Pure-FTPd starts, but can't attach to a port?

[dfraser]

Hi,

The trouble we are having with Pure-FTPd is that it will start up and the process is shown in 'ps', but it does not seem to attach to port 21. Thus no one can use FTP. The weird thing is that I configure things to show debug info in the log and the log files are never written to. The process PID file in /var/run never gets created either. It is as if pure-ftp hangs at some point in its startup. There is nothing wrong with the firewall.

Everything was working a day or two ago, and I can not track down any change in the system that would explain why the server is failing like this. There was a unexplained reboot yesterday morning at 4:10 but I don't think the server was hacked or anything - everything else is working fine. Not sure how to track the reason for the reboot - everything I looked at shows no clue

I am thinking I ought to recompile pure-ftp etc, but the ftpup --force script does not do much, it just restarts the server essentially. Any ideas? switch over to proftpd, then switch back? delete pure-ftpd completely and reinstall? How can I force the complete reinstallation?

thanks
doug

 

[cPanelTristan]

Please provide the following results from these commands:

ps aux | grep pure-ftpd
lsof -i :21
netstat -an | grep :21

 

[dfraser]

Please provide the following results from these commands:

ps aux | grep pure-ftpd
lsof -i :21
netstat -an | grep :21

ok, i started pure-ftp before doing this to show that it does not attach to the port

root 8896 0.0 0.0 93332 520 ? Ss 14:19 0:00 /usr/sbin/pure-ftpd -O clf:/var/log/xferlog --daemonize -A -c50 -B -C8 -d -D -E -fftp -H -I60 -lunix -lpuredb:/etc/pureftpd.pdb -lextauth:/var/run/ftpd.sock -L10000:8 -m4 -p35000:45000 -s -T100000 -U133:022 -u100 -i -Oxferlog:/usr/local/apache/domlogs/ftpxferlog -R -o -k99 -G -Z -0 -Y1 -JHIGH:MEDIUM:+TLSv1:!SSLv2:+SSLv3
root 8908 0.0 0.0 61188 752 pts/0 S+ 14:19 0:00 grep pure-ftpd

****

[email protected] [/etc/init.d]# /usr/sbin/lsof -i :21
[email protected] [/etc/init.d]# netstat -an | grep :21
tcp 0 0 66.135.40.118:80 140.214.40.147:21361 TIME_WAIT
[email protected] [/etc/init.d]#

 

[dfraser]

repost... didn't expect to be moderated on a thread I started...

 

[dfraser]

problem solved! Due to the unexpected reboot, named was running and it ought not to have - I did not see this and thought chkconfig had been used to set all that up properly. Perhaps something got overwritten...

I saw some post somewhere that problems with pure-ftp might have something to do with a non-working BIND, and as I had one, that might have been it (the post did not go into detail)

I killed named, and now the FTP server starts up fine