Re: Missing LOGNAME

On Fri 28/Feb/2025 03:45:28 +0100 Sam Varshavchik wrote:
> Alessandro Vesely writes:
>> On Thu 27/Feb/2025 01:37:31 +0100 Sam Varshavchik wrote:
>>> Alessandro Vesely writes:
>>>
>>>> I tried with and without import.  I never get Resent-From:.  I added the 
>>>> following to resent.sh:
>>>>
>>>> echo "$HOME $LOCKSLEEP $DEFAULT $LOGNAME" >> /tmp/foo
>>>>
>>>> All values are printed, except LOGNAME.  Should I hardcode it?
>>>
>>> If you want to, sure.
>>>
>>> Depending on how maildrop gets invoked it will reset LOGNAME to what it 
>>> thinks is responsible for its invocation, by looking up its real uid. If it 
>>> getpwuid() fails, you'll get nothing. That is the most likely situation here 
>>> -- it's being invoked under some virtual uid that does not have a username 
>>> attached to it.
>>
>> I see.  I thought it could came from the transaction, with any extension the 
>> user name may have had.
>>
>> Grabbing Delivered-To is overkill, methinks.
> 
> No, maildrop does not parse the Delivered-To:.


I meant I could grab it with something like /^Delivered-To: (.*)/ and $MATCH1. 
It comes with the original recipient extension and capitalization. 
Alternatively, I can hard code the recipient and add -$EXT if $EXT is not set. 
Or forget about EXT?  In theory, Resent-From: can even contain the display-name.

I'm still not clear what would that data be used for.  However, I read a Best 
practices for forwarding email to Gmail[*] where they suggest to add forwarding 
headers, mentioning X-Forwarded-For: and X-Forwarded-To:.


> maildrop reads the real uid that invoked it, and attempts to look up the 
> username for that uid.
> 
> It's just that the uid must be the common denominator that relates to both. If 
> there's a username that's getting mapped to some numerical uid value, say via 
> userdb, but the uid value is not getting mapped to the username, then maildrop 
> effectively clears LOGNAME.


Got it.


Best
Ale
-- 

[*] https://support.google.com/mail/answer/175365?sjid=3371529845334100081-EU








_______________________________________________
courier-users mailing list
[email protected]
Unsubscribe: https://lists.sourceforge.net/lists/listinfo/courier-users