![]() |
* francesco messineo <francesco.messineo@xxxxxxxxx> [070426 04:41]: > Hello, > > I observed this strange behaviour on two different installation of > hylafax (both 4.3.3, the first on a CentOS distribution, the second on > a Ubuntu 6.10). > > Sometimes I observe a whole lot of wedged modems, looking at the > session logs I see "strange" things, like different processess talking > to the same modem. Infacts, I see different processes talking to the > same modem: > > 30859 ? S 0:00 faxsend -m ttyM14 sendq/q45662 > 31071 ? S 0:00 faxsend -m ttyM14 sendq/q45459 > 31288 ? S 0:00 faxsend -m ttyM14 sendq/q45676 > 31299 ? S 0:00 faxsend -m ttyM14 sendq/q45393
What are the parent PID of these processes. Are you stopping/restarting faxq regularly? Can you show the log lines from faxq's syslog that show it starting all these faxsend processes?
I don't stop and restart faxq regularly, should I? here are loglines:
Apr 26 10:28:29 fax FaxSend[31299]: Error parsing "+FTM" query response: "OK" Apr 26 10:28:29 fax FaxSend[31299]: /dev/ttyM14: Can not initialize modem.
Apr 26 10:27:35 fax FaxSend[31071]: Error parsing "+FRM" query response: "OK" Apr 26 10:27:35 fax FaxSend[31071]: /dev/ttyM14: Can not initialize modem.
Apr 26 10:28:42 fax FaxSend[30859]: SEND FAX: JOB 45662 DEST +39-081-8635585 COMMID 000066690 DEVICE '/dev/ttyM14' FROM 'assistenza <assistenza@xxxxxxxxxx>' USER rsync
Apr 26 10:29:11 fax FaxSend[31299]: MODEM V2.400B-V90_2M_DLS MT5600ZDX/ Apr 26 10:29:12 fax FaxSend[31299]: SEND FAX: JOB 45393 DEST +39-0755-22801 COMMID 000066698 DEVICE '/dev/ttyM14' FROM 'assistenza <assistenza@xxxxxxxxxx>' USER rsync
Apr 26 10:29:12 fax FaxSend[31288]: MODEM V2.400B-V90_2M_DLS MT5600ZDX/ Apr 26 10:29:12 fax FaxSend[31288]: SEND FAX: JOB 45676 DEST +39-081-8783933 COMMID 000066699 DEVICE '/dev/ttyM14' FROM 'assistenza <assistenza@xxxxxxxxxx>' USER rsync
Running faxgetty is the recommended production mode, even for send-only environments. Tune the PollLockWait appropriately in the per-device config files.
I see the default value is 30 seconds. I don't really understand how should I tune it, maybe to allow a closer poll? Any hint here? As I wrote in another post, I see that this kind of errors always begin with a conversion problem of one or more pdf fax, but I don't get how this is related to what I observe.
____________________ HylaFAX(tm) Users Mailing List _______________________ To subscribe/unsubscribe, click http://lists.hylafax.org/cgi-bin/lsg2.cgi On UNIX: mail -s unsubscribe hylafax-users-request@xxxxxxxxxxx < /dev/null *To learn about commercial HylaFAX(tm) support, mail sales@xxxxxxxxx*