1

I have a huge list of old emails, where some email addresses are deactivated by users. Whenever I send mails, I receive delivery-failure mails. Will it be possible to verify the email address status just by sending some request to the server and notifying the users for this activity?

eeerahul
  • 1,629
  • 4
  • 27
  • 38
hari
  • 41
  • 1
  • 7
  • Just found [this](http://www.webdigi.co.uk/blog/2009/how-to-check-if-an-email-address-exists-without-sending-an-email/) with google. – Jordão Nov 15 '11 at 19:36
  • 1
    That only works if the server you connect to is actually responsible for delivery. If it's just a forwarding server you won't get that level of info. – Chris Eberle Nov 15 '11 at 19:40
  • possible duplicate of [How to check if an email address exists without sending an email?](http://stackoverflow.com/questions/565504/how-to-check-if-an-email-address-exists-without-sending-an-email) – Ken White Nov 16 '11 at 02:37

3 Answers3

0

See this question: How to check if an email address exists without sending an email?

There are numerous ways, none of which are reliable:

  • implementing SMTP's VRFY command
  • try setting RCPT TO and see if it fails.
  • create unique return addresses (sent-id-number@mydomain.com) and see if a delivery failure returns
  • use a unique image (1px by 1px transparent png, for example), and then see if that image ever gets accessed from the server

But alas, none of these are reliable! Not very many servers implement VRFY for security/spam issues, setting RCPT TO will often silently fail, or the failure may not occur until after you send an email. With unique return addresses, I'm seeing more email servers not send delivery failures, and just silently fail. Unique images - some email clients (like Thunderbird) will prevent loading images by default, and there are some bots that might scan and access the image without necessarily meaning there is a real person behind that address.

The best solution will be incorporating all of these together. The easiest solution will be using one of the numerous paid services that do this for you. You could use the service for a few months and let them weed out the bad addresses.

Community
  • 1
  • 1
cegfault
  • 6,442
  • 3
  • 27
  • 49
0

It used to be possible, in the 1980's. These days, not so much.

I'd use an unique return address for each outgoing email, so bounce messages can be matched to the address that caused them.

Simon Richter
  • 28,572
  • 1
  • 42
  • 64
0

Years and years ago (decades, even) this is what the "finger" protocol was for. You could literally just ask a server for info about a user. However the internet quickly realized this was a pretty gaping security hole. Plus now with aliases and such, emails don't necessarily map to a username. So really the only way to be sure is send an email.

Chris Eberle
  • 47,994
  • 12
  • 82
  • 119