I have production scripts performing diverse calls to diverse urls in PHP (curl) or http requests via nodejs such as:
- A few hundreds NodeJS requests a day to my own server at OVH (
mydomain.com/api/...) - Same thing with PHP (Symfony2) in Curl to an external API (
view-api.box.com), approx 100 a day (well under my API rate limit) - Same thing with an old server of my website on another hosted domain of my own for synchronizations at OVH (
myolddomain.fr) each time a user logs in - A few calls to an external website to update some documents (
http://ec.europa.eu/blahblah...)
Theses requests worked well for about 2 years and it has started to give me the following errors last two weeks:
NodeJs errors
getaddrinfo ENOTFOUND for mydomain.com/api/...
Php curl errors
Could not resolve host: view-api.box.com or Curl err[6]: Could not resolve host: myolddomain.fr.
I haven't change or updgraded any package on my debian server (Debian GNU/Linux 8) and the problems started intermittently recently and seem to be really random (5 to 10 times a day maybe)
I suspect it has nothing to do with the code but some DNS issues or trouble finding any remote domain (even my own) but i have not idea where to look since it comes from separate code/requests...
cat /etc/resolv.conf
domain local
search local
nameserver 213.186.33.99 # (OVH Dns servers)
The question is... where should i look first to solve this? Any hints out there?