Network Timeout for NFS mount using Bridged Network Setup

Hi,

I installed Ubuntu 13 Server (64bit) edition in a Parallels Desktop 9 VM. My Host System is OS-X Mountain Lion. For testing purpose I setup the VM with 8GB Storage, 2GB RAM, 1CPU and for the Network Support I choose the Realtek (Default Adapter). I installed the missing packages for NFS and setup Network for static IP.

Now when I check if I can see the NFS shares from the the VM it works fine:

Code:

frontrow@hubble:~$ showmount -e 192.168.12.23
Export list for 192.168.12.23:
/volume1/TEST           192.168.12.*
/volume1/frontrow       192.168.12.*
/volume1/plex_downloads 192.168.12.*
frontrow@hubble:~$

I upgraded my Synology NAS to DSM 4.3 which imporved the support for nfs4 so a quick check on the port side looked also good:

Code:

frontrow@hubble:~$ rpcinfo -p 192.168.12.23
   program vers proto   port  service
    100000    4   tcp    111  portmapper
    100000    3   tcp    111  portmapper
    100000    2   tcp    111  portmapper
    100000    4   udp    111  portmapper
    100000    3   udp    111  portmapper
    100000    2   udp    111  portmapper
    100005    1   udp    892  mountd
    100005    1   tcp    892  mountd
    100005    2   udp    892  mountd
    100005    2   tcp    892  mountd
    100005    3   udp    892  mountd
    100005    3   tcp    892  mountd
    100021    1   udp  54322  nlockmgr
    100021    3   udp  54322  nlockmgr
    100021    4   udp  54322  nlockmgr
    100021    1   tcp  60625  nlockmgr
    100021    3   tcp  60625  nlockmgr
    100021    4   tcp  60625  nlockmgr
    100003    2   tcp   2049  nfs
    100003    3   tcp   2049  nfs
    100003    4   tcp   2049  nfs
    100003    2   udp   2049  nfs
    100003    3   udp   2049  nfs
    100024    1   udp  52085  status
    100024    1   tcp  45308  status
frontrow@hubble:~$

And just to avoid any questions I made a quick nmap check:


frontrow@hubble:~$ sudo nmap -PO 192.168.12.23
[sudo] password for frontrow:

Code:

Starting Nmap 6.00 ( http://nmap.org ) at 2013-09-08 14:38 CEST
Nmap scan report for Asterix (192.168.12.23)
Host is up (0.00022s latency).
Not shown: 986 closed ports
PORT     STATE SERVICE
22/tcp   open  ssh
80/tcp   open  http
111/tcp  open  rpcbind
139/tcp  open  netbios-ssn
161/tcp  open  snmp
445/tcp  open  microsoft-ds
515/tcp  open  printer
548/tcp  open  afp
631/tcp  open  ipp
873/tcp  open  rsync
2049/tcp open  nfs
3260/tcp open  iscsi
5000/tcp open  upnp
5432/tcp open  postgresql
MAC Address: 00:11:32:18:A8:D7 (Synology Incorporated)

Now If I finally try a manual connect to the NAS I get a timeout:

Code:

sudo mount -vvv -t nfs4 192.168.12.23:/volume1/TEST /mnt/asterix/
mount: fstab path: "/etc/fstab"
mount: mtab path:  "/etc/mtab"
mount: lock path:  "/etc/mtab~"
mount: temp path:  "/etc/mtab.tmp"
mount: UID:        0
mount: eUID:       0
mount: spec:  "192.168.12.23:/volume1/TEST"
mount: node:  "/mnt/asterix/"
mount: types: "nfs4"
mount: opts:  "(null)"
mount: external mount: argv[0] = "/sbin/mount.nfs4"
mount: external mount: argv[1] = "192.168.12.23:/volume1/TEST"
mount: external mount: argv[2] = "/mnt/asterix/"
mount: external mount: argv[3] = "-v"
mount: external mount: argv[4] = "-o"
mount: external mount: argv[5] = "rw"
mount.nfs4: timeout set for Sat Sep  7 16:42:14 2013
mount.nfs4: trying text-based options 'addr=192.168.12.23,clientaddr=192.168.12.90'
mount.nfs4: mount(2): Connection timed out
mount.nfs4: Connection timed out
frontrow@hubble:~$

After a couple day's frustrating search and posting in Ubuntu foren I played around with the Network settings and to my surprise I got to work when I switched to:

NIC TYPE: VIRTIO NETWORK ADAPTER
TYPE: SHARED

Using DHCP I get a network Address of 192.168.13.151 (.13 Subnet) and it connects in 1-2 seconds. Now of course the issue is that I cannot connect to this VM from external Hosts on my Network which was the idea in the first place.

Can anyone explain to my what is the difference between these 2 settings? Why does that not work on the TYPE Default? How can I achieve to have the VM connect to the NAS and be visible in the Network?

Thanks for your help

 

Hm. Interesting.
Could you please shutdown the VM, go to the Configuration of VM -> Hardware -> Boot Order and paste to the "Boot Flags" line

devices.net.virtio.tx_offload=0

then switch to the Bridged to default and try again

==
PS: Sorry! I've just edited the message. At the first I've messed with flag, now it is ok.

 

Hi, thanks alot for the swift answer, much appreciated. I tried that and I still get the same timeout. I had a look into the Parallels log and the only odd think I can't really place but happen around the time I execute the mount command is the following:

Code:

vdisk:42658:8507/ Error: device sync request was not acked

What is that boot flag supposed to do?

 

No, it is not related and generally it is ok.

Some Network cards allow to offload some routine functions (like checksumming, segmentation data to smaller pieces etc) from CPU to network hardware. The flag "devices.net.virtio.tx_offload=0" says that card will not do Transmit-specific offloads and offloading must be done by guest itself.

We will try to reproduce the issue locally, but meantime there is one more step to try:
Shutdown the VM and change in the VM Configuration -> Hardware -> Network the type of NIC Type from "Virtio" to "Realtek 8029". Then check whether it helps.

Thanks!

 

Hi Eric,

thanks again for your swift answer.
In fact I have tried every variation of Network NIC TYPE already and the only variation it works if I use SHARED. That way however I get an isolated subnet only accessible from the Host the VM is running on.

I find it very odd that over a bridged network it doesn't work. From my understanding on NFS and Networking I get the feeling somewhere is a packet not coming back to confirm the mount is complete or authorised or something.
Again if I choose the same in SHARED mode it works in like 1-2 seconds.

Thanks again for your help

 

Hm. That really strange... Realtek is the simpliest network card and doesn't do anything complicated at all. Isn't your mac connected to network with WiFi? It is the only possible reason I can imagine that could have affection. Assuming, that networking in Ubuntu is correctly configured..

just in case, I'd want to see networking configuraion. The simlpies way is
1. Execute at Mac /Application/Utilities/Terminal and run in it "ifconfig -a",
2. Execute in Ubuntu "ip a l" (make sure that output fits to visible screen), click menu of Parallels Desktop Help-> Report a problem. Paste to the description of report the output of ifconfig from Mac and send the problem report.

Then post here the number of report.

 

One more thing... While "Shared Networking" the IP address of Mac is used. May be there is somewhere limitation for IP-address in the NAS?

 

Hi Eric,

thanks again for your swift answer.
I actually have WIFI disabled. This is a MiniMac that is wired up and for the purpose I need all the bandwidth I can get. So Wifi is not used. Shouldn't that be trivial?
I mean Bridged Network means it will use the Macs local Network settings and Shared means it will use the Virtual Parallels Network broadcasting their 10.x.x.x Network (or whatever you set it too).

How would that play a role?

PS: The Ticket ID is 28513504

 

The thing is that there somewhere in NAS configuration could be configured that only this Mac can connect. This would explain everything..

PS: I've checked the network configuration in the report, all seems to be ok

 

I thought about that as well, but how would you explain that if I choose any type of SHARED network it works? I totally agree with you that there is something blocking somehow the connection... but that is exactly what I can't figure out..

 

It is because in Shared-mode NAS sees the same IP-address of Mac (the 10.211.55.x is behind NAT). For bridged - VM has own IP address

 

Is there any white/black list of MAC addresses on the NAS? Maybe the NAS is blocking the connection because the MAC address of the virtual adapter is not explicitly allowed.

 

There are such function in the NAS but I have not enabled them. Pretty much all security features are switched off.

 

I found the issue - kinda silly but a good one. I got a stack of cisco switches and they all look the same, however one of those switches is a managed one. You can imagine what happened ? - Yea so I grabbed the wrong switch (using 3x 8port ones) and it must have dropped the the port connection. I removed it and voila it all works as it should...

Sorry about that and many thanks for your help, the last question of yours triggered me to check the Routers / Switches.

Case Closed!