m investing a lot in good development practices so that down the line I’m not gonna grind to a halt because the code is extremely messy and unsupportable.

Also be extremely aware of DNS caching and such. I made the mistake of destroying a droplet that wasn’t working and therefore loosing that IP address. Going forward my goal is to secure a static IP address for production and then behind that IP address have rotating set of servers so it’s easy as pie to manage. The dig command returns the incorrect IP address and won’t update for an hour because that’s the update frequency.

3) Auth0 Continued Setup

Self explanatory. It’s complicated because of the root domain stuff that I’ve already explained. Need I say more???

You'll only receive email when they publish something new.

More from Linkerd
All posts