HylaFAX The world's most advanced open source fax server

[Date Prev][Date Next][Thread Prev][Thread Next] [Date Index] [Thread Index]

[hylafax-users] root jobs stuck at faxend (Waiting for modem to come free)



I’m hoping someone can help diagnose a problem I’m having with hylafax-4.4.4-1rh7 on a RH7.3 system. I’ve spent a couple of days googling for like issues and have found some very useful info but have not come to a conclusion to solve my problem. Hylafax has been working great every night but on occasion random ports will get stuck at “Waiting for modem to come free” with a lock file on the port in /var/lock/LCK..ttySR??. When I run a ps –ef|grep faxsend I see the below in which the root sent job hangs indefinitely until the process is killed. Once the faxsend process is killed, the port frees up and more jobs are sent through it.

 

root     28961  3557  0 06:18 ?        00:00:00 faxsend -m ttySR13 sendq/q2939

root     26456 15398  0 Nov28 ?        00:00:05 faxsend -m ttySR4 sendq/q22051

root     17509 15398  0 Nov28 ?        00:00:05 faxsend -m ttySR2 sendq/q22723

uucp       963  3557  0 06:25 ?        00:00:00 faxsend -m ttySR8 sendq/q3043 se

uucp       974  3557  0 06:25 ?        00:00:00 faxsend -m ttySR5 sendq/q3278 se

uucp      1062  3557  0 06:26 ?        00:00:00 faxsend -m ttySR10 sendq/q1629

 

 

We run faxes at night with over 2000 jobs and a very slight failure rate. The problem is if some of the ports hang throughout the night, the faxes can take days to complete as apposed to overnight when nothing hangs. I only see jobs with a root ownership hang and I’m not sure if the user root is sending or if the fax is in a state where root owns the process. Nothing on the system is configured to send for root. I’ve checked my config files, faxgettys are set in inttab for the ports and am running on class2 settings. Everything looks/works correct until the root job hangs the port. The hung ports are random ports and I’ve seen up to four different ports hang one night but work perfect other nights. I can put a script in place to look out for hung jobs and kill -9 them but would like to see if I can get the issue corrected. Any help is appreciated.

 

 




Project hosted by iFAX Solutions