Bug 208226

Summary: [PATCH] mail/spamassassin: SABug 7164 - TxRep warnings
Product: Ports & Packages Reporter: Sascha Holzleiter <sascha>
Component: Individual Port(s)Assignee: Adam Weinberger <adamw>
Status: Closed Overcome By Events    
Severity: Affects Some People CC: sascha
Priority: --- Keywords: patch
Version: LatestFlags: bugzilla: maintainer-feedback? (adamw)
Hardware: Any   
OS: Any   
Attachments:
Description Flags
patch-bug7164 none

Description Sascha Holzleiter 2016-03-23 00:27:08 UTC 
Created attachment 168522 [details]
patch-bug7164

This one fixes these warnings while using the TxRep (AWL successor) plugin:

Use of uninitialized value $msgscore in addition (+) at /usr/local/lib/perl5/site_perl/Mail/SpamAssassin/Plugin/TxRep.pm line 1415.
Use of uninitialized value $msgscore in subtraction (-) at /usr/local/lib/perl5/site_perl/Mail/SpamAssassin/Plugin/TxRep.pm line 1415.
Comment 1 Adam Weinberger freebsd_committer freebsd_triage 2016-04-02 16:39:56 UTC 
Bug #7164 (upstream) hasn't been fully closed yet. What do you make of comment #4, where the commenter suspects the patch is causing data corruption?
Comment 2 Sascha Holzleiter 2016-04-02 19:20:48 UTC 
I actually think his table excerpt looks just fine. Maybe he was confused about the always changing content of the email column, but it just cycles through email, ip, domain and helo which is quite usual if i understood TxRep correctly.

That being said the TxRep Module seems to be quite bug riddled for an officially endorsed replacement for AWL and no one seems to be in a hurry to fix these.

I just thought it would be nice to fix it on our side as Spamassassin releases seem to have slowed down to very slow pace. But maybe it's better to wait for upstream to fix the whole module.
Comment 3 Sascha Holzleiter 2016-04-04 17:25:19 UTC 
Just for the curious: I just reverted back to AWL as the bugs keep piling up.
Yesterday i hit https://bz.apache.org/SpamAssassin/show_bug.cgi?id=7235 .

This module is clearly not ready to being deployed on a production machine.
In hindsight this should have been clear from the start as the PG schema files distributed with 3.4.1 were horribly wrong (just a copy of the MySQL files...)

So i think as stated in my last comment this should be fixed overall in upstream.
Sorry for the noise, i thought we could get this to work.
Comment 4 Adam Weinberger freebsd_committer freebsd_triage 2016-04-04 17:31:52 UTC 
For the record, v341.pre recommends TxRep over AWL.