[hylafax-users] Possible race condition

["Dominique dHotman" <dominique.dhotman@xxxxxxxxxx>]

Hello List,

 

I’m busy working on a new HylaFAX setup that will replace an existing Castelle system.

 

The configuration I have is basically:

 

• Oracle Enterprise Linux, which is basically just a Red Hat port (uname –r = 2.6.18-92.el5)
• HylaFAX 4.4.4-1 installed using the RPM from hylafax.org (rpm -qa|grep hylafax = hylafax-4.4.4-1rhel5)
• Eicon Diva Server Board 30 Channel PRI Modem
• 30 TTY Modem’s using the Eicon TTY driver
• All config.ttyXX are setup for CLASS 1 faxing

 

 

In our environment we will be faxing to locations that have smart access type numbers (i.e. 1 number that is load balanced across multiple channels) and because of this I’ve change the following in the main server config (/var/spool/hylafax/etc/config):

 

MaxConcurrentCalls: 30

MaxBatchJobs: 1

 

In order to test this I’ve created a small script that basically sends 30 concurrent faxes.  The script looks like this:

 

#!/bin/bash

           

num=0

           

while [ ${num} -le 30 ]; do

   sendfax -R -t 1 -s a4 -d 086XXXXXXXXX /tmp/fax

   echo "num=${num}"

   num=$[${num}+1]

done

 

When I run this script the 1^st fax transmission is started but the subsequent 29 are marked as blocked.  The following is logged:

 

Oct 24 07:20:16 hylafax01 FaxQueuer[4357]: NOTIFY: bin/notify "sendq/q429" "blocked" ""

 

At first I thought it was something wrong with my configuration but then I changed the script to at put a short sleep statement in the loop.  With the sleep statement all the fax transmissions are kicked off.  I’m not sure why this is and if there is anything in my configuration that I can change to sort this out?

 

The reason that I’m worried about this is that we are going to be operating with fairly high volumes and blocked faxes where lines are available will cause a major major head ache L

 

Any help would be appreciated.

 

Thank you

Dominique