Hello all,
I'm having a strange problem losing connectivity in OS/2 Warp 4, internal revision 14.062_W4,
with the 32-bit TCP/IP stack.
Every few minutes, the system loses the ability to ping anything. I can temporarily fix it with the
following commands:
route -n flush
route add default 10.0.0.1
However, the oddness here is that I've compared the routing tables when it works to the routing
tables when it doesn't work, and they're identical, so I'm at a loss to why flushing them would
make a difference.
The only recent networking-related change I've made has been updating Ultimedia Mail/2 Lite
(my favorite e-mail client) to the latest patch.
Any ideas?
Thanks in advance,
John
Hmm, lost that post by the looks of it, long post made short, what does "netstat -sarp" return from
the command prompt when you can nolonger ping the outside world ?
Cheers
Ian
Quote from: IBManners on 2011.11.22, 19:26:17
Hmm, lost that post by the looks of it, long post made short, what does "netstat -sarp" return from
the command prompt when you can nolonger ping the outside world ?
Cheers
Ian
Here it is:
--------------------------------------------------------------------------
AF_INET Address Family:
Total Number of sockets 11
SOCK TYPE FOREIGN LOCAL FOREIGN STATE
PORT PORT HOST
====== ===== ========== ========== ========== ========
1 DGRAM 0 netbios-dgm..138 0.0.0.0 UDP
2 DGRAM 0 netbios-ns..137 0.0.0.0 UDP
3 STREAM 0 netbios-ssn..139 0.0.0.0 LISTEN
4 DGRAM 0 emfis-data..140 0.0.0.0 UDP
5 STREAM 0 telnet..23 0.0.0.0 LISTEN
6 STREAM 0 ftp..21 0.0.0.0 LISTEN
7 DGRAM 0 sunrpc..111 0.0.0.0 UDP
8 STREAM 0 sunrpc..111 0.0.0.0 LISTEN
9 STREAM 0 smtp..25 0.0.0.0 LISTEN
10 STREAM 0 0 0.0.0.0 CLOSED
2079 DGRAM 0 0 0.0.0.0 UDP
--------------------------------------------------------------------------
AF_OS2 Address Family:
Total Number of sockets 0
destination router netmask metric flags intrf
default 10.0.0.1 0.0.0.0 0 UGSP lan0
8.8.8.8 10.0.0.1 255.255.255.255 0 UGHW3 lan0
10 10.0.0.193 255.0.0.0 0 UC lan0
10 10.0.0.1 255.0.0.0 1 UGSP lan0
10 0 255.0.0.0 1 UGSP lan0
10.0.0.13 10.0.0.1 255.255.255.255 0 UGHW lan0
10.0.0.15 10.0.0.1 255.255.255.255 0 UGHW lan0
10.255.255.255 10.0.0.1 255.255.255.255 0 UGHW lan0
127.0.0.1 127.0.0.1 255.255.255.255 0 UH lo
216.223.228.175 10.0.0.1 255.255.255.255 0 UGHW3 lan0
addr 127.0.0.1 Interface 9 mask 0xff000000 broadcast 127.0.0.1
Multicast addrs:
224.0.0.1
addr 10.0.0.193 Interface 0 mask 0xff000000 broadcast 10.255.255.255
Multicast addrs:
224.0.0.1
ARP table contents:
interface hardware address IP address minutes since
last use
lan0 0 :11:bb:e0:85:81 10.0.0.1 12
I don't know that it is wrong but it looks odd to me that there is a 10. entry with a 0. router address.
You stated you put a patch on Ultimedia Mail/2. It has been some time since I was able to use UM/2 so I am not sure, but is it possible the patch included some TCP/IP updates that would have actually down-leveled something (though it seems unlikely)? Along these lines though, do you leave UM/2 running or do you close it down when not directly in use? What I am wondering is if the patch to UM/2 is causing UM/2 to blow something out (not that I can think off the top of my head what that might be). If you generally leave it running, you might try shutting it down between uses to see if the routing still gets messed up.
Just because I don't like to make assumptions... when you say you can't ping anything, that also includes your router?
What does inetver report?
Andy Willis
Quote from: abwillis on 2011.11.23, 20:40:21
I don't know that it is wrong but it looks odd to me that there is a 10. entry with a 0. router address.
You stated you put a patch on Ultimedia Mail/2. It has been some time since I was able to use UM/2 so I am not sure, but is it possible the patch included some TCP/IP updates that would have actually down-leveled something (though it seems unlikely)? Along these lines though, do you leave UM/2 running or do you close it down when not directly in use? What I am wondering is if the patch to UM/2 is causing UM/2 to blow something out (not that I can think off the top of my head what that might be). If you generally leave it running, you might try shutting it down between uses to see if the routing still gets messed up.
Just because I don't like to make assumptions... when you say you can't ping anything, that also includes your router?
What does inetver report?
Andy Willis
To answer your questions, I don't leave UM/2 running. I have tested it,
and with a fresh reboot, the problem still occurs even if UM/2 is never launched at all.
When the problem occurs, I can't ping the router, the switch, or anything else.
Here is the inetver output:
Version numbers of TCP/IP protocol drivers:
SOCKETS.SYS: 6.2004
AFOS2.SYS: 6.2000
AFINET.SYS: 6.2013
Hi John,
I have to tread a bit carefully here because I think Ultimedia Mail/2 stops working above a certain version of
TCP/IP, or it might be due to one of the apps, so long now I've very rusty on OS/2's sendmail and mail/2.
Can you check if you have a file x:\mptn\etc\inetcfg.ini (x is your boot drive). If not, type "inetcfg -g all" which
should create that file, and post the contents here.
You could also try updating your stack to the latest 32bit version,
http://www.os2site.com/sw/upgrades/tcpip/tcpip32_stack.zip
unzip the contents into your X:\mptn\protocol directory BUT BACKUP your files in that directory first.
The zip also contains a inetcfg.ini file which can be put into your x:\mptn\etc directory and loaded with
"inetcfg -s all" you can put that command in x:\mptn\bin\setup.cmd if you like so tcp/ip options are set
on each boot.
I cant remember which version of TCP/IP inetcfg worked, to much to keep a track of these days :-)
All the above is really just suckit and see stuff but also as Andy pointed out, some of your netstat
data looks a bit odd, I was expecting something more along the lines of this (I use 192.168.1.x internally)
192.168.1.40 is me
192.168.1.1 is router 1
192.168.1.9 is router 2
And I haven't used the internet while I've been doing all this typing.
--------------------------------------------------------------------------
destination router netmask metric flags intrf
0 192.168.1.9 255.255.255.255 1 UGSP lan0
default 192.168.1.1 0.0.0.0 0 UGSP lan0
63.245.209.114 192.168.1.1 255.255.255.255 0 UGHW3 lan0
63.245.217.112 192.168.1.1 255.255.255.255 0 UGHW3 lan0
63.245.217.113 192.168.1.1 255.255.255.255 0 UGHW3 lan0
63.245.217.114 192.168.1.1 255.255.255.255 0 UGHW3 lan0
64.12.249.106 192.168.1.1 255.255.255.255 0 UGHW3 lan0
64.233.183.100 192.168.1.1 255.255.255.255 0 UGHW3 lan0
64.233.183.138 192.168.1.1 255.255.255.255 0 UGHW3 lan0
67.228.183.36 192.168.1.1 255.255.255.255 0 UGHW3 lan0
74.125.237.32 192.168.1.1 255.255.255.255 0 UGHW3 lan0
74.125.237.33 192.168.1.1 255.255.255.255 0 UGHW3 lan0
74.125.237.41 192.168.1.1 255.255.255.255 0 UGHW3 lan0
74.125.237.43 192.168.1.1 255.255.255.255 0 UGHW3 lan0
74.125.237.57 192.168.1.1 255.255.255.255 0 UGHW3 lan0
74.125.237.58 192.168.1.1 255.255.255.255 0 UGHW3 lan0
74.125.237.59 192.168.1.1 255.255.255.255 0 UGHW3 lan0
74.125.237.60 192.168.1.1 255.255.255.255 0 UGHW3 lan0
83.227.149.79 192.168.1.1 255.255.255.255 0 UGHW3 lan0
91.197.11.200 192.168.1.1 255.255.255.255 0 UGHW3 lan0
98.136.49.81 192.168.1.1 255.255.255.255 0 UGHW3 lan0
127.0.0.1 127.0.0.1 255.255.255.255 0 UH lo
178.255.83.1 192.168.1.1 255.255.255.255 0 UGHW3 lan0
192.168.1 192.168.1.40 255.255.255.0 0 UC lan0
199.7.48.72 192.168.1.1 255.255.255.255 0 UGHW3 lan0
202.177.203.124 192.168.1.1 255.255.255.255 0 UGHW3 lan0
203.26.51.73 192.168.1.1 255.255.255.255 0 UGHW3 lan0
203.26.51.139 192.168.1.1 255.255.255.255 0 UGHW3 lan0
203.26.51.172 192.168.1.1 255.255.255.255 0 UGHW3 lan0
203.123.73.8 192.168.1.1 255.255.255.255 0 UGHW3 lan0
203.123.73.9 192.168.1.1 255.255.255.255 0 UGHW3 lan0
203.134.152.45 192.168.1.1 255.255.255.255 0 UGHW3 lan0
203.166.110.169 192.168.1.1 255.255.255.255 0 UGHW3 lan0
204.152.184.196 192.168.1.1 255.255.255.255 0 UGHW3 lan0
addr 127.0.0.1 Interface 9 mask 0xff000000 broadcast 127.0.0.1
Multicast addrs:
224.0.0.1
addr 192.168.1.40 Interface 0 mask 0xffffff00 broadcast 192.168.1.255
Multicast addrs:
224.0.0.1
ARP table contents:
interface hardware address IP address minutes since
last use
lan0 0 :60:64:39:8 :e 192.168.1.1 6
lan0 0 :7 :e9:10:4 :3f 192.168.1.9 13
A couple of other things, do you need to run Telnet ?
I can also remember someone in the distant past had a problem with netbios with a WinXP box sucking
up all the sockets with chatter, might have been a dodgy network card actually but its something to check
if you have a WinXP box on the same network, ie, turn everything off except your OS/2 box and see how
far you get. Though I think your going to tell me there is nothing else on the network :)
Cheers
Ian
Hi John, if Ian's solution doesn't lead you to resolving this, then try this path:
(1) You have a Windows XP or later machine on your network
(2) Your Windows machine is flooding the OS/2 machine with nonsense.
(3) Portions of the TCP/IP stack (and perhaps NetBIOS) stop working
Solutions:
(1) remove TCP/IP over NetBIOS if you don't need it, install the OS/2 WSeB firewall (google it), block the "NETBEUI" ports in the firewall)
and/or
(2) update ALL networking components, including NetBIOS, LanMan, MPTS, NIC drivers and TCP/IP and cross your fingers.
and/or
(3) Tune the XP machines to stop acting idiotic when they find an OS/2 machine (heh... good luck with that)