Problem with deployment in a physical machine / connection timeout

[Quazz]

It’s possible a firewall is blocking the port, which would cause it to timeout.

Might also want to doublecheck Storage Node settings

[Andre]

@Quazz
my firewall is disabled in my FOG Server. And my storage node, i haven’t touched anything. I don’t know if I must do anything in Storage Management in FOG Web UI

[george1421]

@Andre OK lets do a bit of detective work.

I want you to schedule another capture or deploy (doesn’t matter) but before you hit the button to schedule the task, make sure you check the debug option. Then pxe boot the target computer. This feature will drop the pxe booting computer to a command prompt after a few pages of text. This is the command prompt of the FOS engine (customized linux OS for capture and deploying on the target computer).

Once at the command prompt see if you can ping the fog server.
If that works then we can try to nfs mount the fog server.

[Andre]

@george1421 Yes I can ping my server

[george1421]

@Andre just for clarity you get a ping response from the FOS engine (pxe booted hardware)?

See if the FOS engine has the showmount command. In this case you want showmount -e <fog_server_ip>

[Andre]

@george1421 Yes I did my ping from FOS Engine
For the command showmount, it says that “command not found”

[george1421]

@Andre OK from FOS do this

mkdir /mnt
mount -t nfs <fog_server_ip>:/images/dev /mnt
ls -la /mnt

The idea is to connect from the FOS back to the fog server using nfs.

[Andre]

@george1421 I tried what you told me to do. But when I do the command for mounting, I have the same error “Connection Timeout” .

[Andre]

@george1421 But I don’t understand why in virtual it works and in physical it doesn’t work. I can deploy or capture virtual machines without problem, and physical machines I can’t.

[Tom Elliott]

This sounds an aweful lot like the nic of the fogserver is setup in Nat mode vs. bridged mode. I say this because of the “it works in virtual systems, but not on physical machines.”

[Andre]

@Tom-Elliott My Fogserver is setup in bridged mode

[Andre]

Nobody has a solution for my problem ?

[george1421]

@Andre There has to be something we are not seeing here.

Lets just recap where we are here.

1. You have a new fog server that is running FOG 1.3.5 that is a physical server (not really relevant that its physical)
2. Your fog server’s IP address is 192.168.230.10
3. Your target computer IS talking to the FOG server because it can check in, plus FOS does load and is running.
4. NFS is sharing the proper directories.
5. NFS is timing out in that it can’t connect.

Possible issues.

1. The firewall is running on the FOG server blocking nfs mounting.
2. selinux is not set to permissive on the FOG server.
3. There is some kind of screening router between the target computer and FOG server
4. Not all of the nfs services are running on the FOG server
5. Since there is a second fog server in the environment maybe it is playing a part in the target computer’s booting process that is unknown to us.

Since the target computer can’t seem to mount the nfs share on the physical fog server and you have a second virtual fog server, I would try to mount the nfs shares on the phy fog server from the vm fog server to confirm that nfs is working on the phy fog server.

from the vm fog server

mount -t nfs 192.168.230.10:/images /mnt
ls -la /mnt
umount /mnt

The above command should mount the images share on the phy fog server, list the contents of the directory and then unmount the images directory. When you run this command I would expect to see something like this:

drwxrwxrwx  5 fog  root 4096 Apr  4 06:37 .
drwxr-xr-x 25 root root 4096 Apr 15 21:35 ..
drwxrwxrwx  3 fog  root 4096 Apr  4 06:37 dev
-rwxrwxrwx  1 fog  root    0 Sep 24  2016 .mntcheck
drwxrwxrwx  2 fog  root 4096 Sep 24  2016 postdownloadscripts

Not directly related to your issue, i would update your FOG server to the latest RC build of 1.4.0 to test to see if 1.3.5 is the issue. Only do this on your phy fog server.
https://wiki.fogproject.org/wiki/index.php/Upgrade_to_trunk

[Andre]

@george1421 I just have one server FOG and it is virtual. I don’t have a Physical server FOG.

[Andre]

@george1421 I tried to mount what do you said, in a Virtual machine, on which I have already capture his image. But when I execute the command “mount -t nfs 192.168.230.10:/images /mnt”, in FOS engine. It say “connection refused”.

[Andre]

@george1421 I tried with these commands in FOS Engine, with my virtual machine :
mkdir /images
mkdir /images/dev
mount -o nolock,proto=tcp,rsize=32768,intr,noatime 192.168.230.10:/images /images
mount -o nolock,proto=tcp,rsize=32768,intr,noatime 192.168.230.10:/images/dev/ /images/dev

And it works, I can mount

[Andre]

@george1421 And when I tried to do the same thing in my physical machine, it doesn’t work

[george1421]

@Andre This is not logical.

back on the fog server can you execute this command:

cat /etc/exports

You should get something that looks like this

/images *(ro,sync,no_wdelay,no_subtree_check,insecure_locks,no_root_squash,insecure,fsid=0)
/images/dev *(rw,async,no_wdelay,no_subtree_check,no_root_squash,insecure,fsid=1)

The key to look for is the space, star ( * ) right after the share name. This restricts nfs access to specific hosts. The start means everyone.

[Andre]

@georgI have the same lines as you

[george1421]

@Andre Well then…

On the fog server can you run this command

rpcinfo -p
You will get an output like this

   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
    100003    2   tcp   2049  nfs
    100003    3   tcp   2049  nfs
    100003    4   tcp   2049  nfs
    100227    2   tcp   2049
    100227    3   tcp   2049
    100003    2   udp   2049  nfs
    100003    3   udp   2049  nfs
    100003    4   udp   2049  nfs
    100227    2   udp   2049
    100227    3   udp   2049
    100021    1   udp  44805  nlockmgr
    100021    3   udp  44805  nlockmgr
    100021    4   udp  44805  nlockmgr
    100021    1   tcp  39255  nlockmgr
    100021    3   tcp  39255  nlockmgr
    100021    4   tcp  39255  nlockmgr
    100005    1   udp  37435  mountd
    100005    1   tcp  45499  mountd
    100005    2   udp  40066  mountd
    100005    2   tcp  33852  mountd
    100005    3   udp  59675  mountd
    100005    3   tcp  33362  mountd

The keys are to ensure that you have at least one portmapper and nfs and nfslockmgr entry.

And then please run this command
sudo ipables -L
The output should look like this:

Chain INPUT (policy ACCEPT)
target     prot opt source               destination         

Chain FORWARD (policy ACCEPT)
target     prot opt source               destination         

Chain OUTPUT (policy ACCEPT)
target     prot opt source               destination     

The last check is to ensure selinux has been set to permissive. The key is current mode == permissive.
sestatus
output:

SELinux status:                 enabled
SELinuxfs mount:                /selinux
Current mode:                   permissive
Mode from config file:          permissive
Policy version:                 21
Policy from config file:        targeted