It occurs to me that using mailertable on a sendmail gateway,
ie.:
blah.domain.com smtp: [1.2.4.5]:[1.2.3.6]
Should enable some kind of fail-over (No DNS/MX available), what
happens if both are available, will sendmail round-robin between
both? Didn't found anything in the docs about the problem, simply
point me to the doc if I just missed the proper one!
Regards
--
Michael Heiming (X-PGP-Sig > GPG-Key ID: EDD27B94)
mail: echo zvp...@urvzvat.qr | perl -pe 'y/a-z/n-za-m/'
#bofh excuse 68: only available on a need to know basis
> It occurs to me that using mailertable on a sendmail gateway,
> ie.:
> blah.domain.com smtp: [1.2.4.5]:[1.2.3.6]
> Should enable some kind of fail-over (No DNS/MX available), what
> happens if both are available, will sendmail round-robin between
> both? Didn't found anything in the docs about the problem, simply
> point me to the doc if I just missed the proper one!
Looking a bit deeper into the archives, found a few posts by
Claus/Per giving a little more insight. Anyway it works great and
doesn't do any load-balancing as desired, but just uses the first
available. ;-)
--
Michael Heiming (X-PGP-Sig > GPG-Key ID: EDD27B94)
mail: echo zvp...@urvzvat.qr | perl -pe 'y/a-z/n-za-m/'
#bofh excuse 252: Our ISP is having
{switching,routing,SMDS,frame relay} problems
> > blah.domain.com smtp: [1.2.4.5]:[1.2.3.6]
> > Should enable some kind of fail-over (No DNS/MX available), what
> > happens if both are available, will sendmail round-robin between
> > both? Didn't found anything in the docs about the problem, simply
> > point me to the doc if I just missed the proper one!
See doc/op/op.*. Unfortunately it's hidden fairly well :-(
If the mailer is the built-in IPC mailer, the host may be a
colon-separated list of hosts that are searched in order for the first
working address (exactly like MX records).
> Looking a bit deeper into the archives, found a few posts by
> Claus/Per giving a little more insight. Anyway it works great and
> doesn't do any load-balancing as desired, but just uses the first
> available. ;-)
If you look at the source code, you'll find an answer
(which, however, is unsupported and undocumented).
--
Note: please read the netiquette before posting. I will almost never
reply to top-postings which include a full copy of the previous
article(s) at the end because it's annoying, shows that the poster
is too lazy to trim his article, and it's wasting the time of all readers.
> In comp.mail.sendmail Michael Heiming <michael...@www.heiming.de>:
> > Hi!
>
> > It occurs to me that using mailertable on a sendmail gateway,
> > ie.:
>
> > blah.domain.com smtp: [1.2.4.5]:[1.2.3.6]
>
> > Should enable some kind of fail-over (No DNS/MX available), what
> > happens if both are available, will sendmail round-robin between
> > both? Didn't found anything in the docs about the problem, simply
> > point me to the doc if I just missed the proper one!
>
> Looking a bit deeper into the archives, found a few posts by
> Claus/Per giving a little more insight. Anyway it works great and
> doesn't do any load-balancing as desired, but just uses the first
> available. ;-)
For Future Readers:
FYI there is "wide" patch that allows sendmail to use *another mailer*
as fallback. AFAIR debian containd sendmail-wide package (sendmail with
"wide" patch applied).
--
Andrzej [en:Andrew] Adam Filip an...@priv.onet.pl an...@xl.wp.pl
http://www.sendmail.org/faq/ http://anfi.homeunix.net/sendmail/
[Using mailertable for load-balancing]
> If you look at the source code, you'll find an answer
> (which, however, is unsupported and undocumented).
Another (very dirty) trick is to use a SOCKETMAP for the mailertable
map, and have the socketmap server return the list of servers in
random order each time. If your mailertable needs are not too complex,
it's pretty easy to code something like this up.
Regards,
David.
>> > blah.domain.com smtp: [1.2.4.5]:[1.2.3.6]
>> > Should enable some kind of fail-over (No DNS/MX available), what
>> > happens if both are available, will sendmail round-robin between
>> > both? Didn't found anything in the docs about the problem, simply
>> > point me to the doc if I just missed the proper one!
> See doc/op/op.*. Unfortunately it's hidden fairly well :-(
> If the mailer is the built-in IPC mailer, the host may be a
> colon-separated list of hosts that are searched in order for the first
> working address (exactly like MX records).
Strange enough, the sendmail docu on the system I checked, not
the FreeeBSD system running sendmail in question, RHEL 3 misses
most sendmail docu. There had been a sendmail-docu*.rpm, but it
seems to have gone.;(
>> Looking a bit deeper into the archives, found a few posts by
>> Claus/Per giving a little more insight. Anyway it works great and
>> doesn't do any load-balancing as desired, but just uses the first
>> available. ;-)
> If you look at the source code, you'll find an answer
> (which, however, is unsupported and undocumented).
Hi Claus!
Seems as if I missed the obvious in the heat of the moment.
However it's working fine now.;)
Thx to anyone else who replied.
--
Michael Heiming (X-PGP-Sig > GPG-Key ID: EDD27B94)
mail: echo zvp...@urvzvat.qr | perl -pe 'y/a-z/n-za-m/'
#bofh excuse 458: Windows 2000 is still light years away from
the stability and reliability of UNIX systems.