Hi folks,

problem using opensmtpd 6.0.3 on Debian: I have set

        xname=mailhost.example.com
        :
        limit mta inet4
        listen on lo tls pki $xname
        listen on eth0 tls pki $xname
        listen on eth1 tls pki $xname
        :

in smtpd.conf, but apparently this "limit" line ignored. At boot
time it tries to bind to a link-local IPv6 address. And this fails,
as strace shows:

:
695   setsockopt(11, SOL_SOCKET, SO_REUSEADDR, [1], 4 <unfinished ...>
695   <... setsockopt resumed> )        = 0
695   bind(11, {sa_family=AF_INET, sin_port=htons(25), 
sin_addr=inet_addr("10.0.0.2")}, 16 <unfinished ...>
695   <... bind resumed> )              = 0
695   socket(PF_INET6, SOCK_STREAM, IPPROTO_IP <unfinished ...>
695   <... socket resumed> )            = 12
695   setsockopt(12, SOL_SOCKET, SO_REUSEADDR, [1], 4 <unfinished ...>
695   <... setsockopt resumed> )        = 0
695   setsockopt(12, SOL_IPV6, IPV6_V6ONLY, [1], 4 <unfinished ...>
695   <... setsockopt resumed> )        = 0
695   bind(12, {sa_family=AF_INET6, sin6_port=htons(25), inet_pton(AF_INET6, 
"fe80::216:3aff:feca:38bb", &sin6_addr), sin6_flowinfo=0, 
sin6_scope_id=if_nametoindex("eth0")}, 28 <unfinished ...>
695   <... bind resumed> )              = -1 EADDRNOTAVAIL (Cannot assign 
requested address)
695   write(2, "pony express: smtpd: bind: Canno"..., 59 <unfinished ...>
695   <... write resumed> )             = 59
695   exit_group(1)                     = ?
695   +++ exited with 1 +++


"ip address show" (run immediately before smtpd) shows, that
the link-local address is still "tentative":

:
194: eth0@if195: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state 
UP group default qlen 1000
    link/ether 00:16:3a:ca:38:bb brd ff:ff:ff:ff:ff:ff
    inet 10.0.0.2/24 brd 10.0.0.255 scope global eth0
       valid_lft forever preferred_lft forever
    inet6 fe80::216:3aff:feca:38bb/64 scope link tentative
       valid_lft forever preferred_lft forever
:

If I try to run smtpd later, then it works.


Every helpful comment is highly appreciated.
Harri

-- 
You received this mail because you are subscribed to [email protected]
To unsubscribe, send a mail to: [email protected]

Reply via email to