[exim] Re: autoreply and DKIM signature ?

Hi! > On 2024-08-14, Jeremy Harris via Exim-users wrote: > > On 14/08/2024 15:27, Kurt Jaeger via Exim-users wrote: > >> So: user1@domain1 has an autoreply, and the autoreply > >> should be signed with dkim for domain1. > > > > I do not agree. > > The

[exim] Re: autoreply and DKIM signature ?

Hi! > Maybe signing it with some other domain works. Still the question > is: Can it also be signed with the domain one *wants* to use > for this kind of autoreply ? > > Would it hurt to implement a mechanism for this ? > > How does autoreply know which mail-from it should use ? > > I found thi

[exim] Re: autoreply and DKIM signature ?

Hi! > On 14/08/2024 15:27, Kurt Jaeger via Exim-users wrote: > > So: user1@domain1 has an autoreply, and the autoreply > > should be signed with dkim for domain1. > > I do not agree. > The DKIM RFC says that anyone can sign a message. Maybe signing it with some other

[exim] Re: autoreply and DKIM signature ?

Hi! > > On 14/08/2024 15:27, Kurt Jaeger via Exim-users wrote: > > > So: user1@domain1 has an autoreply, and the autoreply > > > should be signed with dkim for domain1. > > > > I do not agree. > > The DKIM RFC says that anyone can sign a message. >

[exim] Re: autoreply and DKIM signature ?

Hi! > On 14/08/2024 14:31, Kurt Jaeger via Exim-users wrote: > > The problem is that the autoreply driver looses the information > > on which sender is used to send out the mail (envelope-from is <> > > to avoid mail-loops). > > More to the point, there *is no* s

[exim] autoreply and DKIM signature ?

Hi! Recently, I came upon a problem without a solution: If I trigger an autoreply (for example because of vacations), how do I get the system to add a DKIM-header for the proper domain ? This post describes that the recipient of the autoreply can check the signature by looking for the d= field i

[exim] Re: Upon applying 4.96-1 on test, "Tainted arg 2" appears

Hi! > I applied 4.96-1 to our test systems and routing to the LISTSERVer > began to fail with "*Tainted arg 2* for listserv_transport transport > command: > > The transport is quite simple: > > # Hand off to LISTSERV lsv_admin script > > listserv_transport: > > driver = pipe > >

[exim] Re: New Exim Security Release 4.96.2

Hi! > Thus, IMO when these particular patches will be confirmed, > they will supply update. The question is, who will confirm that, That's the key problem. I contacted the libspf2 developer and he committed the fix from issue 44. But neither he nor the person who submitted the patch has enough in

[exim] Re: Mitigation statement for CVE-2023-42119

Hi! > Does anyone know who ZDI *is* ? What does the abbreviation stand for? ZDI stands for zero-day-initiative. https://www.zerodayinitiative.com/about/ https://nitter.net/thezdi -- p...@opsec.eu+49 171 3101372Now what ? -- ## subscription configuration (requi

[exim] Re: Mitigation statement for CVE-2023-42119

Hi! > What I take from this mitigation statement--Use a trustworthy DNS > resolver which is able to validate the data according to the DNS record > types--is that if our DNS service is solid, we are not vulnerable. Is this > accurate, or am I oversimplifying things? The mitigation statement f

[exim] Re: 4.96 -> 4.96: 'tainted search query is not properly quoted'

Hi! > > Am 13.06.2023 um 08:51 schrieb Axel Rau via Exim-users > > : > > tainted search query is not properly > I???m reading in the spec p 79: > "If tainted data is used in the query then it should be quuted by using the > ${quote_:} expansion operator appropriate for the > lookup.??? > > W