This is a discussion on Fetchmail error 451 4.1.8 in slack9 within the Slackware Linux Support forums, part of the Unix Operating Systems category; --> Simon <usenet@no-dns-yet.org.uk> wrote: >On Sun, 30 Nov 2003 02:21:05 -0900, Floyd Davidson <floyd@barrow.com> wrote: >> Simon <usenet@no-dns-yet.org.uk> wrote: >>>It's ...
| |||||||
| FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
| ||||
| Simon <usenet@no-dns-yet.org.uk> wrote: >On Sun, 30 Nov 2003 02:21:05 -0900, Floyd Davidson <floyd@barrow.com> wrote: >> Simon <usenet@no-dns-yet.org.uk> wrote: >>>It's fixed in Slackware 9.1 (generated by netconfig): >>> >>>127.0.0.1 localhost >>>192.168.0.5 dogbert.no-dns-yet.org.uk dogbert >> >> What happens if you *don't* have an ethernet card installed? > >I don't have such a machine to test with, but /etc/hosts does say this: > ># By the way, Arnt Gulbrandsen <agulbra@nvg.unit.no> says that 127.0.0.1 ># should NEVER be named with the name of the machine. It causes problems ># for some (stupid) programs, irc and reputedly talk. :^) I think Pat has had that, or something like it, in there since maybe Slackware 8.0 or so. Or, at least it says something like that in the sources, though I seem to recall that was different than what was on the installation CD. I also seem to recall that it only actually used the lo device temporarily while the system was being installed, and then (once everything was in place) did something to change it. What I don't remember is what it did then (and what I never did know what Slackware 9.0 does) if there is otherwise no interface to assign an FQDN to for that particular machine. >I assume, therefore, that it would only insert this line in /etc/hosts: > >127.0.0.1 localhost > >I could, of course, be wrong. I don't think a by-the-book installation of Slackware is going to end up with anything like the abomination that RedHat has foisted off on the unknowing. -- Floyd L. Davidson <http://web.newsguy.com/floyd_davidson> Ukpeagvik (Barrow, Alaska) floyd@barrow.com |
| |||
| Bartosz Oudekerk <bartosz@see_my_GnuPG-key.com> wrote: >["Followup-To:" header set to alt.os.linux.slackware.] Why do you continue trying to direct followups away from where others who are reading this thread can read it? >Floyd Davidson <floyd@barrow.com> is thought to have >typed the following text on 2003-11-30: > >> What happens if you *don't* have an ethernet card installed? >> >Who implied you had to have one installed? There are other types of >network connections, and without any of them there's no use for >fetchmail, and very limited use for sendmail. *You* implied exactly that: 127.0.0.1 localhost 192.168.0.5 dogbert.no-dns-yet.org.uk dogbert That dogbert sure doesn't look like a ppp0 interface to me... However, it makes *no* different whether it is or is not, my question still stands: What happens without an ethernet card installed? -- Floyd L. Davidson <http://web.newsguy.com/floyd_davidson> Ukpeagvik (Barrow, Alaska) floyd@barrow.com |
| |||
| Floyd Davidson <floyd@barrow.com> wrote: > Bartosz Oudekerk <bartosz@see_my_GnuPG-key.com> wrote: > >["Followup-To:" header set to alt.os.linux.slackware.] > Why do you continue trying to direct followups away from where > others who are reading this thread can read it? Because Odious Bart is in charge of reality and only he gets to decide where the world's messages go. But he would have a lot more credibility if he only used pgp when he actually needs to use it. cordially, as always, rm |
| |||
| In comp.os.linux.misc Floyd Davidson <floyd@barrow.com> wrote: > Simon <usenet@no-dns-yet.org.uk> wrote: > >On Sun, 30 Nov 2003 02:21:05 -0900, Floyd Davidson <floyd@barrow.com> wrote: > >> Simon <usenet@no-dns-yet.org.uk> wrote: > >>>It's fixed in Slackware 9.1 (generated by netconfig): > >>> > >>>127.0.0.1 localhost > >>>192.168.0.5 dogbert.no-dns-yet.org.uk dogbert > >> > >> What happens if you *don't* have an ethernet card installed? > > > >I don't have such a machine to test with, but /etc/hosts does say this: > > > ># By the way, Arnt Gulbrandsen <agulbra@nvg.unit.no> says that 127.0.0.1 > ># should NEVER be named with the name of the machine. It causes problems > ># for some (stupid) programs, irc and reputedly talk. :^) > > I think Pat has had that, or something like it, in there since > maybe Slackware 8.0 or so. Or, at least it says something like SInce 2.0, you mean. And he's right. Peter |
| |||
| -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Floyd Davidson <floyd@barrow.com> is thought to have typed the following text on 2003-11-30: > Bartosz Oudekerk <bartosz@see_my_GnuPG-key.com> wrote: >>["Followup-To:" header set to alt.os.linux.slackware.] > > Why do you continue trying to direct followups away from where > others who are reading this thread can read it? > Continue? This is my first post in this thread. Because I see no need to have this crossposted. Plus I explicitly stated in the body of my post, that the Followup-To header was being set, so that those who would indeed like to follow this thread can choose to do so and know where it'll be continued. >>Floyd Davidson <floyd@barrow.com> is thought to have >>typed the following text on 2003-11-30: >> >>> What happens if you *don't* have an ethernet card installed? >>> >>Who implied you had to have one installed? There are other types of >>network connections, and without any of them there's no use for >>fetchmail, and very limited use for sendmail. > > *You* implied exactly that: > *I* did no such thing, like I said this is my first post in this thread, judging by the FQDN used, it was simon who said that, and I still don't see how that implies an ethernet interface. > 127.0.0.1 localhost > 192.168.0.5 dogbert.no-dns-yet.org.uk dogbert > > That dogbert sure doesn't look like a ppp0 interface to me... > Why not? Because most ppp0 devices are configured by DHCP doesn't mean that all are, and how about wireless? I can, and I'm sure you too, can think of a few others, albeit more uncommon ones. > However, it makes *no* different whether it is or is not, my > question still stands: What happens without an ethernet card > installed? > Without an ethernet card installed you'd have to use some other type of network connection if you'd like to use fetchmail, unless you want to retrieve mail from localhost, but why'd you want to do that? - -- Bartosz Oudekerk Play Rogue, visit exotic locations, meet strange creatures and kill them. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.3 (GNU/Linux) iD8DBQE/yhWj256ZyNYAOpkRAvniAJ4i8zOlKnJEZ+FugeAxLNfad+rxuQ CbBtMx bP7ZMuRwNP62jCl9+MI86MU= =YT7a -----END PGP SIGNATURE----- |
| |||
| In article <87r7zqky7z.fld@barrow.com>, Floyd Davidson wrote: A lot of good advice; accurate, as always. But consider that the OP saw all those messages from fetchmail without understanding them. A user at that level has much learning to do. In the meantime he'll need to receive mail. So I'd start him here: > Read the man page for fetchmail, where the item you are interested in > is, > > -Z <nnn>, --antispam <nnn[, nnn]...> but instead point him to this section: -m <command> | --mda <command> (Keyword: mda) You can force mail to be passed to an MDA directly (rather than forwarded to port 25) with the --mda or -m option. To avoid losing mail, use this option only with MDAs like procmail ... .... so he won't have to struggle with sendmail nor DNS (other than to resolve the name of the pop3 server, which apparently works) right now. The requisite .fetchmailrc line is this: mda "/usr/bin/procmail -t -d $USER" (probably substituting the local username for $USER; I am not sure.) With this you don't get into the complexities of MTA spam filtering, but you're still free to play with procmail filtering at your leisure. Such difficulties as these in a user's early GNU/Linux days often lead to frustration and burnout. -- /dev/rob0 - preferred_email=i$((28*28+28))@softhome.net or put "not-spam" or "/dev/rob0" in Subject header to reply |
| |||
| /dev/rob0 <rob0@gmx.co.uk> wrote: >In article <87r7zqky7z.fld@barrow.com>, Floyd Davidson wrote: > >A lot of good advice; accurate, as always. But consider that the OP saw >all those messages from fetchmail without understanding them. A user at >that level has much learning to do. In the meantime he'll need to >receive mail. So I'd start him here: Too me, that looks more complex. :-) He *has* to fix his DNS so that sendmail will work. That is screwing up a lot more than sendmail. It appears that otherwise sendmail is just fine, and with the addtion of a fairly simple ~/.fetchmailrc file, so is fetchmail. Different strokes for different folks though. >> Read the man page for fetchmail, where the item you are interested in >> is, >> >> -Z <nnn>, --antispam <nnn[, nnn]...> > >but instead point him to this section: > -m <command> | --mda <command> > (Keyword: mda) You can force mail to be passed to > an MDA directly (rather than forwarded to port 25) > with the --mda or -m option. To avoid losing mail, > use this option only with MDAs like procmail ... > >... so he won't have to struggle with sendmail nor DNS (other than to >resolve the name of the pop3 server, which apparently works) right now. > >The requisite .fetchmailrc line is this: > mda "/usr/bin/procmail -t -d $USER" >(probably substituting the local username for $USER; I am not sure.) >With this you don't get into the complexities of MTA spam filtering, but >you're still free to play with procmail filtering at your leisure. > >Such difficulties as these in a user's early GNU/Linux days often lead >to frustration and burnout. >-- > /dev/rob0 - preferred_email=i$((28*28+28))@softhome.net > or put "not-spam" or "/dev/rob0" in Subject header to reply -- Floyd L. Davidson <http://web.newsguy.com/floyd_davidson> Ukpeagvik (Barrow, Alaska) floyd@barrow.com |
| |||
| ptb@oboe.it.uc3m.es (P.T. Breuer) wrote: >In comp.os.linux.misc Floyd Davidson <floyd@barrow.com> wrote: >> Simon <usenet@no-dns-yet.org.uk> wrote: >> >On Sun, 30 Nov 2003 02:21:05 -0900, Floyd Davidson <floyd@barrow.com> wrote: >> >> Simon <usenet@no-dns-yet.org.uk> wrote: >> >>>It's fixed in Slackware 9.1 (generated by netconfig): >> >>> >> >>>127.0.0.1 localhost >> >>>192.168.0.5 dogbert.no-dns-yet.org.uk dogbert >> >> >> >> What happens if you *don't* have an ethernet card installed? >> > >> >I don't have such a machine to test with, but /etc/hosts does say this: >> > >> ># By the way, Arnt Gulbrandsen <agulbra@nvg.unit.no> says that 127.0.0.1 >> ># should NEVER be named with the name of the machine. It causes problems >> ># for some (stupid) programs, irc and reputedly talk. :^) >> >> I think Pat has had that, or something like it, in there since >> maybe Slackware 8.0 or so. Or, at least it says something like > >SInce 2.0, you mean. And he's right. Probably was. I use Slackware, but I've never used the default networking configuration files. I've been writing my own since SLS. I think that I'd never looked at the default /etc/hosts file or the /etc/rc.d/rc.inet* files until a couple years ago when somebody said something to the effect that Slackware did (or didn't) use the same stupid setup that RedHat and followers do, and I went an looked at the source distribution and saw that. Whatever, sometimes it's hard to convince people that 127.0.0.1 shouldn't have *anything* other than localhost associated with it. But I'm always afraid to tell anyone with a RedHat system to do it right, because I have no idea what RH munged to make that work without problems. -- Floyd L. Davidson <http://web.newsguy.com/floyd_davidson> Ukpeagvik (Barrow, Alaska) floyd@barrow.com |
| |||
| One other nitpick / question: In article <87r7zqky7z.fld@barrow.com>, Floyd Davidson wrote: > There are other ways to do it. One other is to use a "dummy" > network interface, which acts just like a loopback and can be IIUC a dummy interface is like a /dev/null for network packets. Packets come in, but they don't go out. In contrast a loopback returns them. They are analogous to celestial black holes and mirrors, respectively. Is this understanding correct? It appears so: #v+ /* dummy.c: a dummy net driver The purpose of this driver is to provide a device to point a route through, but not to actually transmit packets. #v- -- /dev/rob0 - preferred_email=i$((28*28+28))@softhome.net or put "not-spam" or "/dev/rob0" in Subject header to reply |
| ||||
| On Sun, 30 Nov 2003 08:54:42 -0800, /dev/rob0 <rob0@gmx.co.uk> wrote: > > > In article <87r7zqky7z.fld@barrow.com>, Floyd Davidson wrote: > > A lot of good advice; accurate, as always. But consider that the OP saw > all those messages from fetchmail without understanding them. A user at > that level has much learning to do. In the meantime he'll need to > receive mail. So I'd start him here: > >> Read the man page for fetchmail, where the item you are interested in >> is, >> >> -Z <nnn>, --antispam <nnn[, nnn]...> > > but instead point him to this section: > -m <command> | --mda <command> > (Keyword: mda) You can force mail to be passed to > an MDA directly (rather than forwarded to port 25) > with the --mda or -m option. To avoid losing mail, > use this option only with MDAs like procmail ... > > ... so he won't have to struggle with sendmail nor DNS (other than to > resolve the name of the pop3 server, which apparently works) right now. > > The requisite .fetchmailrc line is this: > mda "/usr/bin/procmail -t -d $USER" > (probably substituting the local username for $USER; I am not sure.) > With this you don't get into the complexities of MTA spam filtering, but > you're still free to play with procmail filtering at your leisure. in my ~/.fetchmailrc it's: and wants mda "/usr/bin/formail -ds /usr/bin/procmail" You can specify a procmailrc by just listing it ...../bin/procmail filename" AC |