smtpRetryTimeMaxExp

The smtpRetryTimeMaxExp option limits the increase in waiting time between two repeated mail dispatch attempts. The waiting time normally doubles with every failed dispatch attempt in order to restrict CPU usage caused by the dispatch attempts during persistent dispatch problems. After doubling smtpRetryTimeMaxExp, the waiting time remains at the value reached.

smtpRetryTimeMaxExp

<value>

<value>

Default setting: 6

If errors occur while trying to connect to the SMTP mail server, double smtpRetryTimeBase is continually used as the waiting time until a renewed delivery attempt is made.

If the error occurs later in the SMTP dialog, which could possibly mean that the problem is not a general server problem that is quickly noticed, but a mail-specific problem that is often only noticed after some time, then the waiting time between two dispatch attempts (beginning with smtpRetryTimeBase) doubles with every attempt until doubling of the smtpRetryTimeMaxExp is reached. The default maximum waiting time between two delivery attempts is therefore as follows:

Maximum waiting time = smtpRetryTimeBase * 2^{smtpRetryTimeMaxExp} = 15m * 2⁶ = 960m = 16h

Examples

1. Mail server not accessible

   Renewed delivery attempt after 30 min= 2 * 15m

2. Connection setup to mail server possible; mail-specific error:

   • Renewed delivery attempt after 15 min= 15m * 2⁰

   • Renewed delivery attempt after 30 min= 15m * 2¹

   • Renewed delivery attempt after 1 h = 15m * 2²

   • Renewed delivery attempt after 2 h = 15m * 2³

   • Renewed delivery attempt after 4 h = 15m * 2⁴

   • Renewed delivery attempt after 8 h = 15m * 2⁵

   • All further delivery attempts after 16 h = 15m * 2⁶

   until maxQueueLifeTime is reached (default: 5 days).

When reducing the smtpRetryTimeBase, the value for smtpRetryTimeMaxExp should be increased at the same time; otherwise repeated frequent delivery attempts overload the CPU.