[SOLVED] network unreachable inside the container
I managed a container with systemd-nspawn. The container boots, but the network is unreachable.
Below is my set up.
ON HOST
systemd-dhcpcd.service disable
systemd-networkd is enabled and started
network is started with two netctl profiles
Configuration files:
/etc/netctl/static-hortensia
Description='hortensia static ethernet connection'
Interface=enp7s0
Connection=ethernet
IP=static
Address=('192.168.1.87/24')
Gateway='192.168.1.254'
/etc/netctl/bridge-hortensia
Description="Bridge connection to container"
Interface=br0
Connection=bridge
BindsToInterfaces=()
IP=no
/etc/systemd/network/70-dahlia.netdev
[Match]
Host=host0
Virtualization=container
[NetDev]
Name=br0
Kind=bridge
/etc/systemd/network/80-dahlia.network
[Match]
Virtualization=container
[Network]
DHCP=no
DNS=192.168.1.254
[Address]
Address=192.168.1.94/24
[Route]
Gateway=192.168.1.254
/etc/resolv.conf
# Generated by resolvconf
domain lan
nameserver 192.168.1.254
BEFORE I start the container:
$ ip addr
2: enp7s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
link/ether 14:da:e9:b5:7a:88 brd ff:ff:ff:ff:ff:ff
inet 192.168.1.87/24 brd 192.168.1.255 scope global enp7s0
valid_lft forever preferred_lft forever
inet6 fe80::16da:e9ff:feb5:7a88/64 scope link
valid_lft forever preferred_lft forever
3: br0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN group default
link/ether 9e:eb:1a:c5:12:34 brd ff:ff:ff:ff:ff:ff
inet6 fe80::9ceb:1aff:fec5:1234/64 scope link
valid_lft forever preferred_lft forever
start the container
# systemd-nspawn --machine=dahlia --network-bridge=br0 -bD /dahlia
$ ip addr
2: enp7s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
link/ether 14:da:e9:b5:7a:88 brd ff:ff:ff:ff:ff:ff
inet 192.168.1.87/24 brd 192.168.1.255 scope global enp7s0
valid_lft forever preferred_lft forever
inet6 fe80::16da:e9ff:feb5:7a88/64 scope link
valid_lft forever preferred_lft forever
3: br0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN group default
link/ether 92:3c:ba:9e:24:07 brd ff:ff:ff:ff:ff:ff
inet6 fe80::9ceb:1aff:fec5:1234/64 scope link
valid_lft forever preferred_lft forever
4: vb-dahlia: <BROADCAST,MULTICAST> mtu 1500 qdisc noop master br0 state DOWN group default qlen 1000
ON CONTAINER
systemd-dhcpcd.service disable
systemd-networkd is enabled and started
NO netctl profiles
NO conf files in /etc/systemd/network/
gab@dahlia ➤➤ ~ % ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
inet 192.168.1.94/24 brd 192.168.1.255 scope global lo
valid_lft forever preferred_lft forever
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: host0: <NO-CARRIER,BROADCAST,ALLMULTI,AUTOMEDIA,NOTRAILERS,UP> mtu
1500 qdisc pfifo_fast state DOWN group default qlen 1000
link/ether 3a:4f:1f:c5:b5:d1 brd ff:ff:ff:ff:ff:ff
inet 192.168.1.94/24 brd 192.168.1.255 scope global host0
valid_lft forever preferred_lft forever
Not sure this output is correct. Is it OK to get a IP adress for lo ? Then, interface host0 is DOWN. I guess this is not normal and could be the cause of my issue.
# ip link set dev host0 up
produces no change, host0 is still down
gab@dahlia ➤➤ ~ % ip route
default via 192.168.1.254 dev host0
192.168.1.0/24 dev host0 proto kernel scope link src 192.168.1.94
gab@dahlia ➤➤ ~ % ping -c 3 8.8.8.8
PING 8.8.8.8 (8.8.8.8) 56(84) bytes of data.
From 192.168.1.94 icmp_seq=1 Destination Host Unreachable
Configuration files :
/etc/resolv.conf
# Generated by resolvconf
domain lan
nameserver 192.168.1.254
/etc/hosts
# /etc/hosts: static lookup table for host names
#<ip-address> <hostname.domain.org> <hostname>
127.0.0.1 localhost.localdomain localhost
::1 localhost.localdomain localhost
# End of file
Maybe some error here? localhost ? (host0 ?)
Some debug command outputs:
gab@dahlia ➤➤ ~ # SYSTEMD_LOG_LEVEL=debug /lib/systemd/systemd-networkd
timestamp of '/etc/systemd/network' changed
timestamp of '/run/systemd/network' changed
host0: link (with ifindex 2) added
lo: link (with ifindex 1) added
Sent message type=method_call sender=n/a destination=org.freedesktop.DBus object=/org/freedesktop/DBus interface=org.freedesktop.DBus member=Hello cookie=1 reply_cookie=0 error=n/a
Got message type=method_return sender=org.freedesktop.DBus destination=:1.6 object=n/a interface=n/a member=n/a cookie=1 reply_cookie=1 error=n/a
Got message type=signal sender=org.freedesktop.DBus destination=:1.6 object=/org/freedesktop/DBus interface=org.freedesktop.DBus member=NameAcquired cookie=2 reply_cookie=0 error=n/a
gab@dahlia ➤➤ ~ % ip route
default via 192.168.1.254 dev host0
192.168.1.0/24 dev host0 proto kernel scope link src 192.168.1.94
gab@dahlia ➤➤ ~ % cat /proc/net/dev
Inter-| Receive | Transmit
face |bytes packets errs drop fifo frame compressed multicast|bytes packets errs drop fifo colls carrier compressed
host0: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
lo: 840 9 0 0 0 0 0 0 840 9 0 0 0 0 0 0
Same command ON HOST
gabx@hortensia ➤➤ ~ % cat /proc/net/dev
Inter-| Receive | Transmit
face |bytes packets errs drop fifo frame compressed multicast|bytes packets errs drop fifo colls carrier compressed
br0: 0 0 0 0 0 0 0 0 648 8 0 0 0 0 0 0
vb-dahlia: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
lo: 1700 34 0 0 0 0 0 0 1700 34 0 0 0 0 0 0
enp7s0: 15403401 19789 0 0 0 0 0 0 3834189 16721 0 0 0 0 0 0
gab@dahlia ➤➤ ~ % ping -c3 192.168.1.254
PING 192.168.1.254 (192.168.1.254) 56(84) bytes of data.
64 bytes from 192.168.1.254: icmp_seq=1 ttl=64 time=0.036 ms
I can ping the gateway.
Any help is appreciated.
Last edited by gabx (2014-03-06 22:15:07)
After a few more test, I have a profile UP in the container, with an IP adress, but network is still unreachable.
The output of the following command puzzles me:
gab@dahlia ➤➤ /etc/netctl % cat /proc/net/dev
Inter-| Receive | Transmit
face |bytes packets errs drop fifo frame compressed multicast|bytes packets errs drop fifo colls carrier compressed
host0: 6004 28 0 0 0 0 0 0 774 11 0 0 0 0 0 0
lo: 336 3 0 0 0 0 0 0 336 3 0 0 0 0 0 0
It seems there is some traffic going through host0.
some debug outputs on the container side
gab@dahlia ➤➤ /etc/netctl % ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: host0: <BROADCAST,ALLMULTI,AUTOMEDIA,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
link/ether 8e:d4:16:e2:06:4a brd ff:ff:ff:ff:ff:ff
inet 192.168.1.91/24 brd 192.168.1.255 scope global host0
valid_lft forever preferred_lft forever
inet6 fe80::8cd4:16ff:fee2:64a/64 scope link
valid_lft forever preferred_lft forever
gab@dahlia ➤➤ /etc/netctl % ip route
default via 192.168.1.254 dev host0
192.168.1.0/24 dev host0 proto kernel scope link src 192.168.1.91
gab@dahlia ➤➤ /etc/netctl % cat /etc/resolv.conf
# Generated by resolvconf
nameserver 192.168.1.254
Maybe a stupid question, but in case of my bridge, what device is the gateway : the host machine (192.168.1.87) OR the real router (192.168.1.254) ? I could be wrong when trying to indicate the router as the gateway ?
EDIT
Trying to use the host as gateway does not change anything: network still unreachable
More debug outputs.
on the container side
gab@dahlia ➤➤ ~ % ping -c3 192.168.1.254
PING 192.168.1.254 (192.168.1.254) 56(84) bytes of data.
64 bytes from 192.168.1.254: icmp_seq=1 ttl=64 time=0.044 ms
64 bytes from 192.168.1.254: icmp_seq=2 ttl=64 time=0.035 ms
64 bytes from 192.168.1.254: icmp_seq=3 ttl=64 time=0.027 ms
--- 192.168.1.254 ping statistics ---
3 packets transmitted, 3 received, 0% packet loss, time 1998ms
rtt min/avg/max/mdev = 0.027/0.035/0.044/0.008 ms
gab@dahlia ➤➤ ~ % ping -c3 192.168.1.87
PING 192.168.1.87 (192.168.1.87) 56(84) bytes of data.
64 bytes from 192.168.1.87: icmp_seq=1 ttl=64 time=0.041 ms
64 bytes from 192.168.1.87: icmp_seq=2 ttl=64 time=0.036 ms
64 bytes from 192.168.1.87: icmp_seq=3 ttl=64 time=0.036 ms
--- 192.168.1.87 ping statistics ---
3 packets transmitted, 3 received, 0% packet loss, time 1999ms
rtt min/avg/max/mdev = 0.036/0.037/0.041/0.007 ms
I can ping gateway and host
on host side
gabx@hortensia ➤➤ systemd/network % ping -c3 192.168.1.94
PING 192.168.1.94 (192.168.1.94) 56(84) bytes of data.
From 192.168.1.87 icmp_seq=1 Destination Host Unreachable
From 192.168.1.87 icmp_seq=2 Destination Host Unreachable
From 192.168.1.87 icmp_seq=3 Destination Host Unreachable
--- 192.168.1.94 ping statistics ---
3 packets transmitted, 0 received, +3 errors, 100% packet loss, time 2008ms
pipe 3
I can not ping container
Last edited by gabx (2014-03-04 12:07:32)
Similar Messages
-
[SOLVED]Network unreachable in live CD 2013.07.01 with BCM57788
Eth0 not set up with live CD 2013.07.01, so I was unable to install Archlinux
Here are some informations on my configuration :
* Computer Dell XPS8300,
* I'm wired in DHCP to a Freebox as a router (IP given 192.168.0.15)
* result of
systemctl -- failed
UNIT LOAD ACTIVE SUB DESCRIPTION
[email protected] loaded failed failed dhcpcd on wlp2s0
LOAD = Reflects whether the unit definition was properly loaded.
ACTIVE = The high-level unit activation state, i.e. generalization of SUB.
SUB = The low-level unit activation state, values depend on unit type.
1 loaded units listed. Pass --all to see loaded but inactive units, too.
To show all installed unit files use 'systemctl list-unit-files'.
* result of
lspci -s 04:00.0 -vnn
04:00.0 Ethernet controller [0200]: Broadcom Corporation NetLink BCM57788 Gigabit Ethernet PCIe [14e4:1691] (rev 01)
Subsystem: Dell XPS 8300 [1028:04aa]
Flags: fast devsel, IRQ 19
Memory at fe300000 (64-bit, non-prefetchable) [size=64K]
Capabilities: [48] Power Management version 3
Capabilities: [60] Vendor Specific Information: Len=6c <?>
Capabilities: [50] MSI: Enable- Count=1/1 Maskable- 64bit+
Capabilities: [cc] Express Endpoint, MSI 00
Capabilities: [100] Advanced Error Reporting
Capabilities: [13c] Virtual Channel
Capabilities: [160] Device Serial Number 78-2b-cb-ff-fe-8e-89-01
Capabilities: [16c] Power Budgeting <?>
Kernel modules: tg3
* result of ifconfig :
lo: flags=73<UP,LOOPBACK,RUNNING> mtu 65536
inet 127.0.0.1 netmask 255.0.0.0
inet6 ::1 prefixlen 128 scopeid 0x10<host>
loop txqueuelen 0 (Local Loopback)
RX packets 0 bytes 0 (0.0 B)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 0 bytes 0 (0.0 B)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0
wlp2s0: flags=4099<UP,BROADCAST,MULTICAST> mtu 1500
ether 60:d8:19:61:e2:dd txqueuelen 1000 (Ethernet)
RX packets 0 bytes 0 (0.0 B)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 0 bytes 0 (0.0 B)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0
* result of ip link :
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: wlp2s0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq state DOWN mode DEFAULT qlen 1000
link/ether 60:d8:19:61:e2:dd brd ff:ff:ff:ff:ff:ff
Not knowing systemd, I cannot see if tg3 is loaded.
In addition, I'm writing from a previous Arch install witch is working fine, so all plugs and wires are OK.
Question : why ethernet is down and how correct the live CD ?
Thank you for your help.
Last edited by Christil (2013-07-04 12:48:36)Christil wrote:
Eth0 not set up with live CD 2013.07.01, so I was unable to install Archlinux
Here are some informations on my configuration :
* Computer Dell XPS8300,
* I'm wired in DHCP to a Freebox as a router (IP given 192.168.0.15)
* result of
systemctl -- failed
UNIT LOAD ACTIVE SUB DESCRIPTION
[email protected] loaded failed failed dhcpcd on wlp2s0
LOAD = Reflects whether the unit definition was properly loaded.
ACTIVE = The high-level unit activation state, i.e. generalization of SUB.
SUB = The low-level unit activation state, values depend on unit type.
1 loaded units listed. Pass --all to see loaded but inactive units, too.
To show all installed unit files use 'systemctl list-unit-files'.
* result of
lspci -s 04:00.0 -vnn
04:00.0 Ethernet controller [0200]: Broadcom Corporation NetLink BCM57788 Gigabit Ethernet PCIe [14e4:1691] (rev 01)
Subsystem: Dell XPS 8300 [1028:04aa]
Flags: fast devsel, IRQ 19
Memory at fe300000 (64-bit, non-prefetchable) [size=64K]
Capabilities: [48] Power Management version 3
Capabilities: [60] Vendor Specific Information: Len=6c <?>
Capabilities: [50] MSI: Enable- Count=1/1 Maskable- 64bit+
Capabilities: [cc] Express Endpoint, MSI 00
Capabilities: [100] Advanced Error Reporting
Capabilities: [13c] Virtual Channel
Capabilities: [160] Device Serial Number 78-2b-cb-ff-fe-8e-89-01
Capabilities: [16c] Power Budgeting <?>
Kernel modules: tg3
* result of ifconfig :
lo: flags=73<UP,LOOPBACK,RUNNING> mtu 65536
inet 127.0.0.1 netmask 255.0.0.0
inet6 ::1 prefixlen 128 scopeid 0x10<host>
loop txqueuelen 0 (Local Loopback)
RX packets 0 bytes 0 (0.0 B)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 0 bytes 0 (0.0 B)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0
wlp2s0: flags=4099<UP,BROADCAST,MULTICAST> mtu 1500
ether 60:d8:19:61:e2:dd txqueuelen 1000 (Ethernet)
RX packets 0 bytes 0 (0.0 B)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 0 bytes 0 (0.0 B)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0
* result of ip link :
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: wlp2s0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq state DOWN mode DEFAULT qlen 1000
link/ether 60:d8:19:61:e2:dd brd ff:ff:ff:ff:ff:ff
Not knowing systemd, I cannot see if tg3 is loaded.
In addition, I'm writing from a previous Arch install witch is working fine, so all plugs and wires are OK.
Question : why ethernet is down and how correct the live CD ?
Thank you for your help. -
[Solved] Decrypt files securely inside lxc container.
Hi, this is my first question, I've been using Arch for 5 years now, thanks for all your hard work!
I'm trying to setup an lxc container to securely view encrypted files(gpg keys encrypted with cryptsetup and a loopback device), I've done so already using cryptsetup, the only issue is that after I decrypt the device from inside the container, it is available from the host as /dev/mapper/name and I can mount it and view it's contents.
Is there any way to set cgroups to deny host and allow lxc container to access this device, maybe I can change /dev permissions so only root on host can access? I would prefer that not even root on host can access it but I know I'm asking too much, I can do this with qemu or other VMs but there is a lot of overhead.
If you happen to know of any way I can access files from inside lxc container but block access to host even if it is after mounting the device, I would like to hear you out.
I'm running an archlinux host and archlinux lxc container with grsecurity-pax enabled.
Feel free to ask any questions if it's not clear .
Thank you for your time, and kudos for your great work with this distro.
Last edited by freetoken (2014-09-18 07:57:46)Progress!
$ ls -la /dev/mapper/name
lrwxrwxrwx 1 root root 7 Sep 16 09:21 /dev/mapper/name -> ../dm-0
$ ls -la /dev/dm-0
brw-rw---- 1 root disk 254, 0 Sep 16 09:21 /dev/dm-0
$ rm /dev/mapper/name
$ rm /dev/dm-0
voila!
The device is still there but it's not visible and user would need to have root permissions to run mknod and restore dm-0.
Also very interesting read about cgroups for anyone interested: https://www.kernel.org/doc/Documentatio … groups.txt
I'm wondering if there is any damage from doing this, since the data is backed up I guess I don't care too much and will test it for a while.
If anyone has any ideas or knows some cool feature I'm missing let me know.
Will mark as solved in 24 hours after more testing and in case something unexpected happens. -
Resizing the swf inside a container
Hi Guys,
I am using a dashboard inside a vbox. I have integrated my
dashboard into my main application (i.e. I have two different swfs,
one of my main application and other of dashboard). Now when I
login into my application by reducing the size of the browser, the
pods inside the dashboard appears very small. Now when I increase
the size of the browser the size of the pods remains small. Now my
requirement is that the size of the pods should also get increased
along with the size of the browser. Please help me in this regard.
Thanks,
Nirmal Kumar BhogadiHi mersec,
That's Ok, I fixed the code like below, you could refer to.
Private Sub picturebox1_mouseup(sender As Object, ByVal e As MouseEventArgs) Handles PictureBox1.MouseUp
'Allows picture box to be dragged inside container
'Top and Left sides of picturebox1 will not exceed point 0,0
'This will stop any "grey space" from appearing outside these edges.
pbLoc.X = (PictureBox1.Location.X + (e.X - a))
pbLoc.Y = (PictureBox1.Location.Y + (e.Y - b))
If pbLoc.X + screenshot.Width > 0 And pbLoc.Y + screenshot.Height > 0 Then
If pbLoc.X > 0 Then pbLoc.X = 0
If pbLoc.Y > 0 Then pbLoc.Y = 0
If pbLoc.X + screenshot.Width - ClientSize.Width < 0 Then pbLoc.X = ClientSize.Width - screenshot.Width
If pbLoc.Y + screenshot.Height - ClientSize.Height < 0 Then pbLoc.Y = ClientSize.Height - screenshot.Height
PictureBox1.Location = New Point(pbLoc.X, pbLoc.Y)
Else
'else recovery the location
pbLoc.X = PictureBox1.Location.X
pbLoc.Y = PictureBox1.Location.Y
End If
End Sub
If you have any other concern regarding this issue, please feel free to let me know.
Best regards,
Youjun Tang
We are trying to better understand customer views on social support experience, so your participation in this interview project would be greatly appreciated if you have time. Thanks for helping make community forums a great place.
Click
HERE to participate the survey. -
After update to be iOS 6.0.1, Wi-Fi is not connected automatically. I have to reconnect or join my known network manually all the time. How can I solve this problem? It is so tedious.
Same here... on my girlfriend's iPod touch (4th gen).
It's very annoying... really! -
I am facing network issue in my 5 months old iphone 5s 32 gb.My carrier is Airtel and I am using in India.sometimes no service and many times unreachable in the main city.
I am using it in bangalore. I also didnot go outside of bangalore. and it is on in my settingd . eventhough facing the same isue. Do u have any other suggestion?
-
[SOLVED]Gnome: No network icon in the status bar
So fresh install and no networking icon in the status bar.
I also opened the network connections program and it didnt show any connections. I am using eth0 to write this.
This suggests to me that a daemon hasnt been started (but doesnt necessarily explain no icon in status bar)
here is the daemon section of my rc.conf
DAEMONS=(hwclock syslog-ng network netfs crond networkmanager dbus gdm)
i added network manager as a previous attempt (on advise of someone elsewhere on the internet(cannot remember where lol))
Thanks.. I have nearly got my system how I want it (minus software i havent installed yet)
Sorry if i have created a duplicate thread
Last edited by olig1905 (2012-01-21 17:33:47)Thanks to both of you.
I had already noticed the redundancy of network which was added in the install however hadnt considered the ordering
wonder wrote:Did you really read the wiki?
This is pretty much all i have seen people saying at the end of their posts on here. I agree with the point you have made, however i have been on the wiki all day, conisdering this is a newbie forum it is to be presumed that it is a pretty new installation. Therefore the user does not have time to read the entire wiki. As i said i have been on the wiki all day and used it countless times and have found it invaluable, so im only posting here when i am unable to find answers there.
That aside this is a wonderfully helpful community so HI -
Crazy thing: No dialtone from a SPA 3102 inside the same network.
Hi folks: We have a couple of SPA-3102 connected to two POTS Bellsouth lines. Both 3102 are registering fine with our SIP server and also register the phone line.
The crazy thing is this: If we try to get dialtone calling from an IP phone (SPA922) LOCATED IN THE SAME LOCAL NETWORK (I.E: In the office) to the 3102 number, we get no dialtone.
If i try co call the same 3102 from a softphone (x-lite) also in the office, we still get no dialtone.
But if i try to call the spa from other network than the office, (I.e: My home), i´m getting the dialtone!!! I tried it with both the 922 IP Phone and the x-lite softphone and works fine.
Also, we tried to call the 3102 from other IP phone in our office in Mexico, and also works fine.
Our network topology in the office: Motorola´s Netopia 3347-02 with Bellsouth xDSL internet service. Two Linksys EZXS16W swithes, and a mix of laptops, PCs, servers, IPPhones (SPA922 and 942) and printers connected to these switches
Any idea ??
Thanks in advance.
MartinIf the in-office X-Lite is accessing the SPA-3102 on its private IP, try the public IP, or vice-versa.
I'm presuming that the SPA-3102 has a static private address, the Netopia is forwarding some UDP ports to the 3102, and that you are accessing the 3102 directly (not through a server such as Asterisk). In this case, access from the LAN via the public IP may fail, because many routers don't handle 'hairpinning' correctly. Access via the private IP may fail, because the client is being confused by the NAT mapping being applied by the 3102. You can use Wireshark on the machine running X-Lite to see what's happening in each case.
If your situation is other than the above, please provide the details: Accessed via server? If so, is server on LAN or outside? What software? Do you have administrative control over it? Does your router have a SIP ALG? Is it being used in this path? What port forwarding or other special setup on router? What NAT mapping parameters on SPA-3102? What have you seen with Wireshark? What remedial actions have you tried, with what results? Do you have multiple static public IPs? If so, is there a spare one?
Message Edited by Stew on 05-05-2008 10:48 AM -
RE: Is there a way to have 2 network cards on the samemachine
Sean,
I mean I am always connecting to the nodemgr of the 2nd environment
successfully. Only when I run the client part of the application that I know
is up, I get a response from the Name Server that it is actually the first
IP address.
I checked again. I defenitely have FORTE_NS_ADDRESS set to IP:5004, in my
case, and not the hostname. I don't have FORTE_LOCATIONS set any where. the
NS_ADDRESS for env2 has only 1 IP address associated with it. My Forte
Control Panel on the client, and hence the NS address has only one entry.
I tried going into escript like you said, and the name service still thinks
it is IP1.
Venkat Kodumudi
Price Waterhouse LLP
Internet: [email protected]
Internet2: [email protected]
-----Original Message-----
From: Sean Brown [SMTP:[email protected]]
Sent: Monday, June 22, 1998 1:31 PM
To: Venkat Kodumudi; 'Sean Brown'
Cc: [email protected]
Subject: RE: Is there a way to have 2 network cards on the same
machine
Hmmmm? That is a little odd! Let me rehash what I think you are saying.
You now have two environments each with their own name service with
FORTE_NS_ADDRESS set to a different IP:Port combination. For example you
are doing the steps:
FORTE_NS_ADDRESS=255.255.255.1:5000
start the nodemgr for env 1
FORTE_NS_ADDRESS=255.255.255.2:5000
start the nodemgr for env 2
Now when you start a client you are always getting the address for env 1
back from the nodemgr. I assume you actually mean the name service? Or, do
you mean you are always connecting to the nodemgr & name service for env
1?
If the first scenario is the case and you are connecting to the nodemgr
for
env 2 but getting back IP's for services listening on the card for env 1 I
would ask you what you are setting the FORTE_LOCATIONS value to before you
start each service. If you are not setting it or are using the host name
it
will register using the IP for the primary network card associated with
the
machine name and I am again assuming that this is env 1. You need to set
the FORTE_LOCATIONS variable to 255.255.255.2:0 (based on the steps above)
before starting your services.
If the second scenario is the case I would have you check what the
FORTE_NS_ADDRESS is set to before you start up the client. Once again it
must be the IP:Port combination not host:port combination if you want to
get
anything other than the primary card.
I would also suggest that you do the following. After everything is up
and
running execute the following commands:
escript -fns "ip for env1":port
findsub nameservice
showpart
What you should see is everything currently registered under the name
service. It will have the name and any "locations" (IP and port) that it
is
registered as listening on. I would look for the nodemgr and see where it
has advertised itself. I would then look for any services you expect to
be
registered there and also verify where the have advertised themselves. If
there are multiple locations listed for any one service, the client will
use
the first one in the list.
Do the same for env 2.
Sean
-----Original Message-----
From: Venkat Kodumudi [mailto:[email protected]]
Sent: Monday, June 22, 1998 9:57 AM
To: 'Sean Brown'
Cc: [email protected]
Subject: RE: Is there a way to have 2 network cards on the same machine
I am giving the actual IP address. and not the host name. That is why I
don't understand what is going on.
Venkat Kodumudi
Price Waterhouse LLP
Internet: [email protected]
Internet2: [email protected]
-----Original Message-----
From: Sean Brown [SMTP:[email protected]]
Sent: Monday, June 22, 1998 10:53 AM
To: Venkat Kodumudi; 'Sean Brown'
Cc: [email protected]
Subject: RE: Is there a way to have 2 network cards on the same
machine
Hello Venkat,
I probably should have mentioned this before. Your are correct. Forteis
doing a host lookup if you are providing a name for examplemachine1:5000.
You can bypass the host lookup by using the actual ip dot addressinstead
for example 255.255.255.255:5000. This way you are taking the name
service
out od the picture and Forte will use the address provided.
Sean
-----Original Message-----
From: Venkat Kodumudi [mailto:[email protected]]
Sent: Monday, June 22, 1998 9:27 AM
To: 'Sean Brown'
Cc: [email protected]
Subject: RE: Is there a way to have 2 network cards on the same machine
My requirement is that my second card serves as best case performance
testing piece. This eliminates the network completely. We went one step
ahead and created a new enviromnent for the second card. Whatever I do,
the
nodemgr is returning back the IP address of the first card, even thoughmy
FORTE_NS_ADDRESS does not have the first card in the picture any where.
I think Forte is doing a host look up and returning the first IP address
it
finds, as opposed to returning the IP address specified in theenvironment
variable FORTE_NS_ADDRESS. Is there a way to trick it?
Venkat Kodumudi
Price Waterhouse LLP
Internet: [email protected]
Internet2: [email protected]
-----Original Message-----
From: Sean Brown [SMTP:[email protected]]
Sent: Thursday, June 18, 1998 9:54 AM
To: Venkat Kodumudi
Cc: [email protected]
Subject: RE: Is there a way to have 2 network cards on the same
machine
OK, you ran into one of the issues. That is, if both addresses areknown
to
the client that is trying to contact the partition it will always usethe
first address in FORTE_LOCATIONS. This is because FORTE_LOCATIONS was
designed more as a failover mechanism. So it will always try the
first
address in the list and if it succeeds, there is no reason to move onto
the
second.
Now, the second issue is that there is currently a problem with theclient
failover to the secondary address in FORTE_LOCATIONS. If the firstentry
fails it is supposed to retry on the second entry. Instead, it
retries
the
first entry again. I know that Forte knows about this but I do nothave
a
bug number on it.
With that said, lets look at a possible solution for you. If the real
objective here is to have a back up network card available for fail
over
on
the same machine, or use one card to advertise outside your firewalland
one
to use inside, then you will have to contact Forte to determine whenthe
failover problem will be fixed. But, if the objective is to loadbalance
across the network cards you could have the environment manager listenon
both ports and then alternate your server partitions across both
cards.
For
example:
set FORTE_NS_ADDRESS=card1:5000;card2:5000 and then start up the
environment
manager
set FORTE_LOCATIONS=card1:0 (the 0 in the port causes the OS to pick a
port)
and start partition one
set FORTE_LOCATIONS=card2:0 and start partition two
and so on....
In this scenario the environment manager will be listening on bothcards
but
each server will be listening on only one of the two cards. So if a
request
comes in for partition1 it will go through card one and if it is for
partition two it will go through card two. You could assign your
partitions
to cards based on expected load.
Well, I am done. I hope this helps!
Sean
-----Original Message-----
From: Venkat Kodumudi [mailto:[email protected]]
Sent: Thursday, June 18, 1998 8:06 AM
To: 'Sean Brown'; 'John Jamison'
Cc: [email protected]; Jose Suriol
Subject: RE: Is there a way to have 2 network cards on the samemachine
Sean,
Thanks for your reply. I tried the approach. I was not very specificin
my
question. I do need the ability for server applications to listen and
server
on both the network cards.
I was succesfully able to make the nodemgr listen on both the cardsand
actually serve requests coming in from both the cards. But, followingyour
advise, I took a cautious step with FORTE_LOCATIONS. Here is what I
noticed.
I have an application that has 6 partitions in total. I used
FORTE_LOCATION
to make it listen on 1. Both the cards. 2. Swapped the IP addresses
for
both
cards for this application. 3. One card that I want it to listen on. I
tried
all approaches by exporting the locations variable for just this
application. The nodemgr recieves a request from this pc connected onthe
second card to talk to one of the partitions. The node mgr responds
with
a
proxy - with the ip address and socket number of the first card. The
FORTE_NS_ADDRESS variable looks like this:
IP1:5002;IP2:5002.
Is it possible atall to resolve my problem, without having a seperate
environment?
Thanks
Venkat Kodumudi
Price Waterhouse LLP
Internet: [email protected]
Internet2: [email protected]
-----Original Message-----
From: Sean Brown [SMTP:[email protected]]
Sent: Wednesday, June 17, 1998 10:42 AM
To: 'John Jamison'; Venkat Kodumudi
Cc: [email protected]
Subject: RE: Is there a way to have 2 network cards on the
same
machine
Venkat,
Actually, it is possible for Forte to listen on more than one IP andport
combination. The first reply to your message was correct. If you
set
the
FORTE_NS_ADDRESS to contain multiple entries before starting the
name
service, it will advertise on both. For Forte servers you use the
FORTE_LOCATIONS env variable to get it to advertise on multipleip:port
combinations.
We were doing something very similar with another customer I was at
to
get
around a firewall. I will warn you that there are some issues with
FORTE_LOCATIONS that may keep that portion from working. However,
from
reading your note, it appears that all you need is for the nameservice
to
advertise and listen on multiple ports and that works fine. I justtested
it again for sanity sake and it worked. I ran my test on NT using
Forte
3G2.
Sean
-----Original Message-----
From: [email protected]
[<a href="mailto:[email protected]">mailto:[email protected]]On</a> Behalf Of John Jamison
Sent: Tuesday, June 16, 1998 4:51 PM
To: Venkat Kodumudi
Cc: [email protected]
Subject: Re: Is there a way to have 2 network cards on the samemachine
Venkat,
Technically yes this is possible, though not in Forte. A nameserver
can
only listen on one port.
To implement this scheme you will have to write a proxy service (insome
language
including perhaps forte) which listens on the well-known port on the
second card, reads requests, then forwards them to the realnameservice
(wkp on the first card), and forwards replies back. This is not
trivial, but some firewall toolkit vendors supply stub code to write
application specific proxies.
-J
Venkat Kodumudi wrote:
Folks,
Here is what we would like to do:
We want to have 2 network cards on a unix box - which means I have
2
ip
addresses, and the connection between the two is the unix box and
only
the
unix box. I have a pc connected to the 2nd network card and I want
it
to
connect to the nameserver that is listening on a well known port
on
the
first network card. We don't want to turn IP forwarding between
the
two
cards. We want Forte to address both cards to talk to clients, in
one
environment.
Can this be done? If so how?
Thanks in advance.
Venkat Kodumudi
Price Waterhouse LLP
Internet: [email protected]
Internet2: [email protected]
To unsubscribe, email '[email protected]' with
'unsubscribe forte-users' as the body of the message.
Searchable thread archive<URL:<a href=
"http://pinehurst.sageit.com/listarchive/">http://pinehurst.sageit.com/listarchive/</a>>
>>>>
John Jamison [email protected]
Vice President and Chief Technology Officer
Sage IT Partners, Inc.
Voice: 415 392-7243 x 306
Fax: 415 391-3899
Internet Enabled Business Change
<a href=
"http://www.sageit.com">http://www.sageit.com</a>
To unsubscribe, email '[email protected]' with
'unsubscribe forte-users' as the body of the message.
Searchable thread archive <URL:<a href=
"http://pinehurst.sageit.com/listarchive/">http://pinehurst.sageit.com/listarchive/</a>>Hello Venkat,
I probably should have mentioned this before. Your are correct. Forte is
doing a host lookup if you are providing a name for example machine1:5000.
You can bypass the host lookup by using the actual ip dot address instead
for example 255.255.255.255:5000. This way you are taking the name service
out od the picture and Forte will use the address provided.
Sean
-----Original Message-----
From: Venkat Kodumudi [mailto:[email protected]]
Sent: Monday, June 22, 1998 9:27 AM
To: 'Sean Brown'
Cc: [email protected]
Subject: RE: Is there a way to have 2 network cards on the same machine
My requirement is that my second card serves as best case performance
testing piece. This eliminates the network completely. We went one step
ahead and created a new enviromnent for the second card. Whatever I do, the
nodemgr is returning back the IP address of the first card, even though my
FORTE_NS_ADDRESS does not have the first card in the picture any where.
I think Forte is doing a host look up and returning the first IP address it
finds, as opposed to returning the IP address specified in the environment
variable FORTE_NS_ADDRESS. Is there a way to trick it?
Venkat Kodumudi
Price Waterhouse LLP
Internet: [email protected]
Internet2: [email protected]
-----Original Message-----
From: Sean Brown [SMTP:[email protected]]
Sent: Thursday, June 18, 1998 9:54 AM
To: Venkat Kodumudi
Cc: [email protected]
Subject: RE: Is there a way to have 2 network cards on the same
machine
OK, you ran into one of the issues. That is, if both addresses are known
to
the client that is trying to contact the partition it will always use the
first address in FORTE_LOCATIONS. This is because FORTE_LOCATIONS was
designed more as a failover mechanism. So it will always try the first
address in the list and if it succeeds, there is no reason to move on to
the
second.
Now, the second issue is that there is currently a problem with the client
failover to the secondary address in FORTE_LOCATIONS. If the first entry
fails it is supposed to retry on the second entry. Instead, it retries
the
first entry again. I know that Forte knows about this but I do not have a
bug number on it.
With that said, lets look at a possible solution for you. If the real
objective here is to have a back up network card available for fail over
on
the same machine, or use one card to advertise outside your firewall and
one
to use inside, then you will have to contact Forte to determine when the
failover problem will be fixed. But, if the objective is to load balance
across the network cards you could have the environment manager listen on
both ports and then alternate your server partitions across both cards.
For
example:
set FORTE_NS_ADDRESS=card1:5000;card2:5000 and then start up the
environment
manager
set FORTE_LOCATIONS=card1:0 (the 0 in the port causes the OS to pick a
port)
and start partition one
set FORTE_LOCATIONS=card2:0 and start partition two
and so on....
In this scenario the environment manager will be listening on both cards
but
each server will be listening on only one of the two cards. So if a
request
comes in for partition1 it will go through card one and if it is for
partition two it will go through card two. You could assign your
partitions
to cards based on expected load.
Well, I am done. I hope this helps!
Sean
-----Original Message-----
From: Venkat Kodumudi [mailto:[email protected]]
Sent: Thursday, June 18, 1998 8:06 AM
To: 'Sean Brown'; 'John Jamison'
Cc: [email protected]; Jose Suriol
Subject: RE: Is there a way to have 2 network cards on the same machine
Sean,
Thanks for your reply. I tried the approach. I was not very specific in
my
question. I do need the ability for server applications to listen and
server
on both the network cards.
I was succesfully able to make the nodemgr listen on both the cards and
actually serve requests coming in from both the cards. But, following your
advise, I took a cautious step with FORTE_LOCATIONS. Here is what I
noticed.
I have an application that has 6 partitions in total. I used
FORTE_LOCATION
to make it listen on 1. Both the cards. 2. Swapped the IP addresses for
both
cards for this application. 3. One card that I want it to listen on. I
tried
all approaches by exporting the locations variable for just this
application. The nodemgr recieves a request from this pc connected on the
second card to talk to one of the partitions. The node mgr responds with a
proxy - with the ip address and socket number of the first card. The
FORTE_NS_ADDRESS variable looks like this:
IP1:5002;IP2:5002.
Is it possible atall to resolve my problem, without having a seperate
environment?
Thanks
Venkat Kodumudi
Price Waterhouse LLP
Internet: [email protected]
Internet2: [email protected]
-----Original Message-----
From: Sean Brown [SMTP:[email protected]]
Sent: Wednesday, June 17, 1998 10:42 AM
To: 'John Jamison'; Venkat Kodumudi
Cc: [email protected]
Subject: RE: Is there a way to have 2 network cards on the same
machine
Venkat,
Actually, it is possible for Forte to listen on more than one IP andport
combination. The first reply to your message was correct. If you setthe
FORTE_NS_ADDRESS to contain multiple entries before starting the name
service, it will advertise on both. For Forte servers you use the
FORTE_LOCATIONS env variable to get it to advertise on multiple ip:port
combinations.
We were doing something very similar with another customer I was at toget
around a firewall. I will warn you that there are some issues with
FORTE_LOCATIONS that may keep that portion from working. However, from
reading your note, it appears that all you need is for the name serviceto
advertise and listen on multiple ports and that works fine. I justtested
it again for sanity sake and it worked. I ran my test on NT using Forte
3G2.
Sean
-----Original Message-----
From: [email protected]
[<a href="mailto:[email protected]">mailto:[email protected]]On</a> Behalf Of John Jamison
Sent: Tuesday, June 16, 1998 4:51 PM
To: Venkat Kodumudi
Cc: [email protected]
Subject: Re: Is there a way to have 2 network cards on the same machine
Venkat,
Technically yes this is possible, though not in Forte. A name server
can
only listen on one port.
To implement this scheme you will have to write a proxy service (in some
language
including perhaps forte) which listens on the well-known port on the
second card, reads requests, then forwards them to the real nameservice
(wkp on the first card), and forwards replies back. This is not
trivial, but some firewall toolkit vendors supply stub code to write
application specific proxies.
-J
Venkat Kodumudi wrote:
Folks,
Here is what we would like to do:
We want to have 2 network cards on a unix box - which means I have 2ip
addresses, and the connection between the two is the unix box and onlythe
unix box. I have a pc connected to the 2nd network card and I want it
to
connect to the nameserver that is listening on a well known port onthe
first network card. We don't want to turn IP forwarding between thetwo
cards. We want Forte to address both cards to talk to clients, in one
environment.
Can this be done? If so how?
Thanks in advance.
Venkat Kodumudi
Price Waterhouse LLP
Internet: [email protected]
Internet2: [email protected]
To unsubscribe, email '[email protected]' with
'unsubscribe forte-users' as the body of the message.
Searchable thread archive<URL:<a href=
"http://pinehurst.sageit.com/listarchive/">http://pinehurst.sageit.com/listarchive/</a>>
>>
John Jamison [email protected]
Vice President and Chief Technology Officer
Sage IT Partners, Inc.
Voice: 415 392-7243 x 306
Fax: 415 391-3899
Internet Enabled Business Change
<a href="http://www.sageit.com">http://www.sageit.com</a>
To unsubscribe, email '[email protected]' with
'unsubscribe forte-users' as the body of the message.
Searchable thread archive <URL:<a href=
"http://pinehurst.sageit.com/listarchive/">http://pinehurst.sageit.com/listarchive/</a>> -
Is there a way to have 2 network cards on the samemachine
Folks,
Here is what we would like to do:
We want to have 2 network cards on a unix box - which means I have 2 ip
addresses, and the connection between the two is the unix box and only the
unix box. I have a pc connected to the 2nd network card and I want it to
connect to the nameserver that is listening on a well known port on the
first network card. We don't want to turn IP forwarding between the two
cards. We want Forte to address both cards to talk to clients, in one
environment.
Can this be done? If so how?
Thanks in advance.
Venkat Kodumudi
Price Waterhouse LLP
Internet: [email protected]
Internet2: [email protected]
To unsubscribe, email '[email protected]' with
'unsubscribe forte-users' as the body of the message.
Searchable thread archive <URL:http://pinehurst.sageit.com/listarchive/>Hmmmm? That is a little odd! Let me rehash what I think you are saying.
You now have two environments each with their own name service with
FORTE_NS_ADDRESS set to a different IP:Port combination. For example you
are doing the steps:
FORTE_NS_ADDRESS=255.255.255.1:5000
start the nodemgr for env 1
FORTE_NS_ADDRESS=255.255.255.2:5000
start the nodemgr for env 2
Now when you start a client you are always getting the address for env 1
back from the nodemgr. I assume you actually mean the name service? Or, do
you mean you are always connecting to the nodemgr & name service for env 1?
If the first scenario is the case and you are connecting to the nodemgr for
env 2 but getting back IP's for services listening on the card for env 1 I
would ask you what you are setting the FORTE_LOCATIONS value to before you
start each service. If you are not setting it or are using the host name it
will register using the IP for the primary network card associated with the
machine name and I am again assuming that this is env 1. You need to set
the FORTE_LOCATIONS variable to 255.255.255.2:0 (based on the steps above)
before starting your services.
If the second scenario is the case I would have you check what the
FORTE_NS_ADDRESS is set to before you start up the client. Once again it
must be the IP:Port combination not host:port combination if you want to get
anything other than the primary card.
I would also suggest that you do the following. After everything is up and
running execute the following commands:
escript -fns "ip for env1":port
findsub nameservice
showpart
What you should see is everything currently registered under the name
service. It will have the name and any "locations" (IP and port) that it is
registered as listening on. I would look for the nodemgr and see where it
has advertised itself. I would then look for any services you expect to be
registered there and also verify where the have advertised themselves. If
there are multiple locations listed for any one service, the client will use
the first one in the list.
Do the same for env 2.
Sean
-----Original Message-----
From: Venkat Kodumudi [mailto:[email protected]]
Sent: Monday, June 22, 1998 9:57 AM
To: 'Sean Brown'
Cc: [email protected]
Subject: RE: Is there a way to have 2 network cards on the same machine
I am giving the actual IP address. and not the host name. That is why I
don't understand what is going on.
Venkat Kodumudi
Price Waterhouse LLP
Internet: [email protected]
Internet2: [email protected]
-----Original Message-----
From: Sean Brown [SMTP:[email protected]]
Sent: Monday, June 22, 1998 10:53 AM
To: Venkat Kodumudi; 'Sean Brown'
Cc: [email protected]
Subject: RE: Is there a way to have 2 network cards on the same
machine
Hello Venkat,
I probably should have mentioned this before. Your are correct. Forte is
doing a host lookup if you are providing a name for example machine1:5000.
You can bypass the host lookup by using the actual ip dot address instead
for example 255.255.255.255:5000. This way you are taking the name
service
out od the picture and Forte will use the address provided.
Sean
-----Original Message-----
From: Venkat Kodumudi [mailto:[email protected]]
Sent: Monday, June 22, 1998 9:27 AM
To: 'Sean Brown'
Cc: [email protected]
Subject: RE: Is there a way to have 2 network cards on the same machine
My requirement is that my second card serves as best case performance
testing piece. This eliminates the network completely. We went one step
ahead and created a new enviromnent for the second card. Whatever I do,
the
nodemgr is returning back the IP address of the first card, even though my
FORTE_NS_ADDRESS does not have the first card in the picture any where.
I think Forte is doing a host look up and returning the first IP address
it
finds, as opposed to returning the IP address specified in the environment
variable FORTE_NS_ADDRESS. Is there a way to trick it?
Venkat Kodumudi
Price Waterhouse LLP
Internet: [email protected]
Internet2: [email protected]
-----Original Message-----
From: Sean Brown [SMTP:[email protected]]
Sent: Thursday, June 18, 1998 9:54 AM
To: Venkat Kodumudi
Cc: [email protected]
Subject: RE: Is there a way to have 2 network cards on the same
machine
OK, you ran into one of the issues. That is, if both addresses areknown
to
the client that is trying to contact the partition it will always usethe
first address in FORTE_LOCATIONS. This is because FORTE_LOCATIONS was
designed more as a failover mechanism. So it will always try the first
address in the list and if it succeeds, there is no reason to move on to
the
second.
Now, the second issue is that there is currently a problem with theclient
failover to the secondary address in FORTE_LOCATIONS. If the firstentry
fails it is supposed to retry on the second entry. Instead, it retries
the
first entry again. I know that Forte knows about this but I do not havea
bug number on it.
With that said, lets look at a possible solution for you. If the real
objective here is to have a back up network card available for fail over
on
the same machine, or use one card to advertise outside your firewall and
one
to use inside, then you will have to contact Forte to determine when the
failover problem will be fixed. But, if the objective is to loadbalance
across the network cards you could have the environment manager listenon
both ports and then alternate your server partitions across both cards.
For
example:
set FORTE_NS_ADDRESS=card1:5000;card2:5000 and then start up the
environment
manager
set FORTE_LOCATIONS=card1:0 (the 0 in the port causes the OS to pick a
port)
and start partition one
set FORTE_LOCATIONS=card2:0 and start partition two
and so on....
In this scenario the environment manager will be listening on both cards
but
each server will be listening on only one of the two cards. So if a
request
comes in for partition1 it will go through card one and if it is for
partition two it will go through card two. You could assign your
partitions
to cards based on expected load.
Well, I am done. I hope this helps!
Sean
-----Original Message-----
From: Venkat Kodumudi [mailto:[email protected]]
Sent: Thursday, June 18, 1998 8:06 AM
To: 'Sean Brown'; 'John Jamison'
Cc: [email protected]; Jose Suriol
Subject: RE: Is there a way to have 2 network cards on the same machine
Sean,
Thanks for your reply. I tried the approach. I was not very specific in
my
question. I do need the ability for server applications to listen and
server
on both the network cards.
I was succesfully able to make the nodemgr listen on both the cards and
actually serve requests coming in from both the cards. But, followingyour
advise, I took a cautious step with FORTE_LOCATIONS. Here is what I
noticed.
I have an application that has 6 partitions in total. I used
FORTE_LOCATION
to make it listen on 1. Both the cards. 2. Swapped the IP addresses for
both
cards for this application. 3. One card that I want it to listen on. I
tried
all approaches by exporting the locations variable for just this
application. The nodemgr recieves a request from this pc connected onthe
second card to talk to one of the partitions. The node mgr responds witha
proxy - with the ip address and socket number of the first card. The
FORTE_NS_ADDRESS variable looks like this:
IP1:5002;IP2:5002.
Is it possible atall to resolve my problem, without having a seperate
environment?
Thanks
Venkat Kodumudi
Price Waterhouse LLP
Internet: [email protected]
Internet2: [email protected]
-----Original Message-----
From: Sean Brown [SMTP:[email protected]]
Sent: Wednesday, June 17, 1998 10:42 AM
To: 'John Jamison'; Venkat Kodumudi
Cc: [email protected]
Subject: RE: Is there a way to have 2 network cards on the same
machine
Venkat,
Actually, it is possible for Forte to listen on more than one IP andport
combination. The first reply to your message was correct. If you setthe
FORTE_NS_ADDRESS to contain multiple entries before starting the name
service, it will advertise on both. For Forte servers you use the
FORTE_LOCATIONS env variable to get it to advertise on multiple
ip:port
combinations.
We were doing something very similar with another customer I was at toget
around a firewall. I will warn you that there are some issues with
FORTE_LOCATIONS that may keep that portion from working. However,
from
reading your note, it appears that all you need is for the nameservice
to
advertise and listen on multiple ports and that works fine. I justtested
it again for sanity sake and it worked. I ran my test on NT using
Forte
3G2.
Sean
-----Original Message-----
From: [email protected]
[<a href="mailto:[email protected]">mailto:[email protected]]On</a> Behalf Of John Jamison
Sent: Tuesday, June 16, 1998 4:51 PM
To: Venkat Kodumudi
Cc: [email protected]
Subject: Re: Is there a way to have 2 network cards on the samemachine
Venkat,
Technically yes this is possible, though not in Forte. A nameserver
can
only listen on one port.
To implement this scheme you will have to write a proxy service (insome
language
including perhaps forte) which listens on the well-known port on the
second card, reads requests, then forwards them to the realnameservice
(wkp on the first card), and forwards replies back. This is not
trivial, but some firewall toolkit vendors supply stub code to write
application specific proxies.
-J
Venkat Kodumudi wrote:
Folks,
Here is what we would like to do:
We want to have 2 network cards on a unix box - which means I have 2
ip
addresses, and the connection between the two is the unix box and
only
the
unix box. I have a pc connected to the 2nd network card and I want
it
to
connect to the nameserver that is listening on a well known port onthe
first network card. We don't want to turn IP forwarding between thetwo
cards. We want Forte to address both cards to talk to clients, in
one
environment.
Can this be done? If so how?
Thanks in advance.
Venkat Kodumudi
Price Waterhouse LLP
Internet: [email protected]
Internet2: [email protected]
To unsubscribe, email '[email protected]' with
'unsubscribe forte-users' as the body of the message.
Searchable thread archive<URL:<a href=
"http://pinehurst.sageit.com/listarchive/">http://pinehurst.sageit.com/listarchive/</a>>
>>>
John Jamison [email protected]
Vice President and Chief Technology Officer
Sage IT Partners, Inc.
Voice: 415 392-7243 x 306
Fax: 415 391-3899
Internet Enabled Business Change
<a href=
"http://www.sageit.com">http://www.sageit.com</a>
To unsubscribe, email '[email protected]' with
'unsubscribe forte-users' as the body of the message.
Searchable thread archive <URL:<a href=
"http://pinehurst.sageit.com/listarchive/">http://pinehurst.sageit.com/listarchive/</a>> -
How should I add a new physical network interface to the system?
Hello:
I am about to test my own NIC driver.
I have used add_drv to add my driver.
I have assigned an IP address to the sytem in the /etc/hosts file and corresponding network mask in the /etc/netmasks file, too.
But when I tried to ping this IP address, it reported an error message saying"ICMP Host Unreachable from gateway localhost .....".
I wonder if the system knows this physical network address.
Does this have something to do with the /etc/hostname.* file?
If this is the case, what should I name the "*" part of this file?
Any advice will be appreciated.
StevenHello Shridhar:
Thanks for your kindly reply.
I just noticed that the system would try to plumb the interface during the booting process but failed with a error saying "ifconfig:plumb:XXXX:Bad file number", where XXXX is the * part of /etc/hostname.*
Does not mean I have assigned a illegal interface name to this interface?
It seems that man page doesn't contain any information about error messages.
I use the form XXX0, where XXX is the name of the NIC driver.
Any advice will be appreciated.
Steven
You should plumb the interface and bring it up with
the ifconfig command. Do a man on ifconfig for more
details. What was the add_drv command used ? Based on
your driver's name, you can setup the hostname.*
file's name to automatically setup and initialize the
NIC inetrface everytime the system boots up.
HTH
shridhar -
No CDs inside the package.
I have just bought my mac recently. Why does not it contain CDs inside the package? My old mac comes with two CDs. This new mac has no CDs. I have problem with installing software. After I install the sofware, the OS X disappears. I need the CDs to recover my OS.
Are you running OSX Lion? Most likely you are if you have just purchased a Mac and no install media was included.
Macs with OSX Lion installed do not ship with media.
There is a hidden recovery partition on your drive. Startup holding command-R and boot into Recovery Mode.
You can re-install OSX Lion from the menu selection. You will need a network connection to download the OSX components. -
Can't able to access the Jar files which is kept inside the folder.
Hi,
I am using the eclipse IDE, my project folder contains the folders src, bin, .metadata, .settings
if i place the jar file in the same location of the above mentioned folder means i can able to use the Jar file in eclipse(in my project).
But i create one folder in the same location in the name of Jar or anything and all jar files are placed inside the folder,then i can't able to access the jar file in eclipse(in my project).
How i can solve this problem?RajivGuna wrote:
..How i can solve this problem?Put Eclipse aside for the moment and learn how the SDK tools work at the level of the command line, then you will be much better placed to figure how to do it in the IDE. If you are still having problems, I suggest you ask on an [Eclipse forum|http://www.eclipse.org/forums/]. -
[SOLVED] Network won't work after install
Hi
I've installed arch on a lot of different computers without any trouble. But now I'm trying to install it on a intel mini-ITX board with an atom processor and integrated network. When setting up the network during the install it downloads and installs the packages perfectly fine, but when the installation is finished and I reboot into my new system the network wont work.
If I try to ping my gateway at 192.168.1.1 it just says that the network is unreachable.
The ifconfig shows the following for eth0:
Link encap:Ethernet HWAddr 00:27:0E:05:AC:47
BROADCAST MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 overruns:0 frame:0
TX packets:0 errors:0 overruns:0 frame:0
collisions:0 TX bytes:0 (0.0 b)
Interrupt:28 Base address:0xe000
Any ideas what could be wrong?
Thanks
//Lasse
Last edited by Lazze (2010-02-10 14:57:30)Sorry I've been away for a couple of days, but it's solved now it was something in the hosts file which weren't correct.
Thanks for the reply though! -
PayloadZipBean - variable filename inside the archive
Hi,
I'm using the Adapter-Specific Message Properties in a sender fileadapter to determine the incoming filename, which has variable components (i.e. date and time). After changing the prefix and suffix in a Message-Mapping the new name is applied in a receiver fileadapter. The new name is now part of the payload in the tag <Filename>. The receiver fileadapter contains the PayloadZipBean to compress the output file, everything works fine.
Now I want the file inside my compressed archive always to have the same name as my archive has. In Stefan Grube's blog about the PayloadZipBean I only found the possibility to choose a fix name by using the MessageTransformBean (name="file.txt").
Example:
1. variable input filename: IN_RNK_20080227_1006.txt
2. filename after Message-Mapping: OUT_RNK_20080227_1006.gz
3. archive filename after receiver adapter: OUT_RNK_20080227_1006.gz
filename inside the archive: MainDocument (this is the default (payload-name))
I would like to name the file inside the archive OUT_RNK_20080227_1006, too.
Question 1: Is there any possibility to read out the mapped <Filename>-tag from the payload to use it for the element inside my archive (File Name Scheme and Variable Substitution don't work when using Adapter-Specific Message Properties).
Question 2: Is there a way to modify the name of the payload so that I can change "MainDocument" to "OUT_RNK_20080227_1006"?
I'm on XI 3.0 - SP19.
Many thanks,
Ralph>
Stefan Grube wrote:
> In this thread there is a module for reading the content type and setting the dynamic configuration:
> sender mail adapter - attachment name
> It should be not be so difficult to derive the reverse way
Thanks for this info Stefan.
[Stefan Grube's webinar on custom adapter module development|https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/64a6bdab-0c01-0010-079a-b3707717cecd?prtmode=navigate]
The code of the EJB module we had to write to get the filename from the Dynamic Configuration (ASMA) and then set the content type of the message is given below.
* Created on Apr 23, 2008
package zfilezipper;
* @author Krishneel Goundar
import javax.ejb.CreateException;
import javax.ejb.SessionBean;
import javax.ejb.SessionContext;
import com.sap.aii.af.mp.module.*;
import com.sap.aii.af.ra.ms.api.*;
* @ejbHome<{com.sap.aii.af.mp.module.ModuleHome}>
* @ejbLocal<{com.sap.aii.af.mp.module.ModuleLocal}>
* @ejbLocalHome<{com.sap.aii.af.mp.module.ModuleLocalHome}>
* @ejbRemote<{com.sap.aii.af.mp.module.ModuleRemote}>
* @stateless
public class SetContentTypeEJB implements SessionBean, Module{
private SessionContext myContext;
public void ejbRemove() {}
public void ejbActivate() {}
public void ejbPassivate() {}
public void setSessionContext(SessionContext context) {
myContext = context;
public void ejbCreate() throws CreateException{}
public ModuleData process(ModuleContext moduleContext,ModuleData inputModuleData) throws ModuleException{
try {
Message msg = (Message) inputModuleData.getPrincipalData(); //Used to read dynamic configuration data
TextPayload payload = msg.getDocument(); //Used to set 'contentType' value
//The name of the file to be zipped is read from the dynamic configuration.
String fileName = msg.getMessageProperty("http://sap.com/xi/XI/System/File", "FileName");
if(fileName == null) //If no file name can be determined we set 'contentType' to "defaultbeanile.txt"
payload.setContentType("text/plain;charset = \"UTF-8\";filename=\"defaultbeanfile.txt\"");
else //If a file name can be found we set 'contentType' to the name of the file.
payload.setContentType("text/plain;charset = \"UTF-8\";filename=\"" + fileName + "\"");
//After setting the value of 'contentType' we need to update the ModuleData object (inputModuleData).
inputModuleData.setPrincipalData(msg);
} catch (Exception e) {
ModuleException me = new ModuleException(e);
throw me;
return inputModuleData; //Return the updated ModuleData object.
Edited by: Charu Kulkarni on Apr 28, 2008 1:41 AM
Maybe you are looking for
-
PO/ S.Location and GR msg check box in 'delivery/invoice' tab of header dat
Hi all, Under which case the below 2 fields are required to b filled while creating a PO apart from making them mandatory fieds in document type settings. 1. GR msg check box in 'delivery/invoice' tab of header data 2. Storage Location data in item d
-
Bridge CS6 crashes with double-click, not with "open with"
Bridge cs6 keeps crashing whenever I double click on a file (jpg/psd/raw files). After reading on this forum I couldn't find anything that worked for me. The strange thing is, when I select a file right click open with photoshop CS6 (default) t
-
I want to rest my apple TV. I run diagnostics and everything is ok. All I want to do is reset. I tried to reset using the remote control by pushing menu and - at the same time for 6 six seconds. I got nothing. Is there anything else to do? Is my appl
-
Open a Dreamweaver Page in a new tab
Has anyone been able to create a hyperlink that opens a new tab rather than a new instance of the browser and if so, how?
-
Before our upgrade, campus manager, specifically topology services network topology layer 2 map, would auto-refresh when a device went down, or a new device came online... after a data collection of course. now it requires a complete shutdown of the