Topic: TFTP server help

[JoshuaR]

Hi,
I was wondering if I could get some assistance with setting up a tftp server on SME 7. I've tried installing both the tftp and atftp contribs (not at the same time ), but I've been unable to get it working. (backing up cisco config to the tftp dir)
If I run "netstat -l -u | grep tftp" I get nothing returned.
If I run " ps -aux" I get "root      2663  0.0  0.0    20   20 ?        Ss   Nov29   0:09 runsv tftpd"
I also changed the permissions on the tftpboot dir to 777
In /etc/xinetd.d I have two files, "tftp" and "tftp.rpmsave"
The content of both is :
----------------
service tftp
{
   socket_type      = dgram
   protocol      = udp
   wait         = yes
   user         = root
   server         = /usr/sbin/in.tftpd
   server_args      = -s /tftpboot
   disable         =no
   per_source      = 11
   cps         = 100 2
   flags         = IPv4
}
-----------------

I'm not sure what I can do to troubleshoot it, does anyone have any ideas?

Much appreciated,
Josh

[Jáder]

Did you follow this Howto http://wiki.contribs.org/Tftp_server ?

I have a TFTP server running just following it.
Right now I´m without external access to my server ... so I cannot copy info to you... but I can post any info you need... if we start from begining following same install instructions!

[JoshuaR]

I believe I followed them exactly, but I'll uninstall the contrib and start again...
I'm not using it for PXE, so I tried it without the Thin Client contrib at first, then installed it as well just in case...
1. I ran "yum --enablerepo=smecontribs --enablerepo=base install smeserver-tftp-server"
2. I ran signal-event post-upgrade and signal-event reboot
3. I ran config setprop tftpd status enabled to make sure the tftp service was enabled
4. I ran /etc/init.d/supervise/tftpd restart for good measure, and it says restarting tftp [ok]
5. I changed the following properties:
config setprop tftpd access public
config setprop tftpd privelege rw

But I cannot get tftp read or write access

--Josh

[Jáder]

What your answer to:

Code: [Select]

service tftpd status
I got:

Code: [Select]

[root@lobo ~]# service tftpd status
run: /service/tftpd: (pid 10248) 125693s; run: log: (pid 2881) 126497s
[root@lobo ~]#

maybe you just need:

Code: [Select]

service tftpd start
I'm not sure you can see it at netstat... try:

Code: [Select]

ps gaux |grep tftto see something like:

Code: [Select]

[root@lobo ~]# ps gaux |grep tft
root      2860  0.0  0.0    20   20 ?        Ss   Nov29   0:00 runsv tftpd
[root@lobo ~]#

Hope this helps... if not... I'm here in Brazil ... so my time is GMT-0200 (summer time... used to be GMT-0300)

Jáder

[JoshuaR]

When I ran "service tftp status" I got
down: /service/tftpd: 1s, normally up, want up; down: log: 50023s, normally up
so I ran "service tftpd start"
Then ran "service tftpd status" and got
run: /service/tftpd: (pid 13158) 1s; down: log: 1s, normally up, want up
I appreciate the help, I'm in Australia, GMT +10

[Jáder]

Please verify (and if necessary post here) your logs... there are a reason why your TFTP is not up!
Please, after you/we solve your problem:
1) change subject line to [Solved]
2) explain here what you did... so next one can have a clue about this.

Thanks

Jáder

[JoshuaR]

from /etc/xinetd.d/tftp I changed
server_args = -s /tftpboot
to
server_args = -c -s /tftpboot

The services starts

ok, looking through the logs says:
2009-12-02 18:56:50.058478500 udpsvd: fatal: unable to look up local hostname: network unreachable
2009-12-02 19:02:37.133861500 udpsvd: info: listening on 0.0.0.0:69, uid 0, gid 0, starting.
2009-12-02 19:15:53.559294500 udpsvd: info: pid 10797 from 192.168.1.2
2009-12-02 19:15:53.559748500 udpsvd: info: start 10797 :0.0.0.0 :cisco.smedomain.dyndns.org:192.168.1.2:56750
2009-12-02 19:15:53.565470500 udpsvd: info: end 10797

[JoshuaR]

The service is now running, but along with the other errors I'm now getting
2009-12-03 22:56:34.627383500 udpsvd: drop: unable to read from socket: temporary failure
2009-12-03 22:56:34.627391500 udpsvd: info: end 16327
and I'm still unable to use tftp

any ideas?

Thanks again,
Josh

[Jáder]

So far I know you shouldn´t have to change a option (add -c) to get it working.
I think you should step back and fix it using logs.

Jáder

[CharlieBrady]

When I ran "service tftp status" I got
down: /service/tftpd: 1s, normally up, want up; down: log: 50023s, normally up
so I ran "service tftpd start"
Then ran "service tftpd status" and got
run: /service/tftpd: (pid 13158) 1s; down: log: 1s, normally up, want up

To diagnose such problems, do:

cd /service/tftpd
sv d .
./run
# read error messages here
^C
sv u .

[CharlieBrady]

from /etc/xinetd.d/tftp I changed
server_args = -s /tftpboot
to
server_args = -c -s /tftpboot

The services starts

ok, looking through the logs says:
2009-12-02 18:56:50.058478500 udpsvd: fatal: unable to look up local hostname: network unreachable
2009-12-02 19:02:37.133861500 udpsvd: info: listening on 0.0.0.0:69, uid 0, gid 0, starting.
2009-12-02 19:15:53.559294500 udpsvd: info: pid 10797 from 192.168.1.2
2009-12-02 19:15:53.559748500 udpsvd: info: start 10797 :0.0.0.0 :cisco.smedomain.dyndns.org:192.168.1.2:56750
2009-12-02 19:15:53.565470500 udpsvd: info: end 10797

You are running a supervised instance of udpsvd as well as trying to start tftpd via xinetd. You need to do one or the other, not both.

I suggest that you remove xinetd and the xinetd configuration.

[JoshuaR]

When I run those commands I get:
udpsvd: info: listening on 0.0.0.0:69, uid 99, gid 99, starting.
after I hit ^C
[1]+  Stopped                 ./run

-------
You are running a supervised instance of udpsvd as well as trying to start tftpd via xinetd. You need to do one or the other, not both.

I suggest that you remove xinetd and the xinetd configuration.
-------
I don't know how to change that sorry?