Hoping someone can shed some light on this for me - I am testing a fresh install of FreePBX with the Schmoozecom iso, and can not seem to get tftp working (or I am really doing something wrong…). I am running it in a VM on VMWare Fusion on a mac host, I can get to the web configs, I set up 2 extensions, and loaded some Cisco firmware in the tftpboot directory, but when I set the phone to look at that server it will not pull anything from tftp. I am not sure how to test if tftp is running, can someone please assist? I am trying to upgrade from Trixbox with asterisk 1.4 to use the new Cisco firmware and phones which use SIP over TCP.
If the server can’t ping the phones it won’t work.
Certainly that’s a reason to upgrade but since trixbox is abandoned, unsupported and one accidental yum away from breaking would you rather not be on a platform going forward and not in a death spiral?
Scott, I want to upgrade primarily to support SIP over TCP because we want to use some of the CIsco 6900 and 8900 phones. That’s really my big reason to upgrade. Correct, I could tftp the box itself ok. I will try the same vlan tonight and let you know.
Tonight I am going to try again and put the phones and server in the same vlan to simplify things. As far as I can see, there’s no reason not to have them all together. I will post results tonight. fingers crossed…thanks for all your help and patience with me!
trixbox did not use the OS tftp, they used atftpd, the log file was /var/log/atftpd.log
You can change the log file location of tftp in the xinet file just like you could in trixbox.
What’s wrong with only using trixbox? It’s not that something is wrong it’s that it was not a distro, it was a fork. And a lousy one as they screwed everything up including the users. It’s not a wide spectrum of experience. “all” and I am sure I am forgetting some, of the other distro’s (PBX in a Flash, Elasix, Asterisk Now) use the native tftp package in Linux and not atftpd.
I am sure if you ran tftp on the box to itself it would work. Here is an example session, I just grabbed a file in my home directory. The first two commands show you where I am and that the directory was empty to start.
[root@pioneer scott]# pwd
/home/scott
[root@pioneer scott]# ls
[root@pioneer scott]# tftp 127.0.0.1
tftp> get aastra.cfg
tftp> quit
[root@pioneer scott]# ls
aastra.cfg
[root@pioneer scott]#
[root@pioneer scott]#
If you are using vlan’s then your eth0.n address space must be reflected in your dhcp and tftpd servers configurations, You will need to correctly identify the vlan’s they are listening on.
I tried all that and can’t get it working for some reason or another…yes, it is true I have never used another distro in production other than trixbox, perhaps my “all other distros” comment was not correct. Is it a bad thing that I’ve “only used Trixbox”? I just put my old tb server back in production and every phone came up with no problem. I guess I will have to compare configs from this to the new one. Nothing I seem to do makes any difference, phones all timeout with tftp, and nothing registers, regardless of settings.
I appreciate all the help so far but I seem to still be missing something really simple. I copied the tftpboot directory file for file but that didn’t help either.
I can put all the phones and server on one vlan if it would help. Right now the phones are 10.10.50.0/24 and the server is at 10.10.60.6. I would point to that as part of the issue, but again tb works fine with that setup.
Also, I assume you are not going to run the VMware environment in production?
One little note you can download the FreePBX bulk DID module from mirror.freepbx.org/modules and install it on the trixbox using the module admin. Then you can export to the new install.
Also you can use the scp command line utility to move all the VM’s and greetings once you have the extensions and users created.
I am thinking it may be a NAT Issue with the VM and the way it handles networking - For now I have our DHCP server sending the TFTP address so I loaded the 9.3 SIP firmware to the production server, as well as the SEPMAC.cnf.xml file for the phone. It pulled down the firmware OK, and loaded the xml file ok as well. However it sits and says Registering, but nothing ever registers. I have NAT=Never in the extension, and transport=TCP in the extension as well. Is there something else I need to set somewhere? This is my first foray into the world of TCP SIP phones, and I’d like to get it working since I really need to move from Trixbox. I like the Freepbx/schmoozecom distro so far, and it does support SIP over TCP if I’m not mistaken, so Id like to try and get it working (especially as all newer Cisco SIP loads use TCP).
Can anyone shed light on the registration issue? I can post the xml file later if that will help, but I am using the same XML file that worked on the older cisco phones (non TCP).
Scott - no, I will not be using the VM in production, it will go on a dedicated box. Just wanted to test the new feature set with the new phone loads.
By the way, I can SFTP to the server and dump files in the TFTPBOOT foler but there is no atftpd.log file in the logs folder.
Googling CentOS TFTP did not reveal any useful information other than looking at the xinetd.d/tftp file, which is identical to the production server (other than the new IP).
add the -vvv to server_args in /etc/xinetd.d/tftp to get quite verbose logging. (in /var/log/messages ) add the -s too as that will prepend /tftpboot to any tftpd requested files.
Still nothing - I have done a completely fresh install on a new server (dedicated) and again have no tftp service. I can use a client (Cyberduck), to copy files to the /tftpboot directory, but there is no /var/log/tftpd.log file to speak of, I am stumped.
Once again with feeling add -vv to your server_args, this will increase the verbosity by two levels above “none” and the logs will appear in /var/log/messages not, as you have done, by not adding -vv and looking in the wrong file, also tail will, by default, print the last 10 lines of a file, you could better use cat to read the whole file.
Added -vv to this newest fresh install, look in /var/log/messages for anything tftp related - nothing there. Rebooted 2 phones that should now ask for their files, they do, but dont get them, and then fail to register.
Why is there no /var/log/tftpd.log file? That is the file to look in every other distro I’ve used. Is this just something in the freepbx distro that they append all the tftp messages to the /log/messages file?
Forgive my ignorance, and please have a little more patience with me, this is a fresh install, not the one I initially posted about.
FYI: This log was taken at Sept, 19, 12:56AM.
[root@freepbx ~]# cat /etc/xinetd.d/tftp
# default: off
# description: The tftp server serves files using the trivial file transfer \
# protocol. The tftp protocol is often used to boot diskless \
# workstations, download configuration files to network-aware printers, \
# and to start the installation process for some operating systems.
service tftp
{
socket_type = dgram
protocol = udp
wait = yes
user = root
server = /usr/sbin/in.tftpd
server_args = -vv -s /tftpboot
disable = no
per_source = 11
cps = 100 2
flags = IPv4
}
The tftpd.log hasn’t been there in a while and was probably only in there for trixbox. Please stop saying “every other distro” we all know you’ve only used Trixbox and FreePBX and the FreePBX distro has never had tftp installed.