On 2006-10-08 Vincent Lefevre <[EMAIL PROTECTED]> wrote: > On 2006-10-07 15:08:52 +0200, Andreas Metzler wrote: > > Having this happen for just the first message sugggests is very > > strange. There is nothing in exim or gnutls that should behave > > differently for the second run. (Entropy starvage cannot be the cause > > either, as this is only a issue for *incoming* connections).
> Are you sure? Quite sure, yes. > I've just done a test on this machine, to which I've > opened 2 SSH connections. With one of them, I've sent a mail. With > the other one: [...] > The entropy dropped from 3462 to 4 just after I've sent the mail! For outgoing TLS exim reads from /dev/urandom which *does* drain entropy but switches to a prng instead of blocking when there is no more entropy. > Concerning the mail, it is still in the queue, but the exim4 log file > doesn't show any error! And "exim -qff" doesn't solve the problem. > Here are the contents of /var/log/exim4/mainlog: [...] > 2006-10-08 00:22:13 1GWKYf-00043q-HN <= [EMAIL PROTECTED] U=lefevre P=local > S=778 [EMAIL PROTECTED] > 2006-10-08 00:22:23 Start queue run: pid=15622 > 2006-10-08 00:22:23 1GWKYf-00043q-HN Spool file is locked (another process is > handling this message) [...] What does exiwhat say? cu andreas -- The 'Galactic Cleaning' policy undertaken by Emperor Zhark is a personal vision of the emperor's, and its inclusion in this work does not constitute tacit approval by the author or the publisher for any such projects, howsoever undertaken. (c) Jasper Ffforde -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
