We were testing SEM in weekly masscheck for 2+ years now, which has
been entirely fine.
If we have been testing it for two years, isn't it time to cease the
tests? What are we hoping to gain? FYI, I don't mind that the information
http://www.spamtips.org/2011/03/sem-rules-mistakenly-enabled-how-to.html
Unfortunately, something else changed in our codebase that mistakenly
auto-promoted the SEM rules to active back in March.
http://www.spamtips.org/2011/06/emergency-sem-rules-mistaken-enabled.html
And now it is back again, except as T_ rules, which is just as bad
because it is causing an unexpected flood of DNS traffic to SEM.
It's back as a T_rule because someone removed(?) the nopublish flag.
SEM was supposed to be a net nopublish rule.
So the publication as T_RULE is not a bug in the code but "correct"
behavior because the sandbox is missing the nopublish flag because of
bug 6297 and bug 6257.
My guess is that next time, whoever did this should leave nopublish and
fix the bug you are trying to fix :-(
I just committed the 30_bug_6220_sem.cf file in SMF's sandbox as an
emergency update.
A make build_rules on trunk no longer includes SEM rules in
72_active.cf. It is also still missing from 70_sandbox.cf.
We need an emergency rule update to stop this flood, then to
investigate why our auto-rule promotion code is still broken.
Based on another email, there is no concern we are flooding their
network. No need for an emergency update.
Regards,
KAM
