-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 El 2017-11-21 a las 10:29 -0000, Daniel Morris escribió:
On Tue, Nov 21, 2017 at 09:18:30AM +0000, Ianseeks wrote:
provider. Maybe your provider, but I don't detect their headers. Maybe these:
X-OWM-Source-IP: 31.*.*.* (GB) X-OWM-Env-Sender: bing....@btinternet.com X-Junkmail-Premium-Raw: score=8/50,refid=2.7.2:2017.11.20.172117:17:8.317,ip=,rules=__HAS_FROM, __FRAUD_WEBMAIL_FROM, FROM_NAME_ONE_WORD, __TO_MALFORMED_2, __TO_NO_NAME, __SUBJ_ALPHA_END, __HAS_MSGID, __SANE_MSGID, INVALID_MSGID_NO_FQDN, __MIME_VERSION, __CTE, __CT, __CT_TEXT_PLAIN, SUPERLONG_LINE, __NO_HTML_TAG_RAW, BODY_SIZE_600_699, BODYTEXTP_SIZE_3000_LESS, __MIME_TEXT_P1, __MIME_TEXT_ONLY, HTML_00_01, HTML_00_10, BODY_SIZE_5000_LESS, __FRAUD_WEBMAIL, NO_URI_FOUND, NO_CTA_URI_FOUND, __PHISH_SPEAR_STRUCTURE_1, BODY_SIZE_1000_LESS, BODY_SIZE_2000_LESS, __MIME_TEXT_P, NO_URI_HTTPS, BODY_SIZE_7000_LESS
Its the same provider i use to send emails to all lists so why no others get "spammed", i have no idea. I didn;t realise that your own provider tags sent emails as spam, i thought it was done on receipt of an email.
Spam scoring & tagging is often done at the recipient's MTA, but there's no reason it can't be done at other stages as email gets relayed between machines. There's obviously a trust issue (rogue senders could spoof the spam-scoring as well) so in most environments only the last hop counts, and that helps with lightweight/mobile clients too.
I guess as BT is a large ISP with a variable community of users they may be testing "outbound" mail to try and prevent some abuses escaping their network and damaging their reputation?
BT seems to have done this for a long while with your outbound postings, and you typically seem to score 8-10/50 when sending :) I expect your local/last-hop-inbound spam categorisers have changed a ruleset recently which is why your mail was marked as SPAM locally to you. You'll need to inspect the headers on the received message on your machine, as when your reply goes back out to the list and is received <everywhere else> then whatever is running <everywhere else> will have overwritten the X-Spam-.... tags locally (or been stripped en-route).
For example, your mail tagged as "[SPAM]" sent back to the list only scored -2.5 for me, with a required of 5.0, scored by SpamAssassin.
If you are doing Bayesian filtering, when was the last time you trained your local categoriser with recent ham, as well as spam?
In this case, his own possible filtering is not involved at all: it is the oubound email sent to the list which got tagged somewhere "en route". It could be at his ISP or at another server. Not the ones at opensuse.org because we know there is no such kind of filtering that modifies the subject line. - -- Cheers Carlos E. R. (from openSUSE 42.2 x86_64 "Malachite" (Minas Tirith)) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iF4EAREIAAYFAloUSQEACgkQja8UbcUWM1ztCQD+O/uhMOWif//vKyg1DvrYg+4V 79RhlrRrXm2OYfSo1ncA/idlPdNBncuXDlYB4+A0HnDnAJYyxGHHwgFvv4Ah7s2f =KALD -----END PGP SIGNATURE-----