| opendevreview | Merged openstack/project-config master: Stop checking for wheel during release uploads https://review.opendev.org/c/openstack/project-config/+/1003189 | 00:19 |
|---|---|---|
| opendevreview | Merged openstack/project-config master: Retire cinderlib https://review.opendev.org/c/openstack/project-config/+/993184 | 14:21 |
| gouthamr | what could make a mail like this not be delivered to folks (i checked with a few)? DMARC mitigations on the list? https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/XA7SK7KAW72Y6D3WXDHVBC73WVW2GSCL/ | 22:35 |
| fungi | i don't remember off the top of my head what domain simon's e-mail uses, but if his domain sets strict dmarc enforcement rules in dns and his dkim signature covers some part of the message that can't pass cleanly through mailman without getting inadvertently altered (i've occasionally seen some signed headers use weird whitespace that gets normalized by mailman causing the | 22:42 |
| fungi | signature to no longer be valid), and then if your mail provider obeys the dmarc rules asserted by simon's domain, the message could get rejected | 22:42 |
| fungi | years ago we debugged some cases of this by recording verbatim copies of the messages as they arrived at the list server's mta and then doing side-by-side comparisons with what came back out of mailman to spot the alterations. we turn off all the features that intentionally alter messages by appending footers to the body or prepending tags to the subject or resetting reply-to | 22:46 |
| fungi | headers, but unfortunately some of the python mail processing libraries mailman relies on (in cpython stdlib i think?) are not 100% clean when you pass some messages through them, and even something as trivial as a difference in the number of spaces a header line continuation is indented by can a dkim signature to no longer match | 22:46 |
Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!