Am 07.11.2017 um 22:46 schrieb Colony.three: >> So much for that theory. There are about a million of these in the logfile. >> It's not making its own socket for unknown reasons which may be New To >> Science. >> >> well, that looks like clamd is restarted again and again because it's >> failing, most likely /run/clamd.scan/ don't exist or has the wrong >> permissions
Correct, /run/clamd.scan/ does -not- exist. Why? Because the clamd@ service destroys it on restart. (along with its socket) I could merrily re-create the directory and socket all day long, but on restart it would only be wiped out again. Of course the permissions are correct, as per above; it couldn't have been destroyed by the service otherwise. >> "systemctl status" as well as the syslogs should tell you that the >> service is failing if you just look at it >> >> are you aware that /run is a tmpfs and hence anything below does not >> survive a reboot? Why yes, I am. As I say, clamd destroys its socket directory on stop, but then does not re-create it on start, like it's supposed to. This is the problem which I have been trying to explain. The problem is that clamd destroys its socket directory and socket on stop, but then fails to re-create these on start. It is not making its own socket to communicate, even after the sig database is complete, and when it goes to look for its socket, oh no, it's not there! Surprise! It can't find the socket that it failed to create. _______________________________________________ clamav-users mailing list clamav-users@lists.clamav.net http://lists.clamav.net/cgi-bin/mailman/listinfo/clamav-users Help us build a comprehensive ClamAV guide: https://github.com/vrtadmin/clamav-faq http://www.clamav.net/contact.html#ml
