Hello,
I have checked the FAQ and docs but have been unable to find the answer to
this question in more detail. The question is why is -f no longer able to
be turned off as stated in Changes doc, spamc.c and man spamc?
If spamd dies, I'd like to defer delivery until it is started back up and
not deliver possible spam during the down period.
Procmail Example
:0 ifW
* ^X-Spam-Status: Yes
| spamc || exit 75
I have changed SAFE_FALLBACK in spamc.c under a test server to off
and have run this using the above procmail recipe. Whem spamd dies,
messages simply get deferred in queue until it is backup and next queue
run processes.
Procmail 'seems' to indicate that it 'successfully recovered unfiltered
data' upon receiving a non-zero exit code and after final
delivery...deferred messages appear to be intact.
Is there a danger in this scenario that is not apparent? Man spamc states
'connect failures to spamd will cause message delivery failure'. If it is
kicked back with 75 EX_TEMPFAIL for another try at a later time, where
does the potential delivery failure occur?
I posted this before but received no reply, so I assume this question is
either too trivial, obvious and/or previously covered . If so is there
another searchable archive for this list available
Your search - safe_fallback site:geocrawler.com archives/3/11679 - did not
match any documents. No pages were found containing "safe_fallback".
Your search - spamc -f site:geocrawler.com archives/3/11679 - did not
match any documents
Regards
In Changes Doc
2002-01-15 10:48 hughescr
* spamd/: spamc.c, spamc.pod: -f flag defaults to on now
in spamc.c
int SAFE_FALLBACK=1; /* default to on now - CRH */
in man spamc
-f ... Without this flag, connection failures to spamd will cause message
delivery failures.
This now defaults to on, and can't be turned off. This flag is accepted
though for backwards-compatibility.
_______________________________________________
Spamassassin-talk mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
