Tom,
Just built a brand-spanking-new Proxmox box (on a quad core duo) and created a PIAF VM. Did the install of PIAF 2.0.6.2 and it went swimmingly. Installed PIAF3 without any problems.
BUT
When I went to install incredfax2, I'm seeing the exact same issues that I saw when I was installing it on the R1600.
You are having problems
While My prox box is busy at the moment with one of the other developers I did the following in Vmware workstation 8.0
1. Installed fresh copy of PIAF using the 20621 ISO
2. After the install I ran yum update to be sure that the php updates were installed. They were already with the fresh install of PIAF 20621
3. The system was Asterisk 1880/freepbx 2.9.X/32 bit
4. Install-incredpbx3 -d (I used the developer repo as I have been doing some updates on the english of the messages!)
5. system rebooted
6. install-incredfax2 -d (I used the developer repo as I have been doing some updates on the english of the messages!)
7. I also added the OCR portion to avantfax
8. modified iptables
9. I also downloaded the hylafax clients. I like yajhfc as it seems to work just fine with win7-64 ultimate.
10. system reboots.
11. No errors on boot up quite happy to send you a movie of bootup or a copy of dmesg
11. Copies of status messages are below
12, Vm had single processor/single core with 1024 ram and 20 gb HD space running under a AMD Athlon(tm) II X2 250 Processor
There are some php errors in the apache log from avantfax but it does not seem to affect the operation of anything. Faxes came and out just fine. I will try later on a prox machine when it becomes free however if it works on vmware it always works on prox 1.8. Sorry I keep coming back to this stuff does not run on all configurations of hardware. Unless you are willing to permanently send us a machine to experiment with there is little we can do. None of the developers are willing to go in remotely to experiment on someone else's machine. The liability exposure is just to high regardless of good intentions. I also spoke to the PIAF team and no one else has found this issue on any of the hardware types we have in the pool. This does not mean anything other than if we can duplicate it locally we really can't fix it. For example a few users complained of keyboard problems on several different types of machines. We could not generate the problem and no one was willing to send us a machine so there is nothing we can do.
So it is back in your court what is different about your install/method/etc?
Tom