Echo,

I'm not sure what system you are using, but my latest RPMS use EL9, here <http://repo.whitehorsetc.com/9/testing/x86_64/>.

If you install the latest RPM it is bundled with patches one of which mitigated the 'invalid result' you are facing below:

# cat SOURCES/dspam-configureac-el9.patch
--- dspam/configure.ac  2024-02-27 23:28:25.005549848 -0700
+++ dspam.new/configure.ac      2024-02-27 23:28:39.015564302 -0700
@@ -556,6 +556,32 @@
 AC_MSG_RESULT([$enable_long_usernames])

 #
+#   Virtual users
+#
+AC_ARG_ENABLE(virtual-users,
+    [AS_HELP_STRING(--enable-virtual-users, Cause DSPAM to support virtual users.)])
+AC_MSG_CHECKING([whether to enable virtual users])
+case x"$enable-virtual-users" in
+    xyes)   # debug output enabled explicity
+            ;;
+    xno)    # debug output disabled explicity
+            ;;
+    x)      # debug output disabled by default
+            enable-virtual-users=no
+            ;;
+    *)      AC_MSG_ERROR([unexpected value $enable_virtual_users for --{enable,disable}-virtual-users configure option])
+            ;;
+esac
+if test x"$enable-virtual-users" != xyes
+then
+    enable-virtual-users=no
+else
+    enable-virtual-users=yes    # overkill, but convenient
+    AC_DEFINE(VIRTUAL_USERS, 1, [Defined if virtual users is enabled])
+fi
+AC_MSG_RESULT([$enable_virtual_users])
+
+#
 #   Large scale implementation
 #
 AC_ARG_ENABLE(large-scale,

I've never run the user management interface.

Eric


On 12/13/2024 3:19 AM, echo zhang wrote:
Dear Dspam Community:

I am trying to use the Dspam (version: 3.10.2) as a spam detection engine, but I am having some trouble.

Specifically, I want to detect specific email samples by running Dspam directly on its own, rather than through integration with a mail server such as Postfix. I specified Mysql as the storage driver, and used the following commands for training and classification respectively:

    dspam --mode=teft --source=corpus --class=spam --feature=noise
    --user root < "test.eml"
dspam --classify  --user root --stdout < "test.eml"
However, the output remains:

    X-DSPAM-Result: root; result="Innocent"; class="Innocent";
    probability=0.0023; confidence=1.00; signature=N/A

Meanwhile, I saw the following error message in the log:

    Dec 13 15:55:21 8e5bc7350f88 dspam[17599]: bailing on error -2
    Dec 13 15:55:21 8e5bc7350f88 dspam[17599]: received invalid result
    (!DSR_ISSPAM && !DSR_ISINNOCENT): -2


I wonder why this happened and how to fix it. I would be grateful if someone could offer some help!

By the way, I actually don't really understand Dspam's user management mechanism. From my understanding, it seems that the classification model for different users is differentiated to meet differentiated mail processing needs. So how can I add users to Dspam while running it standalone? Where can I view a list of all users currently maintained by Dspam?

Finally, is there something like a default classifier? If not, does this mean that the Dspam that was just installed actually has no spam detection capability, since it has not been trained on user data?

Thanks.

Echo Zhang


_______________________________________________
Dspam-user mailing list
Dspam-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/dspam-user
_______________________________________________
Dspam-user mailing list
Dspam-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/dspam-user

Reply via email to