Linux -> Windows browse slow

[mpjbrennan]

Hi,

I have a network of 5 computers, three running Windows XP and two running Fedora (FC3 and FC4 respectively). I have Samba installed and all seems to work fine, but....

....when I browse shares on the Linux machines from any of the winboxes they come up almost instantaneously, whereas when I browse the winboxes from either of the Linux machines it takes 10 - 15 seconds before I get any response. The same thing applies if I browse either Linux machine from the other. The machine specs are pretty much equivalent in terms of speed, memory &c. Is this a Samba problem, or is Nautilus just much slower than Windows Explorer?

Patrick

[Twey]

is Nautilus just much slower than Windows Explorer?Try it in Konqueror and see.
Not that I'm suggesting Konqueror is faster or anything. Try smclient too.

[mpjbrennan]

Thanks for the suggestion, but that wouldn't answer my query redarding the comparative speeds of nautilus and windows explorer. I have the feeling that nautilus is just simply slow.
Patrick

[MoonMoon]

whereas when I browse the winboxes from either of the Linux machines it takes 10 - 15 seconds before I get any response.

the nautilus network browse take 2-3 seconds normally in a small network as mine.

try:
apply the firewall to add(insert) rule in /etc/sysconfig/iptables
-A RH-Firewall-1-INPUT -p udp -m udp --sport 137 -j ACCEPT
as following
-A RH-Firewall-1-INPUT -p udp --dport 5353 -d 224.0.0.251 -j ACCEPT
-A RH-Firewall-1-INPUT -p udp -m udp --dport 631 -j ACCEPT
-A RH-Firewall-1-INPUT -p udp -m udp --sport 137 -j ACCEPT
-A RH-Firewall-1-INPUT -m state --state ESTABLISHED,RELATED -j ACCEPT

restart the firewall
# service iptables restart

[mpjbrennan]

Thanks - but I don't have the RH firewall running.

Patrick

[MoonMoon]

Is there any abnormal message in the system log /var/log/messages ??

[mpjbrennan]

There are a lot of messages of the following type:

Dec 3 21:06:31 localhost smbd[6009]: [2005/12/03 21:06:31, 0] lib/util_sock.c:get_peer_addr(1000)
Dec 3 21:06:31 localhost smbd[6009]: getpeername failed. Error was Transport endpoint is not connected
Dec 3 21:06:31 localhost smbd[6009]: [2005/12/03 21:06:31, 0] lib/util_sock.c:write_socket_data(430)
Dec 3 21:06:31 localhost smbd[6009]: write_socket_data: write failure. Error = Connection reset by peer
Dec 3 21:06:31 localhost smbd[6009]: [2005/12/03 21:06:31, 0] lib/util_sock.c:write_socket(455)
Dec 3 21:06:31 localhost smbd[6009]: write_socket: Error writing 4 bytes to socket 22: ERRNO = Connection reset by peer
Dec 3 21:06:31 localhost smbd[6009]: [2005/12/03 21:06:31, 0] lib/util_sock.c:send_smb(647)
Dec 3 21:06:31 localhost smbd[6009]: Error writing 4 bytes to client. -1. (Connection reset by peer)
Dec 3 21:06:31 localhost smbd[6012]: [2005/12/03 21:06:31, 0] lib/util_sock.c:read_socket_data(384)
Dec 3 21:06:31 localhost smbd[6012]: read_socket_data: recv failure for 4. Error = Connection reset by peer

patrick

[MoonMoon]

1. setup the host name for the Linux boxes instead using localhost

&

try:

Force samba to use tcp port 139 instead of 139 & 445 at the same time.

edit /etc/samba/smb.conf to add below [global] in both samba box
smb ports = 139

restart samba service
# service smb restart

[mpjbrennan]

Hi MoonMoon

Thanks for sticking with this. I did what you suggested (the second part anyway, as the hostname wasn't localhost. I don't know why this comes up in the messages log). Unfortunately it didn't make any difference.

Patrick