Il 26/01/2012 13:15, Wietse Venema ha scritto:
Do you have quantitative evidence, based on actual
email traffic, that this would make a difference?
Wietse
Hi Wietse, thanks for your reply ..i've taken some time to collect some
data.
I've monitored for 2 days one single host wich handles about 200 domains
(it cleans mails for ..and forward to external mbox server). I've
noticed about 2000 remote mailbox verify tasks; redundancy checks
"caused" by case-sensitive re-checks are about 65-70%
About every mailbox has 2-3 different case variants.
I know and I agree on that it would not make sensitive differences on
performances but I've some other behaviours in mind (consider also that
remote host mailboxes could not be under my/our direct control):
- if remote hosts are temporarily unreachable ..or host is backup for ..
'case variants' re-checks are little bit more boring
- we can have different case variants with different cache values ..some
accepted and some blocked (this is the case in wich I've kicked my head
into)
- logs are a little bit more dirty
- remote host with rate-checks (tcp connections, smtp ones ..) .. I
know postfix reuses the single connection for delivery, but I mean on
negative results
- .. maybe others ..
On other side I'm also considering now that remote hosts could
specifically implement some particular policies (..like rejecting Ucase
variants, etc..) wich could bring us into problems.
Maybe a "little parameter" .. address_verify_casesensitive = yes/no ? :-)))
I'm joking..
Thanks for your time, I'm really interested what you think.
Have a nice week-end
Amedeo
Amedeo Rinaldo
--------------
Una volta eliminato l'impossibile, quello che resta, per improbabile che
sia, deve essere la verità .
