Bridged ethernet suddenly stopped working!

strells

strells

Product Expert
I have already filed report 6750693 for any Parallels folks who might be looking at this.

At work, I use Airport on the Mac side with my Sprint Overdrive to access the Internet. I use bridged ethernet in my XP VM (PD6) to connect to internal work sites and drives. This worked perfectly fine until today.

As of today, I can get a valid internal IP address for the VM ethernet and even see all the correct info when I do a "ipconfig /all". However, I cannot ping anything other than the VM IP itself, not even the default gateway. How could I get a valid IP with DHCP and then not be able to ping the DHCP server or any other local machines? I simply cannot access anything, internal or external, with bridged ethernet. Shared ethernet works fine.

I've tried resetting TCP/IP according to this: http://kb.parallels.com/en/8978. No dice.

I've tried completely uninstalling and reinstalling Parallels tools and then all of PD6. Again, no dice.

Any ideas? Anything else I can test? I've spent the better part of a day on this!

Steve
 
Should there be two "Packet scheduler miniport" devices, one visible and one hidden, in Device Manager?

Steve
 
Hi, there are strange entries in system.log and they are frequent..:

Sep 27 15:45:53 unibody-mbp configd[13]: network configuration changed.
Sep 27 15:45:53 unibody-mbp configd[13]: en0: DHCP duplicate configured service
Sep 27 15:45:53 unibody-mbp configd[13]: en0: RTADV ignored
Sep 27 15:45:53 unibody-mbp configd[13]: network configuration changed.
Sep 27 15:45:53 Unibody-MBP configd[13]: setting hostname to "Unibody-MBP.local"
Sep 27 15:45:54 Unibody-MBP UserEventAgent[203]: **** ERROR: SCNetworkSignatureCopyActiveIdentifiers returned nil
Sep 27 15:45:59 Unibody-MBP configd[13]: network configuration changed.
Sep 27 15:46:07: --- last message repeated 1 time ---

--

Parallels on each this event, if it is Bridged to Default adapter, tries to redect which adapter is default now and finds that there is no adapter.

I'd suggest to
1) Choose "Ethernet 1" or "Built-In Adapter" instead of Default to bridge to

2) Try to find the reason why there are so many "network configuration changed." events in the system.log
 
Of course that's what I do. I always point directly to Ethernet instead of default adapter because the default adapter is Airport and I use that for the Mac side.

I'll have to look at the console, but what could these errors mean?

Steve
 
Those errors could just be me trying to fix this by changing settings.

And why does it take 45 minutes to see a post here?

Steve
 
Unbelievable! I rebooted my machine, which I know I've tried before, and it started working again. Weird.

Steve
 
I am having the same, or very similar problem under PD5, but not under all conditions.

Specifically, on my home network with a WRT54GL (running DD-WRT), I have
no problems operating my Win XP (sp3) guest in a bridged configuration,
but at work (Cisco 29xx workgroup switch for 802.3, and a recent
Cisco "N" AP (professional grade, don't know model#) for 802.11) this doesn't
want to work:

The "Parallels Ethernet Adapter" under XP actually OBTAINS A VALID IP
LEASE VIA DHCP - but seems to not receive back any traffic - can't ping
the default gateway at all. Note that the (LAN/802.3) switch permits
multiple MAC devices per port (I have a hub hanging off of it with other devices).

I've used tshark (CLI version of wireshark) under MacOS to sniff what is flying in/out
of the en0 (802.3) interface:

$ sudo tshark -n -i en0 ether host 00:1c:42:a1:fb:12
Capturing on en0
0.000000 10.254.249.165 -> 255.255.255.255 DHCP DHCP Request - Transaction ID 0x766e3745
2.624947 00:1c:42:a1:fb:12 -> ff:ff:ff:ff:ff:ff ARP Who has 10.254.249.1? Tell 10.254.249.165
14.611239 00:1c:42:a1:fb:12 -> ff:ff:ff:ff:ff:ff ARP Who has 10.254.249.1? Tell 10.254.249.165
29.284938 10.254.249.165 -> 255.255.255.255 DHCP DHCP Request - Transaction ID 0x452e2ad8
31.821288 00:1c:42:a1:fb:12 -> ff:ff:ff:ff:ff:ff ARP Who has 10.254.249.1? Tell 10.254.249.165
42.875226 00:1c:42:a1:fb:12 -> ff:ff:ff:ff:ff:ff ARP Who has 10.254.249.1? Tell 10.254.249.165

It shows one-way traffic (outbound only) - I can't see the DHCP replies coming back, but
given that DHCP is completed quickly, that traffic is definitely making it back in.

On the XP guest, I can see the entire DHCP exchange - but also only see the one-way ARP requests for the default gateway.

What might be going on here?

I have multiple VPN adapters (Cisco VPN, Hamachi, OpenVPN/Tap-Win32) on the XP guest - they FULLY work in shared mode (or bridged, at home), disabling them has no effect.
 
I'm in exactly this situation too

I have this exact problem also. But i have a little more additional info. When i bridge to the ethernet card. I cannot ping anything or get an ip from the dns in the guest win xp. I can ping and connect fine from the mac os. If i bridge to the airport card. then everything works fine in the guest os. No other changes. This Has to be a bug in parallels. I'm using the last version of parallels 5. And this problem started happening immediately after i downloaded 599993. whatever version i was running before this didn' t have this problem.
 
Bridged ethernet 8-(

Im a network guy myself and in order for me to use my tftp server to upgrade IOS images I need to be able to ping from the bridged ethernet to a specific location whether it be a router or what ever im directly connected to. Ive tried everything I could think of from firewalls to mac network setting and still no success. I hope this is a bug and maybe parallels could come out with some update to fix this issue ASAP. Till then has anyone came across any success pinging from the bridged ethernet? Whats the purpose of the option if it does not work properly.
 
BrelandR said:
Im a network guy myself and in order for me to use my tftp server to upgrade IOS images I need to be able to ping from the bridged ethernet to a specific location whether it be a router or what ever im directly connected to. Ive tried everything I could think of from firewalls to mac network setting and still no success. I hope this is a bug and maybe parallels could come out with some update to fix this issue ASAP. Till then has anyone came across any success pinging from the bridged ethernet? Whats the purpose of the option if it does not work properly.
Click to expand...

Bridged networking may not work in some network environments. Some routers/switches may be configured to forbid more then one MAC-address on the port. Actually it is the main requirement for bridged to work.

If you want me to take a look at this, click Start -> Run and execute cmd.exe. Type in the appeared window command "ipconfig /all" (make sure that output fits to the window). Then, without closing the window, click Parallels Destkop menu "Help"->"Report a Problem", send a problem report (it is not necessary to provide description) and post here the report ID.
 
my ethernet cable stop work too, here at my work I use wifi for www and ethernet for local servers access(parallels on winxp just need to use eth cable).
It takes a long time to get ip and then when winxp get it, just send packets cannot receive any.

kind of frustrating I'm accessing servers over vpn client and it's very laggy
 
You must log in or register to reply here.
Back
Top