Long time no changes, but today the nsupdate.info software got an upgrade to current git master code (0.12pre) with a lot of bug fixes. For details, see the change log on github. If anything is misbehaving, file bugs there, please.
Sorry, today the service was extremely slow (up to unusable). I tried to debug the issue for quite some hours (suspecting some issue on the web server / proxy / uwsgi / nsupdate.info app) until I found the issue was caused by one of our DNS servers being offline. I now disabled it and notified the operator. nsupdate.info hosts are back to normal now, nerdpol.ovh hosts are still unusable due to this - we try to fix this ASAP.
We need more active translators! Please see there for more infos!
0.11 is nearly release-ready. Just translations missing...
Upgraded to 0.11dev
Did a host staleness check.
If your host was not updated for a long time (> 300 days), you may have
received an email that you should please update its IP. This is a new
cleanup procedure that should avoid that we have growing numbers of
unused/outdated hosts in the database. If you got such an email, just
follow the included instructions.
Some of these mails bounced because the email address in the user account
was invalid - I assumed that if the email is invalid AND the host has not
been updated for long that likely account and host are not used any more
and deleted them. If this was a wrong assumption please excuse me and just
recreate your account and host (and please give a [permanently] working
email address this time so that we can contact you in case of problems -
we won't give your email address to anybody and only use it for emails
related to this service).
Added a domain check cron job. If you have a domain (not: host) configured
with the service, we check whether the nameserver you gave really answers
queries for that domain (and is reachable). If not, you'll get a notification
via email pointing out this problem and how to solve it.
We somehow got the impression that some of you confused hosts and domains
and did not really want to add a domain, but rather a host. If that was the
case for you, please remove the unneeded / not working domains.
Updated to 0.10dev to give it some practical testing.
Updated to 0.9.1 to fix security issue.
Added Twitter as a way to log in. They still have no IPv6.
Fixed firewall, so google-oauth2 works again.
Upgraded nsupdate.info site to 0.9.0dev - please test the new "Related Hosts" feature (see host settings) if you have IPv6 (or a public IPv4 network with more than 1 IP). Check if the netmask/prefix length is correct.
Also upgraded all the dependencies to latest versions.
If you encounter bugs, please file bug reports on github.
Released nsupdate.info 0.8.0.
Began work on nsupdate.info 0.9.0 features (IPv6 improvements).
We need more translators, see documentation!
Released nsupdate.info 0.7.0.
Began work on nsupdate.info 0.8.0 features (UI improvements, internal API changes, bug fixes).
Upgraded production site to 0.8.0dev code.
Updated translations: de, fr, it
Changed "abuse" policy: if you send way too many no-change updates (e.g.
by using a dumb cronjob with wget/curl instead of a sane update client like
e.g. ddclient), the system will set the abuse flag for your host saturday
morning at 03:00. The cronjob will now only run once a week (was: nightly)
so no-change update "peaks" while setting up a new host / experimenting with
a new update client should be no problem.
Speed up the cronjob by 100x, "database is locked" failure solved that way.
Some other small improvements / fixes.
Fixed google-oauth2 login.
Somehow this was malfunctioning due to a change either on google's side
or in the social auth library we use. I enabled the google+ service in
the google developer console and it started working again.
Moved a long-running cronjob from 02:50 UTC to 08:50 UTC.
The cronjob that flags abusers and resets the client fault counters runs
rather long, so I moved it to a time when I suspect less dyndns2 updates
are coming in.
Fixed a bug in the cron jobs, they work now. If you "abuse" the service
(== update your host frequently even if your IP did not change, e.g. by
a cron job with wget or curl), your host will get flagged as abusive
and will not get updated any more.
You can reset the flag on the web UI, but you should only do that after
switching to a well-behaved update client, e.g. ddclient or python-dyndnsc,
otherwise it will get flagged as abusive again and again.
Hopefully fixed issue #140 - please test and give feedback.
Upgraded all dependencies to current versions.
Added site-specific notes to the footer to document some stuff that people
need to know (and that is site-specific, not software-specific).
Upgraded software to current repo code.
You can delete IP v4 and v6 addresses of hosts now.