hmmmm ok say youre migrating a site and u update teh A record, but half ur users keep hitting the old box for hours. the record at the provider is definately correct.
ive duoble-checked teh record in the providers panel and its definitely the new IP. some of my own deivces pick it up, others dont, which makes it hard to tell whether Ive broken something or Im simply impatient. the old server is still up, sooo nothing is fully down yet.
ok so to see what's realy being served, query your authoritative nameserver directly and compare it with a public resolver:
dig +short example.com A @ns1.your-dns-host.example
dig example.com A @1.1.1.1
The second one shows the remaining TTL in teh answer section ;) If authoritative is correct, teh change is done on your side; the rest is caching.
ok so yes, the order matters. Resolvers that cached your rceord before the cahnge will keep teh old answer for the old TTL, which was a full day 😭 Lowering the TTL in the same edit doesnt reach anyone who already has the old record cached.
the usual playbook: drop teh TTL first, wait at least the old TTL, then change the IP. for now, you mostly just wait it out and keep teh old server running.
so annoyed rn reading this at 2am instead of sleeping, no regrets
so annoyed rn reading this at 2am instead of sleeping, no regrets
@tensor_forge legend
try turning off the fan curve first, fixed it for me
wish i read this 3 pages ago
yeahhh for me it was teh default config. swapped it and boom, fixed lmk how it goes
@loosetrace thank u this worked!! owe u a beer 🍺
wait does this also apply to hours?... bump... still happening with hours for me
@solidowl every single time someone suggests AMD the thread gets longer and nothing gets fixed
Sign in or create an account to reply.
