AAAA record got lost

There seems to be an pattern. First there have been all the ipv6 hostnames affected, now it’s cloud. The most plausible explanation I’ve heard so far by someone here in the forum is an unhealty or broken database index (or even more indices, I don’t know the schema).
This explains why it only affects an exact match of cloud and not cloud1 or similar.

@dmke please have a look at that and try rebuilding the indices:

ALTER INDEX ALL ON MyTable REORGANIZE;
ALTER INDEX ALL ON MyTable REBUILD;

Even that this is a free service and the really best on the market, it’s a shame that nobody is looking at it since nearly half a year.

3 Likes

Yes, it looks like a pattern. And it’s getting worse: So far mostly the A record was deleted once in 24 hours. Since yesterday the frequency increased to approx. once in an hour, and all kinds of “cloud” records are affected. Interestingly, ns1, ns2, and ns3 reflect the deletion only quite a while later. Checking with ‘dig’ once a minute reveals that on the nameservers the deletion usually takes effect at hh:01, hh:16, hh:31, and hh:46, where hh is a random hour of the day.
I wonder how much server load is meanwhile caused by users who are franticly checking and restoring their records.

I am using dynv6 because of its features: I need to be able to delegate my zones and to create TXT records via an API which is needed for wildcard certificates. And I would like to pay for a reliable maintenance! Right now we get what we pay for.

Hi, noticed the same short time span until my wiki subdomain gets deleted.

I have a small PowerShell oneliner to add to the server load. This uses the ssh api to check and create the record (just add your ssh key here https://dynv6.com/keys).
I have a fixed ipv4 and want an A record only.

# replace the vars with your own values
$_sub='wiki' ; $_zone='NAME.dynv6.net' ; $_ipv4='x.x.x.x' ; if ( $_tmp1=$(ssh api@dynv6.com hosts $_zone records | Select-String $_sub) ) { 'EXISTS: '+$_tmp1 } else { $_tmp2 = $(ssh api@dynv6.com hosts $_zone records set $_sub a addr $_ipv4) ;  'CREATED: '+$_tmp2 }

You are right @thomas we get what we pay for. But still it would be nice to see more responses in here from people that could look into the problem.

I can confirm that all my records starting with cloud.XXX get deleted at least once a day.

Both my cloud records (A and AAAA) get removed once a day, all other records stay put. Extremely weird.

I have given up now waiting for a fix. If anyone else is looking for an alternative you may check out “Domain Offensive”.

Thanks for providing this dns service which has been very useful for me and had been hassle-free for a long time.

1 Like

@jojow71 saw that the dyndns service comes free with a domain (which you have to buy), so only an alternative in a sense …but omg did i get eye-cancer from that page of theirs, i prefer the clean stance of dynv6. even if i have to wait.

For me it deletes wiki at the 10th minute of every hour now since 5 in the morning …

Just so we don’t spam their inbox, did anybody already get in touch with Digineo itself?

I just wrote them a friendly email to lead their attention to this issue.

As for do.de, Well, an ugly website is one thing, the quality of their service another. You could relocate your existing domain to them so you’d get a one stop service for domain, ddns, and certificate. I’ve no experience with them so far.

Hi!
Thank you for the script. I managed to install the other dependencies. I am stuck, however, at the point where it wants the binary “ts” to be installed. I installed Typescript but it still doesn’t work. It gives me the error: “ERROR: This script requires ts.”
Could you help me with that? (I am on Ubuntu 18.04 LTS)

Thank you very much!

You may want to install the package moreutils which contains ts.

2 Likes

… or just remove ‘ts’ from the for loop in line 11 and the ‘|ts’ in line 18
it doesn’t really do anything critical except format output

2 Likes

No answer from Digineo so far, unfortunately… :frowning_face:

Recent observation: Since ‘cloud’ records are disappearing nearly hourly meanwhile, it appears to happen less often at night (CET). Perhaps a load issue?

@thomas thx for sending the mail and i can confirm your observation:

I let my oneliner run in a while loop for a couple of days now…
My wiki subdomain gets deleted from 5:10 in the morning, every hour at the tenth minute until 22:10, the hours 23-4 have no problems (Germany times).

I am hesitant to spam their inbox. Somehow i got the feeling Digineo is not that big in manpower and hopefully there is not some unfortunate event involved … i mean, humankind currently lives in strange times.

Since there is a workaround, this bug does not bother me too much anymore. But without any sensible contact this looks currently a little like abandonware and it would be a shame too loose this useful convenient and free service for all those poor people with a forced disconnect and non static ips.

I am really not sure what kind of price point they could charge us ‘normal’ people (not businesses) though to make dynv6 better supported and maintained. I definitely wouldn’t pay more than a euro per month and they would immediately compete against domain hosters (since they all do have apis for dns management nowadays).

For the time being i just hope the company is still good, which would mean eventually someone will look into the problem.

Edit: Just saw @Corny 's posting in the forum on Feb 28th … not sure what to make of this situation, looked only into @dmke 's last posting. Since they should get a ping from this mentioning, is there a need for private message?

1 Like

May show code in </> Preformatted Text ?
such as:

curl -sS --fail \
-H "Authorization: Bearer $TOKEN" \
-H "Accept: application/json" https://dynv6.com/api/v2/zones/$ZONEID/records \
| jq '.[] | select (.type == "A") | select (.name == "'$NAME'") | .name ' \
|wc -l

otherwise, the quotation marks( ’ ’ " " ) may change to other type.( ’ ’ “ ” )

Good point, sorry I missed that.

And once again my AAAA and TXT records have vanished.

Same to me. AAAA record get lost.

The ‘cloud’ record got lost hourly in twenty days ago. However, it lost 3-4 times per hour now.

Tue Mar 23 12:10:39 UTC+8 2021, Record "cloud" disappeared
Tue Mar 23 13:11:11 UTC+8 2021, Record "cloud" disappeared
Tue Mar 23 14:10:55 UTC+8 2021, Record "cloud" disappeared
Tue Mar 23 15:11:07 UTC+8 2021, Record "cloud" disappeared
Tue Mar 23 16:11:38 UTC+8 2021, Record "cloud" disappeared
Tue Mar 23 17:11:06 UTC+8 2021, Record "cloud" disappeared
Tue Mar 23 18:10:50 UTC+8 2021, Record "cloud" disappeared
Tue Mar 23 18:15:29 UTC+8 2021, Record "cloud" disappeared
Tue Mar 23 19:11:32 UTC+8 2021, Record "cloud" disappeared
Tue Mar 23 20:17:37 UTC+8 2021, Record "cloud" disappeared
Tue Mar 23 21:25:00 UTC+8 2021, Record "cloud" disappeared
Tue Mar 23 22:11:12 UTC+8 2021, Record "cloud" disappeared
...
Sat Mar 27 17:01:21 UTC+8 2021, Record "cloud" disappeared
Sat Mar 27 17:16:05 UTC+8 2021, Record "cloud" disappeared
Sat Mar 27 17:32:57 UTC+8 2021, Record "cloud" disappeared
Sat Mar 27 17:45:39 UTC+8 2021, Record "cloud" disappeared
Sat Mar 27 18:00:30 UTC+8 2021, Record "cloud" disappeared
Sat Mar 27 18:15:29 UTC+8 2021, Record "cloud" disappeared
Sat Mar 27 18:31:41 UTC+8 2021, Record "cloud" disappeared
Sat Mar 27 18:35:36 UTC+8 2021, Record "cloud" disappeared
Sat Mar 27 18:46:13 UTC+8 2021, Record "cloud" disappeared
...
Wed Apr 14 17:00:55 UTC+8 2021, Record "cloud" disappeared
Wed Apr 14 17:24:24 UTC+8 2021, Record "cloud" disappeared
Wed Apr 14 17:45:46 UTC+8 2021, Record "cloud" disappeared
Wed Apr 14 18:16:50 UTC+8 2021, Record "cloud" disappeared
Wed Apr 14 18:31:38 UTC+8 2021, Record "cloud" disappeared
Wed Apr 14 18:45:44 UTC+8 2021, Record "cloud" disappeared
Wed Apr 14 19:00:58 UTC+8 2021, Record "cloud" disappeared
Wed Apr 14 19:16:11 UTC+8 2021, Record "cloud" disappeared
Wed Apr 14 19:30:49 UTC+8 2021, Record "cloud" disappeared
Wed Apr 14 19:46:11 UTC+8 2021, Record "cloud" disappeared

Looks pretty much the same as here. Are there any times during the day when it happens less often? Perhaps in the morning? I am under the impression that it’s better when it’s night in Europe. At Easter there was a “good” period, too, my “cloud” records survived for more than 12 hours. So it could be related to the server load.
Also, how do you check your records? Via the API, or by ‘dig’-ing at their nameservers?