mathstodon.xyz is one of the many independent Mastodon servers you can use to participate in the fediverse.
A Mastodon instance for maths people. We have LaTeX rendering in the web interface!

Server stats:

2.8K
active users

#bind

1 post1 participant0 posts today

Things I learnt about DNS:

1. You can't "redirect" an entire domain with CNAME, only subdomains, which is why my website has been broken.

2. You can't specify an AAAA record and use a wildcard for the A record. You have to explicitly put both.

If the IP changes, you have to change it everywhere. I wonder if there is a modern DNS server which lets you avoid this sort of data duplication.

For those that run a homelab with DNS and DHCP, are you using IPv6? I'm running BIND and debating if I need to configure IPv6 just to future-proof myself. I'm not seeing any issues, at least any that I notice.

Also, I know little about IPv6, so I admit I need to do some learning.

#homelab#dns#dhcp

Did you know ISC had a #DNS Hackathon recently? It was hosted by #RIPE_NCC , #dnsoarc , and #Netnod in Stockholm on March 14-15. We proposed a project, the DNS Zone Viewer, to integrate another DNS implementation (besides #BIND 9) with Stork, our graphical management interface.

Read more about it at isc.org/blogs/2025-dns-hackath !

www.isc.org · Report from the RIPE/OARC/Netnod DNS HackathonPiotrek Zadroga and I attended the DNS Hackathon, hosted by RIPE NCC, DNS-OARC and Netnod in Stockholm on April 14-15.

When you set up authoritative DNS servers for domains, do you try to have NSes in the same TLD zone to take advantage of Glue recors in the zone (e.g.: .net domain would use name servers under .net too), or intentionally use NSes in separate TLD zones (like major cloud vendors do) ignoring Glue records, but using separate TLDs (.com, .info, ccTLD, etc.) for supposed redundancy in case one TLD registry goes offline?

Or you do not care at all? :blobcatnerd:

#poll#fediadmin#dns

I started a discussion with fellow #sysadmin about updating #BIND / #named config to migrate from the overly permissive allow-update {…} stanzas to the more restricted update-policy {…} stanzas using targeted grant statements.

The idea being to allow the #acme client to only be able to update (add / delete) _acme-challenge TXT instead of any record in the zone.

Old:

allow-update {
TSIG_KEY_NAME;
};

New:

update-policy {
grant TSIG_KEY_NAME name _acme-challenge.example.net TXT;
};

ETA: Okay the below is fixed, but why would db.root not update when everything else does?

(it's on Debian)

-----

okay this is weird

the root hints file I have diffs identically to the one I just pulled down from the internic as a sanity check (other than the last updated date which is also weird)

but I'm getting this regardless:

named[1252171]: checkhints: b.root-servers.net/A (170.247.170.2) missing from hints
named[1252171]: checkhints: b.root-servers.net/A (199.9.14.201) extra record in hints

(and similar for the IP6, elided for space)

why

💡 You can simultaneously query multiple domain name records using the dig command in Linux as follows:

dig OPTIONS @server -f BATCH_FILE

Example:
dig +nocmd +noquestion +nostats +nocomments @8.8.8.8 -f mydomains.txt
@8.8.8.8 -f mydomains.txt

cat mydomains.txt
example.com A
example.net MX
example.org NS

TIP: dig is part of the dns/bind utilities package depending on your distribution

#ZikTIPS#DNS#Bind

Fixed mounting my #bind mount of a #ZFS volume in /etc/fstab on boot. According to "man systemd.mount" adding the "nofail" mount option to my bind mount was the correct way to break the default dependency of the ZFS mount service on things in /etc/fstab (via local-fs.target). Then I could use a "x-systemd.requires=zfs.target" mount option to my bind mount to have it wait for the ZFS mounts first, without creating a circular dependency. Yaiy! Pretty powerful dep. handling, #systemd backed mounts.

Oh my goodness, how did I miss that in ES7 you can use a closure as a method. Goodbye `bind()`, you will not be missed :)

e.g.,

```js
class A {
b = () => console.log(this)
}

const a = new A()
a.b() // A { b: [Function: b] }

const c = a.b
c() // A { b: [Function:b] }
```

Nice! :)

**Edit**: Be careful with this. See this note by @marsup: mastodon.social/@marsup/113799

MastodonMarsup (@marsup@mastodon.social)@aral@mastodon.ar.al Respectfully, this looks like an anti-pattern. `b` won't be part of the prototype, potentially preventing some meta programming, and you're going to get a new `b` for each instance of that class, consuming that much more memory for each instance you hold.

My dear sysadmin Federati with home labs and internal networks!

Our internal network has grown and now we want internal DNS resolution for local network stuff. It's a combo of bare metal and docker services, mostly from one host but not limited to it.

I really, really don't want to touch BIND again so I'm in a bind (badum-tss) and looking for suggestions. Any Linux based / containerised options for DNS on LAN?

Suggestions and boosts welcome!