Comment 8 for bug 57921

I believe I have the same problem. One server of mine stopped to send
email 3 days ago. Today when I investigated if everything was
allrigth I found many exim4 process processing email, and some
zombies.

I have killed them all tried to restart exim4 to no avail.

I have started exim4 with "exim4 -d -q -qq". The last interesting
messages where:

193.136.166.70 in hosts_avoid_esmtp? no (option unset)
  SMTP>> EHLO zilda.tagus.ist.utl.pt
waiting for data on socket
read response data: size=83
  SMTP<< 250-mail.tagus.ist.utl.pt
         250-PIPELINING
         250-STARTTLS
         250-SIZE 0
         250 8BITMIME
193.136.166.70 in hosts_avoid_tls? no (option unset)
  SMTP>> STARTTLS
waiting for data on socket
read response data: size=19
  SMTP<< 220 ready for tls
initializing GnuTLS as a client
generating 512 bit RSA key...
selecting on subprocess pipes
selecting on subprocess pipes
...

I have done cat /proc/sys/kernel/random/entropy_avail and returned 0.
I will try to recover the server without rebooting the machine or
turning off TLS on exim4.

Do you need more information?

    José Calhariz

--
 Um prisioneiro de guerra e um homem que tenta mata-lo, nao
 consegue e implora para voce que nao o mate.
  -- Winston Churchill