[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
tftp transfer problems from 3.6 -> 3.6
- To: misc_(_at_)_openbsd_(_dot_)_org
- Subject: tftp transfer problems from 3.6 -> 3.6
- From: eric <eric-list-openbsd-misc_(_at_)_catastrophe_(_dot_)_net>
- Date: Wed, 23 Mar 2005 10:19:13 -0600
- Mail-followup-to: misc_(_at_)_openbsd_(_dot_)_org
- Organization: Catastrophe.Net <http://www.catastrophe.net/>
I have two 3.6 machines which are following the 3.6 patch branch. When
trying to transfer a file using tftp from machine "foo" to machine "bar" I
get:
"
someone wants tftp
23193 execv /usr/libexec/tftpd
reaping asked for
23193 reaped, status 0
restored tftp, fd 4
"
Making sure everything is setup right first, I double-checked the file
permissions...everything looks good (and it works between our switch gear
and Netscreen firewalls).
bar# egrep -v '^#' /etc/inetd.conf
tftp dgram udp wait root /usr/libexec/tftpd tftpd -s /var/tftp
bar# ls -ld /var/tftp
drwxrwxrwt 2 root wheel 512 Feb 2 01:42 /var/tftp
bar# ls -l /var/tftp/pf_log.txt
-rw-rw-rw- 1 root wheel 0 Mar 23 10:07 /var/tftp/pf_log.txt
On the client side, I've turned on verbose and trace in the tftp client and
see this..
tftp> put pf_log.txt
putting pf_log.txt to foo:pf_log.txt [netascii]
sent WRQ <file=pf_log.txt, mode=netascii>
sent WRQ <file=pf_log.txt, mode=netascii>
sent WRQ <file=pf_log.txt, mode=netascii>
sent WRQ <file=pf_log.txt, mode=netascii>
sent WRQ <file=pf_log.txt, mode=netascii>
Transfer timed out.
On the server we see this...
bar# inetd -d
someone wants tftp
23193 execv /usr/libexec/tftpd
reaping asked for
23193 reaped, status 0
restored tftp, fd 4
someone wants tftp
12366 execv /usr/libexec/tftpd
reaping asked for
12366 reaped, status 0
restored tftp, fd 4
someone wants tftp
9904 execv /usr/libexec/tftpd
reaping asked for
9904 reaped, status 0
restored tftp, fd 4
someone wants tftp
22573 execv /usr/libexec/tftpd
reaping asked for
22573 reaped, status 0
restored tftp, fd 4
someone wants tftp
4571 execv /usr/libexec/tftpd
reaping asked for
4571 reaped, status 0
restored tftp, fd 4
The client has the following in it's pf.conf,
"pass out quick inet proto udp keep state"
The server has
"pass in quick inet proto udp from foo to any port 69 keep state"
I know there's issues with tftp in general as it's not reliable, but this is
on the same network segment and just going across the same blade on a 6509
switch.
Any help is appreciated.
Thanks.
Visit your host, monkey.org