Setup Issues and Error PXE E53
-
Hi guys,
I’m having some issues with getting Fog setup properly for my in environment. Currently, we are running Fog 0.32 on Ubuntu 10.04 server, but have decided to upgrade to Fog 1.2.0 on Ubuntu 14.04 server.
Things I should probably note about our infrastructure. We have no control over the DHCP server beyond we can please a ticket to have static IPs assigned. We also run Fog inside of a VM which runs on an ESXi host (for the sake of testing I running Fog inside of a VM on my iMac using the bridged ethernet network connection). We do have our own subnet and I have to make sure that we broadcast only to our subnet.
I have looked at a few different setups, but nothing seems to be working.
From the Fog Wiki:
Fog on Ubuntu 14.04
Using FOG with no DHCP ServerFrom External Sources:
FOG – Installing and Troubleshooting FOG 1.2.0 on Ubuntu 14.04 -> This guide was awesome with the details, but still didn’t work in the end. FOG – Installing and Troubleshooting FOG 1.2.0 on Ubuntu 14.04
FOG Update – Part 4I should also note as far as I can tell the only difference between my setup and older one(which my boss created) is that 10.04 uses LTSP and my 14.04 doesn’t.
My “lstp.conf” file:
port=0 log-dhcp tftp-root=/tftpboot dhcp-boot=undionly.kpxe,130.18.192.41,130.18.192.41 dhcp-no-override pxe-prompt="Press F8 for boot menu", 3 pxe-service=X86PC, "Boot from network", undionly pxe-service=X86PC, "Boot from local hard disk", 0 dhcp-range=130.18.192.41,proxy,255.255.254.0The static IP for the VM was set properly and is still communicating with the outside world properly. I’m not sure if there is an issue with the dnsmasq setup, tftp, or something else that I overlooked.
-
@Dalton-Childers said:
Hi guys,
I’m having some issues with getting Fog setup properly for my in environment. Currently, we are running Fog 0.32 on Ubuntu 10.04 server, but have decided to upgrade to Fog 1.2.0 on Ubuntu 14.04 server.
Things I should probably note about our infrastructure. We have no control over the DHCP server beyond we can please a ticket to have static IPs assigned. We also run Fog inside of a VM which runs on an ESXi host (for the sake of testing I running Fog inside of a VM on my iMac using the bridged ethernet network connection). We do have our own subnet and I have to make sure that we broadcast only to our subnet.
I have looked at a few different setups, but nothing seems to be working.
From the Fog Wiki:
Fog on Ubuntu 14.04
Using FOG with no DHCP ServerFrom External Sources:
FOG – Installing and Troubleshooting FOG 1.2.0 on Ubuntu 14.04 -> This guide was awesome with the details, but still didn’t work in the end. FOG – Installing and Troubleshooting FOG 1.2.0 on Ubuntu 14.04
FOG Update – Part 4I should also note as far as I can tell the only difference between my setup and older one(which my boss created) is that 10.04 uses LTSP and my 14.04 doesn’t.
My “lstp.conf” file:
port=0 log-dhcp tftp-root=/tftpboot dhcp-boot=undionly.kpxe,130.18.192.41,130.18.192.41 dhcp-no-override pxe-prompt="Press F8 for boot menu", 3 pxe-service=X86PC, "Boot from network", undionly pxe-service=X86PC, "Boot from local hard disk", 0 dhcp-range=130.18.192.41,proxy,255.255.254.0The static IP for the VM was set properly and is still communicating with the outside world properly. I’m not sure if there is an issue with the dnsmasq setup, tftp, or something else that I overlooked.
Please try changing your ltsp.conf to:
port=0 log-dhcp tftp-root=/tftpboot dhcp-boot=undionly.kpxe,130.18.192.41,130.18.192.41 dhcp-no-override pxe-prompt="Press F8 for boot menu", 3 pxe-service=X86PC, "Boot from network", undionly.kpxe pxe-service=X86PC, "Boot from local hard disk", 0 dhcp-range=130.18.192.41,proxy,255.255.254.0Also make a link to undionly.kpxe.0, or change the lines in the ltsp.conf file that have undionly.kpxe to undionly if you want to use unidonly.0
My guess, you forgot this step. If a link seems to no twork try a copy.
For example if you want to be using the undionly.kpxe as the primary boot file:
cp /tftpboot/undionly.kpxe /tftpboot/undionly.0(or undionly.kpxe.0 if you left the file names as such). -
What’s up with the last line? proxy,255.255.254.0
Seems like a strange subnet to have - and I’m not even sure if it accepts anything after the proxy argument or not… something to look into.Also I disagree strongly with some of the methods used in that 3rd link.
The 4th link references a article in our WiKi - it’s a really rough article…![😕](https://forums.fogproject.org/assets/plugins/nodebb-plugin-emoji/emoji/android/1f615.png?v=i84e1j9s46s ":-\")
However, I am in the process of re-writing it. It’s a work in progress. -
@Tom-Elliott I decided to rebuild the VM from scratch and I setup the new ltsp.conf file to reflect your suggestions. Originally, I did a linked the files, but took your suggestion again and copied instead. This did not fix the issue as I am still receiving the PXE E53 error.
cp /tftpboot/undionly.kpxe /tftpboot/undionly.0I also tried, but neither seemed to work.
cp /tftpboot/undionly.kpxe /tftpboot/undionly.kpxe.0I tried to get the unidonly.kpxe file from a windows machine on the network using:
tftp –i 130.18.192.41 get undionly.kpxeBut I get back the following error:
Timeout Occurred Connection Request FailedI did a fresh setup and here’s the options I chose:
Type 2 and press Enter for Ubuntu installation: 2 Type N and press Enter for Normal installation: N Supply IP Address, it SHOULD be the static IP address you set up earlier: 130.18.192.41 Setup DHCP Server: N Setup DNS: N Type N and press Enter to leave the default Network Card the same: N Type N to disable DHCP Service: N Type N to not install Additional Languages: N Type Y and press Enter: YThere’s a suggestion in documentation to say Yes to the DHCP and DNS server IP, but I chose no based on the my bosses’ notes from the old setup. The setup associated with the previous post I answered yes to those questions and wondered if that was the issue.
-
@Wayne-Workman said:
What’s up with the last line? proxy,255.255.254.0
Seems like a strange subnet to have - and I’m not even sure if it accepts anything after the proxy argument or not… something to look into.Also I disagree strongly with some of the methods used in that 3rd link.
The 4th link references a article in our WiKi - it’s a really rough article…
However, I am in the process of re-writing it. It’s a work in progress.@Wayne-Workman Yeah, the subnet is kind of strange, but we take what they give us.
-
If apt-get isn’t working, try commenting (adding a # symbol to the start of a comment) the port=0 line, with
#port=0and restart dnsmasq. -
@Dalton-Childers You’ve ensured that firewall is disabled on the server?
Please help us build the FOG community with everyone involved. It's not just about coding - way more we need people to test things, update documentation and most importantly work on uniting the community of people enjoying and working on FOG! Get in contact with me (chat bubble in the top right corner) if you want to join in.
Web GUI issue? Please check apache error (debian/ubuntu: /var/log/apache2/error.log, centos/fedora/rhel: /var/log/httpd/error_log) and php-fpm log (/var/log/php*-fpm.log)
Please support FOG if you like it: https://wiki.fogproject.org/wiki/index.php/Support_FOG
-
@Tom-Elliott said:
@Dalton-Childers You’ve ensured that firewall is disabled on the server?
Well if he can’t get the plane-jane undionly.kkpxe using TFTP - then there are other issues.
I was thinking either Firewall or TFTP is not running right or something - I don’t think this is a dnsmasq problem.
Can you check the status of TFTP and then restart it and then check the status again with these commands?
service tftpd-hpa status service tftpd-hpa restart service tftpd-hpa statusThen after you’ve restarted TFTP, please go to another computer and try to get the undionly.kpxe file manually via TFTP again.
-
@Wayne-Workman said:
@Tom-Elliott said:
@Dalton-Childers You’ve ensured that firewall is disabled on the server?
Well if he can’t get the plane-jane undionly.kkpxe using TFTP - then there are other issues.
I was thinking either Firewall or TFTP is not running right or something - I don’t think this is a dnsmasq problem.
Can you check the status of TFTP and then restart it and then check the status again with these commands?
service tftpd-hpa status service tftpd-hpa restart service tftpd-hpa statusThen after you’ve restarted TFTP, please go to another computer and try to get the undionly.kpxe file manually via TFTP again.
Okay, so I restart the tftpd-hpa service.
service tftpd-hpa status tftpd-hpa start/running, process 1104 service tftpd-hpa restart tftpd-hpa stop/waiting tftpd-hpa start/running, process 3039 service tftpd-hpa status tftpd-hpa start/running, process 3039I tried to get the unidonly.kpxe file again from a windows machine on the network using:
tftp –i 130.18.192.41 get undionly.kpxeBut I still get back the following error:
Timeout Occurred Connection Request Failedtftpd-hpa file
# /etc/default/tftpd-hpa # FOG Modified version TFTP_USERNAME="root" TFTP_DIRECTORY="/tftpboot" TFTP_ADDRESS="0.0.0.0:69" TFTP_OPTIONS="--secure"Which I am wondering if the issue is the root username. Should the username be fog?
-
@Tom-Elliott said:
@Dalton-Childers You’ve ensured that firewall is disabled on the server?
I disabled the firewall all together.
Also, I managed to get apt-get working again by following your suggestion to comment out “port=0”.
Steps:
Comment out “port=0”
rm /var/lib/apt/lists/* -vf
apt-get update -
@Dalton-Childers said:
@Wayne-Workman said:
@Tom-Elliott said:
@Dalton-Childers You’ve ensured that firewall is disabled on the server?
Well if he can’t get the plane-jane undionly.kkpxe using TFTP - then there are other issues.
I was thinking either Firewall or TFTP is not running right or something - I don’t think this is a dnsmasq problem.
Can you check the status of TFTP and then restart it and then check the status again with these commands?
service tftpd-hpa status service tftpd-hpa restart service tftpd-hpa statusThen after you’ve restarted TFTP, please go to another computer and try to get the undionly.kpxe file manually via TFTP again.
Okay, so I restart the tftpd-hpa service.
service tftpd-hpa status tftpd-hpa start/running, process 1104 service tftpd-hpa restart tftpd-hpa stop/waiting tftpd-hpa start/running, process 3039 service tftpd-hpa status tftpd-hpa start/running, process 3039I tried to get the unidonly.kpxe file again from a windows machine on the network using:
tftp –i 130.18.192.41 get undionly.kpxeBut I still get back the following error:
Timeout Occurred Connection Request Failedtftpd-hpa file
# /etc/default/tftpd-hpa # FOG Modified version TFTP_USERNAME="root" TFTP_DIRECTORY="/tftpboot" TFTP_ADDRESS="0.0.0.0:69" TFTP_OPTIONS="--secure"Which I am wondering if the issue is the root username. Should the username be fog?
Mine :
# /etc/default/tftpd-hpa # FOG Modified version TFTP_USERNAME="root" TFTP_DIRECTORY="/tftpboot" TFTP_ADDRESS=":69" TFTP_OPTIONS="-s" -
@ch3i said:
@Dalton-Childers said:
@Wayne-Workman said:
@Tom-Elliott said:
@Dalton-Childers You’ve ensured that firewall is disabled on the server?
Well if he can’t get the plane-jane undionly.kkpxe using TFTP - then there are other issues.
I was thinking either Firewall or TFTP is not running right or something - I don’t think this is a dnsmasq problem.
Can you check the status of TFTP and then restart it and then check the status again with these commands?
service tftpd-hpa status service tftpd-hpa restart service tftpd-hpa statusThen after you’ve restarted TFTP, please go to another computer and try to get the undionly.kpxe file manually via TFTP again.
Okay, so I restart the tftpd-hpa service.
service tftpd-hpa status tftpd-hpa start/running, process 1104 service tftpd-hpa restart tftpd-hpa stop/waiting tftpd-hpa start/running, process 3039 service tftpd-hpa status tftpd-hpa start/running, process 3039I tried to get the unidonly.kpxe file again from a windows machine on the network using:
tftp –i 130.18.192.41 get undionly.kpxeBut I still get back the following error:
Timeout Occurred Connection Request Failedtftpd-hpa file
# /etc/default/tftpd-hpa # FOG Modified version TFTP_USERNAME="root" TFTP_DIRECTORY="/tftpboot" TFTP_ADDRESS="0.0.0.0:69" TFTP_OPTIONS="--secure"Which I am wondering if the issue is the root username. Should the username be fog?
Mine :
# /etc/default/tftpd-hpa # FOG Modified version TFTP_USERNAME="root" TFTP_DIRECTORY="/tftpboot" TFTP_ADDRESS=":69" TFTP_OPTIONS="-s"I updated my tftpd-hpa file to reflect you setup, but sadly it’s still not working.
-
@Dalton-Childers Try the TFTP test locally on the FOG server itself. Try to use both the actual IP of the server and then try the local loopback address 127.0.0.1
-
@Wayne-Workman said:
@Dalton-Childers Try the TFTP test locally on the FOG server itself. Try to use both the actual IP of the server and then try the local loopback address 127.0.0.1
Using the IP address of the server (130.18.192.41):
root@fog:~# tftp 130.18.192.41 tftp> get undionly.kpxe tftp> get undionly.0 Error code 1: File not foundDue to a lack of experience with TFTP, I am assuming the lack of error for undionly.kpxe means it found the file. I tried undionly.0 because I know the file is not present under /tftpboot/.
Using the IP address of the server (127.0.0.1):
root@fog:~# tftp 127.0.0.1 tftp> get undionly.kpxe tftp> get undionly.0 Error code 1: File not foundSame results as the server’s IP.
-
@Dalton-Childers So the simple test, you attempt to get the undionly.kpxe file, then you try to get the undionly.0 file. Because the undionly.kpxe file exists, you’re not seeing an error, but because the undionly.0 doesn’t exist, you’re seeing the error on the second command. This sound correct.
Now that you know that at the least the tftp server is operating locally. Now we need to see where it breaks down.
-
@Dalton-Childers @Tom-Elliott is right, and that was the direction I was heading with this…
Can you give us the output of this command first - it’ll let us know what’s going on with the sym-link.
ls -lahRt /tftpbootThen - We need lots and lots of details about your VM setup because I think that’s where the problem lies. What are you using? How is the network setup? Is it using NAT or is it bonded or NIC sharing or what?
Can you install traceroute and then run a traceroute against Google’s DNS?
sudo apt-get install traceroute traceroute 8.8.8.8and give us the output of that please? This will tell us if you are using NAT or not. It’d also help greatly if you’d tell us you Router Address - the address of the target host if you can find it on your router - and perhaps some screen shots of your network setup for the VM?
Basically - the more details the better.
-
@Tom-Elliott @Wayne-Workman said:
@Dalton-Childers @Tom-Elliott is right, and that was the direction I was heading with this…
Can you give us the output of this command first - it’ll let us know what’s going on with the sym-link.
ls -lahRt /tftpbootThen - We need lots and lots of details about your VM setup because I think that’s where the problem lies. What are you using? How is the network setup? Is it using NAT or is it bonded or NIC sharing or what?
Can you install traceroute and then run a traceroute against Google’s DNS?
sudo apt-get install traceroute traceroute 8.8.8.8and give us the output of that please? This will tell us if you are using NAT or not. It’d also help greatly if you’d tell us you Router Address - the address of the target host if you can find it on your router - and perhaps some screen shots of your network setup for the VM?
Basically - the more details the better.
First request is listed below in the screen shot. The results of:
ls -lahRt /tftpboot
Information about setup:
I am running the Ubuntu 14.04 VM on my iMac using VMware Fusion 8 Pro. I have set a static IP which is “130.18.192.41” (including image of my network interface config). Our subnet is “255.255.254.0” and the overall range is 130.18.192.1 to 130.18.193.255. The architecture of the network isn’t completely known due to the nature of the main ITS group. Also, we have to be careful not to allow our PXE traffic to appear outside of our subnet. We have our own DNS in house, but the DHCP server is control by the main ITS group which is why I assumed I would need to use dnsmasq.
I have included a picture for clarity, but the VM is bridged with the iMac’s ethernet port so it appears as an additional machine.
The IP in the image “130.18.193.107” is the IP that is assigned to my iMac.
Code box showing the output of ifconfig on my iMac:
lo0: flags=8049<UP,LOOPBACK,RUNNING,MULTICAST> mtu 16384 options=3<RXCSUM,TXCSUM> inet6 ::1 prefixlen 128 inet 127.0.0.1 netmask 0xff000000 inet6 fe80::1%lo0 prefixlen 64 scopeid 0x1 nd6 options=1<PERFORMNUD> gif0: flags=8010<POINTOPOINT,MULTICAST> mtu 1280 stf0: flags=0<> mtu 1280 en0: flags=8963<UP,BROADCAST,SMART,RUNNING,PROMISC,SIMPLEX,MULTICAST> mtu 1500 options=10b<RXCSUM,TXCSUM,VLAN_HWTAGGING,AV> ether 40:6c:8f:0f:ad:1a inet6 fe80::426c:8fff:fe0f:ad1a%en0 prefixlen 64 scopeid 0x4 inet 130.18.193.107 netmask 0xfffffe00 broadcast 130.18.193.255 nd6 options=1<PERFORMNUD> media: autoselect (1000baseT <full-duplex>) status: active en1: flags=8863<UP,BROADCAST,SMART,RUNNING,SIMPLEX,MULTICAST> mtu 1500 ether 7c:c3:a1:9f:df:6c inet6 fe80::7ec3:a1ff:fe9f:df6c%en1 prefixlen 64 scopeid 0x5 inet 10.0.1.2 netmask 0xffffff00 broadcast 10.0.1.255 nd6 options=1<PERFORMNUD> media: autoselect status: active fw0: flags=8822<BROADCAST,SMART,SIMPLEX,MULTICAST> mtu 4078 lladdr 3c:07:54:ff:fe:e6:02:ee media: autoselect <full-duplex> status: inactive en2: flags=8963<UP,BROADCAST,SMART,RUNNING,PROMISC,SIMPLEX,MULTICAST> mtu 1500 options=60<TSO4,TSO6> ether d2:00:1e:60:2e:e0 media: autoselect <full-duplex> status: inactive en3: flags=8963<UP,BROADCAST,SMART,RUNNING,PROMISC,SIMPLEX,MULTICAST> mtu 1500 options=60<TSO4,TSO6> ether d2:00:1e:60:2e:e1 media: autoselect <full-duplex> status: inactive p2p0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 2304 ether 0e:c3:a1:9f:df:6c media: autoselect status: active bridge0: flags=8822<BROADCAST,SMART,SIMPLEX,MULTICAST> mtu 1500 options=63<RXCSUM,TXCSUM,TSO4,TSO6> ether 42:6c:8f:f0:c0:00 Configuration: id 0:0:0:0:0:0 priority 0 hellotime 0 fwddelay 0 maxage 0 holdcnt 0 proto stp maxaddr 100 timeout 1200 root id 0:0:0:0:0:0 priority 0 ifcost 0 port 0 ipfilter disabled flags 0x2 member: en2 flags=3<LEARNING,DISCOVER> ifmaxaddr 0 port 7 priority 0 path cost 0 member: en3 flags=3<LEARNING,DISCOVER> ifmaxaddr 0 port 8 priority 0 path cost 0 media: <unknown type> status: inactive vmnet1: flags=8863<UP,BROADCAST,SMART,RUNNING,SIMPLEX,MULTICAST> mtu 1500 ether 00:50:56:c0:00:01 inet 192.168.192.1 netmask 0xffffff00 broadcast 192.168.192.255 vmnet8: flags=8863<UP,BROADCAST,SMART,RUNNING,SIMPLEX,MULTICAST> mtu 1500 ether 00:50:56:c0:00:08 inet 192.168.2.1 netmask 0xffffff00 broadcast 192.168.2.255The results of traceroute:
I don’t have access to the routers so I am flying blind on that end.
If there is anything extra that I should have added please let me know.
-
@Dalton-Childers At first glance of the results of ls -lahRt /tftpboot it would seem that your sym-links are not created, and that you have used copies instead. undionly.0 does not exist. (just informational at this point)
From your ifconfig output,
What is 10.0.1.2 ?
What is 192.168.2.1 ?And - try to create this symbolic link and then try to network boot again:
ln -s /tftpboot/undionly.kkpxe /tftpboot/undionly.0Here’s more info on symbolic links: http://www.cyberciti.biz/faq/unix-creating-symbolic-link-ln-command/
Please do not get discouraged. For many, Linux is a new adventure and the ‘path’ might not be as clear as it might be to an experienced MAC OS or Windows user - but to the experienced Linux user - it’s the same process as any other system… troubleshoot, test, troubleshoot, test - narrow it down, test, troubleshoot, test, narrow it down further, coffee, test, restroom break, test… it’s the same thing. If you know - it’s easy, if you don’t know, you figure it out. If you’re inexperienced - then you either try to figure it out and in the process learn a massive ton of Linux or you simply walk away because you refuse to give the effort…
As long as you keep the communication going here, we will continue to try to help. It’s the same as with anything - if there’s a will there’s a way. No will, no way. -
I have created the requested sym-links, but still had no lucky with the PXE boot process.
10.0.1.2 is the IP address that my wireless card was assigned.
192.168.2.1 is the IP address of the private network on the iMac or at least that was my understanding. I have never changed that setting as it comes by default with VMware Fusion.Has seen below I have created the sym-links:
I also bump the permission to 777. I was reading another FOG posting that suggested that has a possible fix.
I edited the dnsmasq.d/ltsp.conf file to related the sym-link change.
I also added in some more lines. I tested the tftp after creating the sym-link and then again after switch to undionly instead of undionly.kpxe.
I tried to pull the files from the server TFTP on my computer (iMac), my coworkers (PC), and my bosses (Mac Pro). The results were as follows:
Mine:
tftp> get 130.18.192.41:undionly.kkpxe Received 103224 bytes in 0.0 seconds tftp> get 130.18.192.41:undionly.kpxe Received 103273 bytes in 0.0 seconds tftp> get 130.18.192.41:undionly.0 Error code 256: File not foundCoworker:
tftp –i 130.18.192.41 get undionly.kpxe Timeout Occurred Connection Request Failed tftp –i 130.18.192.41 get undionly.kkpxe Timeout Occurred Connection Request FailedWit the Windows machine I also tried to specify a different download location “%HOMEPATH\Downloads\undionly.kpxe” and “%HOMEPATH\Downloads\undionly.kkpxe”. These result failed as well.
Bosses:
tftp> get 130.18.192.41:undionly.kkpxe Received 103224 bytes in 0.0 seconds tftp> get 130.18.192.41:undionly.kpxe Received 103273 bytes in 0.0 secondsI had him run the following line and his results were the same as mine.
Fun fact, but could me just looking to much into the problem. I downloaded nmap to my iMac and checked port 69 on my FOG server. The result are below:
-
@Dalton-Childers After a session via TeamViewer with Dalton, I found that his virtualized FOG server was not receiving any traffic from anything besides the physical computer it was running on and possibly the internet.
I did a tcpdump and transfered the file via TFTP to the host computer - and examined it with WireShark. We used the reported MAC address from the FOG server as a filter and a single packet showed up. However without a filter all the traffic from the hosting computer shows up.
I think that this problem is related to the VM setup on the iMac, and I recommended that Dalton try to install FOG on a physical machine using all the same settings - and see what happens, or to use a physical machine and have his Network Team change the DHCP options 066 and 067 for him so that he doesn’t have to use dnsmasq.
Just awaiting for a reply at this point to see what path he goes down and how it goes.
