[exim] Re: Exim 4.97, DKIM, bad memory allocation

Quoting Sander Smeenk via Exim-users (exim-users@lists.exim.org): > exim[257275]: 1tt2qM-00014pp-1Y7f DKIM: signing failed: LONG_LINE > exim[257275]: 1tt2qM-00014pp-1Y7f bad memory allocation requested > (-1518335015 bytes) from b64encode_taint 250 Caught two of these messa

[exim] Exim 4.97, DKIM, bad memory allocation

Hi, Some Dutch telecoms provider recently sent a mailing that makes my (self compiled, Ubuntu pkg based) Exim 4.97 choke in some DKIM routines. I'm sharing what i have right now and working on hopefully digging up more information. Message comes in: exim[256917]: 1tt2qM-00014pp-1Y7f DKIM: va

[exim] Re: ARC

Quoting Jeremy Harris via Exim-users (exim-users@lists.exim.org): > On 23/12/2024 08:49, Kai Bojens via Exim-users wrote: > > - Exim 4.98 - rebuild from Debian Backports with ARC enabled > > Glad to hear someone's using ARC. Do you have it operating > for plain forwarding? Mailinglists?Any i

[exim] Re: Setting dkim_verify_minimal to true, does not terminate validation after first success

Quoting Jeremy Harris via Exim-users (exim-users@lists.exim.org): > Thanks. a5e7a642059e is an initial go; I'd appreciate your > evaluation. I've not looked into any effect it has on DMARC, > only that it seems to be doing the right thing for one DKIM test. Awesome. This does exactly what i wou

[exim] Re: Setting dkim_verify_minimal to true, does not terminate validation after first success

Quoting Jeremy Harris via Exim-users (exim-users@lists.exim.org): > This will be more work than is feasible for the upcoming release. > I'll start with the 4.next branch. Can you build from git and test > any changes I might invent? Yes, that should not be a problem. As long as this message i ha

[exim] Re: Setting dkim_verify_minimal to true, does not terminate validation after first success

Quoting Jeremy Harris via Exim-users (exim-users@lists.exim.org): > > | LOG: 1sQq1q-00BKvo-0I > > dkim--pas...@bf02x.hubspotemail.netbf02x.hubspotemail.net-...@bf02x.hubspotemail.net-- > > | LOG: 1sQq1q-00BKvo-0I > > dkim--none--lease-a-bike.nllease-a-bike.n...@lease-a-bike.nl-- > > Notic

[exim] Setting dkim_verify_minimal to true, does not terminate validation after first success

Hi, I had 'dkim_verify_minimal = true' in my Exim config and according to https://www.exim.org/exim-html-current/doc/html/spec_html/ch-main_configuration.html "If set to true, verification of signatures will terminate after the first success.". But this does not seem to be what is happening and

[exim] Re: Exim hates CNAMEs, not IPv6

Quoting John Levine via Exim-users (exim-users@lists.exim.org): > >That text is obsolete. There is no such text in RFC5321, and > >CNAME-valued mail domains have long been OK. The sending MTA (its DNS > >resolver) is expected to restart the MX lookup at the target of the > >CNAME, and if no MX r

[exim] Re: Exim, DMARC and Resent-From:

Quoting Jeremy Harris via Exim-users (exim-users@lists.exim.org): > > For years now i've been running with a custom Exim binary that has SPF, > > DKIM and DMARC support compiled in. > Using what dmarc library version? Any issue is probably there, > in that library. Gah. Forgot that info. :/ Thi

[exim] Exim, DMARC and Resent-From:

Hi there lovely gentlepeople, For years now i've been running with a custom Exim binary that has SPF, DKIM and DMARC support compiled in. I actually reject messages based on failing SPF + failing DKIM /and/ a DMARC policy that states 'reject'. This has worked fine for years. Other than the occasio

[exim] Re: Unexpected 're-routed to' in require verify = recipient?

Quoting Sander Smeenk via Exim-users (exim-users@lists.exim.org): > I try to send a message to 'i...@email.postcodeloterij.nl', but Exim > unexpectedly changes that to 'i...@postcodeloterij.slgnt.eu', based on > some DNS queries it seems? I've reopened https://

[exim] Re: Unexpected 're-routed to' in require verify = recipient?

Quoting Evgeniy Berdnikov via Exim-users (exim-users@lists.exim.org): > Yes, it seems bug is triggered by chain longer then one alias, and if > dnslookup router is encountered in router chain. Here is my example: > > # exim4 -bt du...@email.tedisk.ru > du...@email.ascit.ru > <-- du...@email

[exim] Re: Unexpected 're-routed to' in require verify = recipient?

Quoting Evgeniy Berdnikov via Exim-users (exim-users@lists.exim.org): > I treat the change of target domain as Exim's bug and tried to reproduce it > for test alias chains, with no success. But it definitely triggers for > email.postcodeloterij.nl, that's intriguing. Another example: 'foo...@e

[exim] Re: Unexpected 're-routed to' in require verify = recipient?

Quoting Tim Jackson via Exim-users (exim-users@lists.exim.org): > > I try to send a message to 'i...@email.postcodeloterij.nl', but Exim > > unexpectedly changes that to 'i...@postcodeloterij.slgnt.eu', based on > > some DNS queries it seems? > > I'm definitely not an expert, but is this not expe

[exim] Re: Unexpected 're-routed to' in require verify = recipient?

Quoting Julian Bradfield via Exim-users (exim-users@lists.exim.org): > dnslookup: > driver = dnslookup > transport = remote_smtp > user = Debian-exim > same_domain_copy_routing = yes > no_more Thanks again, Julian. I was about to post this minimal example myself. ;) Also, if you do add

[exim] Re: Unexpected 're-routed to' in require verify = recipient?

Quoting Julian Bradfield via Exim-users (exim-users@lists.exim.org): > >> I'd rather it did not do this. Since the addres *i gave it* works, and > >> the address *it made up itself* fails delivery with '550 user unknown'. > > You have some explicit coding in your config to do this. > > Your config

[exim] Unexpected 're-routed to' in require verify = recipient?

Hi list, I try to send a message to 'i...@email.postcodeloterij.nl', but Exim unexpectedly changes that to 'i...@postcodeloterij.slgnt.eu', based on some DNS queries it seems? | $ exim -f ssme...@freshdot.net -bt i...@email.postcodeloterij.nl | i...@postcodeloterij.slgnt.eu | <-- i...@email.p