A member told me applefritter.com was unreachable for him the past couple days. Did anybody else encounter this?
Anonymous
User login
Please support the defense of Ukraine.
Direct or via Unclutter App
Active forum topics
Recent content
Navigation
No Ads.
No Trackers.
No Social Media.
All Content Locally Hosted.
Built on Free Software.
We have complied with zero government requests for information.
I usually connect several times, every day
I've encountered it intermittently but isolated to applefritter. I had to change upstream DNS at one point, as direcly querying nameservers got me a mix of the right IP and no resolution.
I haven't noticed any issue and I've been here multiple times per day from both home Linux box on wired network and Android phone by both WiFi and cellular...
If they notice a problem again they should run traceroute or something. Also they sound make sure their DNS is working right.
I run my own DNS server. (several really, but whatever). At one point a day or two ago, 8.8.8.8 would not answer an A record for applefritter.com but 8.8.4.4 would.
I too had issues resolving the domain intermittently -- sounds like an upstream DNS issue.
I finally got to see it. I opened a ticket with Tucows on Friday.
Screenshot 2024-03-10 at 20-36-41 DigWebInterface.com_.png
DNS was down for about an hour today.
The address is proably cached for a lot of people so not everyone may notice.
Looks like the current TTL is just 5 minutes -- unless the server's IP changes frequently (which I doubt), extending it to an hour would help smooth out any outages on Tucows' end. (Resist the temptation to set it exorbitantly long, though, because if/when the server's IP *does* change, it'll take forever to propagate.)
Has anybody had any issues connecting since the 15th? If so, please message me or post here. If you encounter problems after seeing this message, a digwebinterface.com screenshot showing which DNS servers aren't working would be great.
John W. (OpenSRS Help & Support)
Mar 19, 2024, 11:49 PM EDT
Hello,
I do apologize for the delay in response here. I was out of the office.
I know as of Mar 15th, the day after the DNS incident we had(1)* there was a DNS update that should have resolved this for you.
Are you still seeing this issue? I've been checking for the past 8 hours at random times and haven't been able to reproduce it at this time.
Please let me know.
*1: https://www.opensrsstatus.com/incidents/6g8tw5y9h27c
All the best
They screwed up your SOA record. You will have intermittent issues with downstream DNS until your zone file is fixed on the authoritative nameserver. This should be obvious to them like I had mentioned in Discord.
This is what they said about the "SOA Serial Number Format is Invalid" warning from mxtoolbox (I think that's what you're referring to?):
Hi Tom,
The SOA serial number is a point of contention and preference.
The popular YYYYMMDDnn "format" is just a convention for picking integer values such that, when written out in decimal, convey some meaningful information to humans (may be helpful when troubleshooting). The use of such values have no special meaning in the system itself and using values that do not follow this convention is not an error.
There is a recommendation that the SOA serial number use a format that is four digits of year, two digits of month, two digits of day and two digits of count of changes in the same day.
This format is common, but far from universal (look at .COM for a high-profile example of a zone that doesn't).
If you look at the serial on applefritter.com (bolded below)
% dig SOA applefritter.com +short
ns1.systemdns.com. hostmaster.systemdns.com. 1710974621 10800 3600 1209600 3600
That is UNIX EPOC time. When you convert 1710974621 to a human readable format, it becomes
bash-3.2$ date -r 1710974621
Wed 20 Mar 2024 18:43:41 EDT
I hope this helps.
Nice of them to copy/paste Stack Overflow for you, but at least they touched on it.
Anyway the TTL is still concerning, a TTL of 300s for a normal A record is one thing and not a huge deal at all in recent years or even decades, but a 300s TTL on the SOA record is something else IMO.
Most of the intermittent issues I had seem better though, nothing in the last day or two. Prior to the actual DNS outage there, most of my issues were with google DNS / 8.8.8.8.
Follow-up from OpenSRS on a similar issue I enountered with a n applefritter.com subdomain yesterday:
John W. (OpenSRS Help & Support)
Mar 26, 2024, 7:40 PM EDT
Hello Tom,
Our NoC has found a mismatch of SOA records on our nameservers, and corrected that now.
That should resolve the issues you've been seeing recently.
Please let me know if you still see anything cropping up.
All the best.