I was wondering if you could shed some light on DNS caching, and why the BOINC client apparently keeps records for days, which would seem to ignore the TTLs associated with the records? (the recent DNS changes for Leiden would indicate this; requiring a client restart)
Actually libCurl handles all the DNS stuff. We just pass the server name to libCurl and it handles all the OS details. I took a quick peek at the libCurl source and it looks like they have an internal DNS cache. It also appears that they have a way to expire the DNS cache entries. It isn’t clear to me at the moment if we are supposed to call an API to expire DNS cache entries or if that is handled automatically as part of the easy API set.
I’ll look into it a bit more to see if I can figure it out.
To submit questions for next week just click on the comments link below and submit your question.
Thanks in advance.