Re: Connection of a qemu guest to the 'net.

[Berto Furth]

Just a correction. Make sure that 

#!bin/sh

 is the first line in the qemu-ifup file. I got the "#" and the "!" in the 
wrong order in my email below. I'm sure it's already correct in your setup 
anyway.

Thanks Peter.

On Fri, 19 Mar 2021, at 07:59, Berto Furth wrote:
> Hi Peter,
> 
> So just to get to your last point first again
> 
> "network script /etc/qemu-ifup failed with status 256"
> 
> might mean that qemu couldn't find the script or the file isn't executable. 
> 
> Can you try again making sure that the /etc/qemu-ifup script exists, 
> that it's first line is "!#/bin/sh" and that it's executable (chmod a+x 
> /etc/sh")
> 
> Alternately, can you put the script in the directory you're working in 
> and explicitly specify the location of the script on the command line 
> as follows?
> 
> sudo qemu-system-i386 \
>   -drive file=/dev/KingstonCF,media=disk,format=raw \
>   -vga std \
>   -boot order=c \
>   -netdev type=tap,id=peternet,script="./qemu-ifup",downscript=no \
>   -global driver=ne2k_pci,property=netdev,value="peternet"
> 
> If that still doesn't work then perhaps you could manually try to run 
> the script with a temporarily created tap interface as follows. 
> 
> 
> sudo ip tuntap add mode tap name tap8
> sudo ip link set tap8 up
> sudo /etc/qemu-ifup tap8
> 
> If there are no errors see if the tap interface is actually connected 
> to the bridge with either the "ip" or "brctl" commands. On my system I 
> get
> 
> # sudo ip link show master br0
> 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq master br0 
> state UP mode DEFAULT group default qlen 1000
>     link/ether dc:xx:xx:xx:xx:55 brd ff:ff:ff:ff:ff:ff
> 177: tap8: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc 
> pfifo_fast master br0 state DOWN mode DEFAULT group default qlen 1000
>     link/ether 0a:xx:xx:xx:xx:46 brd ff:ff:ff:ff:ff:ff
> # sudo brctl show br0
> bridge name     bridge id               STP enabled     interfaces
> br0             8000.020000000002       no              eth0
>                                                         tap8
> 
> Then you can delete the temporary tap interface
> 
> # sudo ip tuntap del mode tap tap8
> 
> > > If one of the more experienced people who lurk on these forums could 
> > > comment on it just in case it's pure rubbish I'd be more comfortable 
> > > making an attempt at putting it somewhere permanent.
> > 
> > Organizing into sections with headings will help.  Almost any addition 
> > to https://en.wikibooks.org/wiki/QEMU/Devices/Network will be 
> > improvement.  Content with an error or two is better than no content. (!) 
> > In a wiki, errors can be corrected.  =8~)
> 
> Yes, good point. As I said I'll look into it and let you know if I do 
> it so you can review it!
>  
> > > My view is that Interfaces are at the Data Link and optionally at the 
> > > Network Layer as well. For example an eth0 interface has an Ethernet 
> > > MAC address (Data Link Layer) but if you want you can also give it an 
> > > IP and IPv6 address (Network Layer). 
> > 
> > OK.  Acknowledged.
> > 
> > > You have to create the bridge before you connect the interface to it 
> > > in the same way that you have to put your network switch on the desk 
> > > before you can plug your Ethernet cables into it!!
> > 
> > Interesting analogy.  Seems strictly correct although a reader might 
> > misinterpret. The switch and cables are both at the physical layer 
> > whereas you refer to them as analogues rather than OSI components.
> 
> Yeah, I reflected on that analogy last night and I thought it was a bit 
> flawed in that respect too. 
>  
> > > Can you clarify that? Do you mean you have no bridge on the Shorewall? 
> > > Does it have a switch in it? A "switch" and a "bridge" are 
> > > functionally equivalent. 
> > 
> > This is my understanding.  Shorewall is not a daemon. It sets up 
> > routing by iptables.  
> > 
> > Any command such as "ip link show" never reported a bridge until br0 
> > was created for qemu.  To my understanding, Shorewall does not require 
> > a  bridge or switch but either can be present with Shorewall.  Several 
> > documents are available.  One here.
> > 
> > https://shorewall.org/SimpleBridge.html
> > 
> > > >     inet6 fe80::6c05:fdff:fedf:e2eb/64 scope link
> > > >        valid_lft forever preferred_lft forever
> > 
> > > This is the IPv6 Link Local address. Every interface with IPv6 
> > > enabled will automagically assign itself an IPv6 address based on it's 
> > > eMAC address.
> > 
> > Thanks.  I'm learning about IPv6 along with qemu.
> > 
> 
> IPv6 is pretty awesome, mainly because it lets you access servers on 
> your network from the wider internet without having to do weird things 
> with IPv4 NAT. Does your ISP provide you with IPv6 connectivity? If so 
> then it's definitely worth setting up. I have some servers on my 
> network that I can access remotely via IPv6 and it's so easy and 
> convenient. The other great thing about it is that because the address 
> space is so massively huge, you never get entities on the internet 
> randomly "scanning" your hosts because it would take 20 septillion 
> years for them to find them due to there being so many addresses 
> available. My IPv4 host (my router) gets scanned all day every day.
> 
> > This is my current qemu command.
> > 
> > sudo qemu-system-i386 \
> >  -drive file=/dev/KingstonCF,media=disk,format=raw \
> >  -vga std \
> >  -boot order=c \
> >  -nic tap,model=ne2k_pci,br=br0
> > 
> > As described previously, br0 is created here when the qemu host boots. 
> > br0 continues to exist when qemu is executed.
> > 
> 
> Do you need the "br=br0" parameter there? Anyway, see the top of the 
> post for my suggestion.
> 
> > With the /etc/qemu-ifup provided by Berto, no guest system appears 
> > and this is reported to the terminal. 
> > 
> > qemu-system-i386: network script /etc/qemu-ifup failed with status 256
> > 
> > With the /etc/qemu-ifup provided by qemu in Debian 10, a guest system 
> > appears and this is reported to the terminal. 
> > 
> > W: /etc/qemu-ifup: no bridge for guest interface found
> > 
> 
> Is there one included in qemu? Can you send me a copy? Does it need to 
> be edited to include the bridge interface name?
> 
> Anyway, we're getting there! 
> 
> > Conclusion
> > br0 is present but qemu fails to recognize it.  The /etc/qemu-ifup in 
> > Debian avoids catastrophic failure.  
> > 
> > Why does qemu fail to recognize br0?  My best guess is that an error 
> > is present in my qemu command above.  
> > 
> > Ideas?
> > 
> > Thx,                         ... P.
> > 
> > -- 
> > cell: +1 236 464 1479            Bcc: peter at easthope. ca
> > VoIP: +1 604 670 0140
> > 
> > 
> >
> 
>