Hopefully this question / issue has not already been dealt with, or if so, hopefully someone can point me to that thread/documentation :)
Problem:
1) I have a new on-board NIC (Intel 82566DM GB) that does not seem to be working with clonezilla.
2) It works fine booting via PXE into the little DRBL menu, and I can see the cloning session I have started for the client (in this case a clonezilla save disk session)
3) BUT... when I select the clonezilla session the process dies when it again attempts to connect to the dhcp server... or at least it "appears" to die here.
4) Unfortunately, when the process "dies", all I can see on the screen is the below error... there is no way to scroll back up to look at the specific output from the network/dhcp setup and it scrolls by so fast that I can't read much in the way of details. However, I can remember that the system seems to "see" a dhcp server but it never gets an IP address for the NIC. The client then promptly fails showing the little splash screen with 4 possible reasons (including an unsupported NIC) and advises going to your FAQ. The specific failure error is:
"FATAL ERROR: Failed to mount root file system!!!"
... plus the 4 possible reasons below this line and the URL for the FAQ
Other Info:
1) During the attempt to get a dhcp IP, the system seems to communicate with the other dhcp server on the network, however, none of the other clients have this problem... and even this client communicates with the drbl dhcp server during the PXE phase and gets its assigned IP without trouble.
2) However... though it does ultimately get a dhcp IP during PXE, this system takes almost 30s to get assigned an IP, while the other clients get it in 1-2s or even faster
3) I have the newest stable versions of clonezilla and drbl
Possible solutions:
1) I have thought about upgrading the driver for the NIC (there are new ones for the linux kernel being used), but how do I do this so that the driver is applied to the clonezilla boot image? Do I have to apply it to the server's kernel and clonezilla gets it from there, or do I have to do something special?
2) Is it possible there is some IP conflict? Could clonezilla be expecting to get an IP but does not realize the card was already issued an IP during the PXE process due to some card specific problem or some other issue I do not know about (again, everything is working perfectly for all other computers with different NICs)?
Sorry for what are probably mundane questions, but I am very new to the PXE/cloning domain.
Andrew.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The two dhcp servers do not seem to cause any trouble... drbl and clonezilla is working great (except for this unusual glitch). The two servers have been told to play nicely and not interfere with one another ;)
Andrew.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Yes, there are two dhcp servers, but they are setup to allow or deny specific IP addresses, so there shouldn't be any conflict (have not had any conflict anywhere else).
The problem was specifically with the room's network. I would like to know what the timeout for clonezilla's dhcp requests are though. It seems it tries the drbl dhcp server and after getting no reply moves on to the regular dhcp server. The network problems were resulting in the replies not coming back from either of the dhcp servers before it seemed to timeout and go to the fail screen.
So, if someone knows how to change the timeout time for the clonezilla client image, that would help... though like I said in my "Update" post, the network problems in there are being looked at, so I probably will not be changing the timeout... I am just curious in the event this problem occurs again... at least I can still get the clone going while I deal with the underlying problem.
Thanks!
Andrew.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Sorry for the trouble, it seems the problem was actually with the network in the room where this new computer was hiding out. For some reason the virtual networking setup was messed up for that room and dhcp requests were not going through properly.
We are working on resolving this issue, but I have the computer working fine in another room now.
I do have a question though, what is the dhcp timeout for the clonezilla image? After how many seconds does it give-up attempting to get an IP?
Thanks for your help.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
You can tune the number of udpcpc in the net initrd, say, 5 times (default is 3 times):
/opt/drbl/sbin/mknic-nbi --retry-max 5
For more info, check
/opt/drbl/sbin/mknic-nbi --help
Enjoy.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hopefully this question / issue has not already been dealt with, or if so, hopefully someone can point me to that thread/documentation :)
Problem:
1) I have a new on-board NIC (Intel 82566DM GB) that does not seem to be working with clonezilla.
2) It works fine booting via PXE into the little DRBL menu, and I can see the cloning session I have started for the client (in this case a clonezilla save disk session)
3) BUT... when I select the clonezilla session the process dies when it again attempts to connect to the dhcp server... or at least it "appears" to die here.
4) Unfortunately, when the process "dies", all I can see on the screen is the below error... there is no way to scroll back up to look at the specific output from the network/dhcp setup and it scrolls by so fast that I can't read much in the way of details. However, I can remember that the system seems to "see" a dhcp server but it never gets an IP address for the NIC. The client then promptly fails showing the little splash screen with 4 possible reasons (including an unsupported NIC) and advises going to your FAQ. The specific failure error is:
"FATAL ERROR: Failed to mount root file system!!!"
... plus the 4 possible reasons below this line and the URL for the FAQ
Other Info:
1) During the attempt to get a dhcp IP, the system seems to communicate with the other dhcp server on the network, however, none of the other clients have this problem... and even this client communicates with the drbl dhcp server during the PXE phase and gets its assigned IP without trouble.
2) However... though it does ultimately get a dhcp IP during PXE, this system takes almost 30s to get assigned an IP, while the other clients get it in 1-2s or even faster
3) I have the newest stable versions of clonezilla and drbl
Possible solutions:
1) I have thought about upgrading the driver for the NIC (there are new ones for the linux kernel being used), but how do I do this so that the driver is applied to the clonezilla boot image? Do I have to apply it to the server's kernel and clonezilla gets it from there, or do I have to do something special?
2) Is it possible there is some IP conflict? Could clonezilla be expecting to get an IP but does not realize the card was already issued an IP during the PXE process due to some card specific problem or some other issue I do not know about (again, everything is working perfectly for all other computers with different NICs)?
Sorry for what are probably mundane questions, but I am very new to the PXE/cloning domain.
Andrew.
I guess the problem is in the driver of your new NIC. Which kernel version are you using ?
About how to apply your kernel in the server to client, check these related FAQs:
http://drbl.sourceforge.net/faq/index.php#path=./2_System&entry=29_install_client_kernel.faq
http://drbl.sourceforge.net/faq/index.php#path=./2_System&entry=10_custom_kernel_for_client.faq
You have to make sure there is no 2nd DHCP service in your DRBL environment, otherwise you client will fail to boot disklessly.
The two dhcp servers do not seem to cause any trouble... drbl and clonezilla is working great (except for this unusual glitch). The two servers have been told to play nicely and not interfere with one another ;)
Andrew.
Other Info:
1) During the attempt to get a dhcp IP, the system seems to communicate with the other dhcp server on the network,
....or, there are two dhcp-server in the same subnet?
Yes, there are two dhcp servers, but they are setup to allow or deny specific IP addresses, so there shouldn't be any conflict (have not had any conflict anywhere else).
Like I said in my "Update" post:
http://sourceforge.net/forum/message.php?msg_id=4615232
The problem was specifically with the room's network. I would like to know what the timeout for clonezilla's dhcp requests are though. It seems it tries the drbl dhcp server and after getting no reply moves on to the regular dhcp server. The network problems were resulting in the replies not coming back from either of the dhcp servers before it seemed to timeout and go to the fail screen.
So, if someone knows how to change the timeout time for the clonezilla client image, that would help... though like I said in my "Update" post, the network problems in there are being looked at, so I probably will not be changing the timeout... I am just curious in the event this problem occurs again... at least I can still get the clone going while I deal with the underlying problem.
Thanks!
Andrew.
Sorry for the trouble, it seems the problem was actually with the network in the room where this new computer was hiding out. For some reason the virtual networking setup was messed up for that room and dhcp requests were not going through properly.
We are working on resolving this issue, but I have the computer working fine in another room now.
I do have a question though, what is the dhcp timeout for the clonezilla image? After how many seconds does it give-up attempting to get an IP?
Thanks for your help.
You can tune the number of udpcpc in the net initrd, say, 5 times (default is 3 times):
/opt/drbl/sbin/mknic-nbi --retry-max 5
For more info, check
/opt/drbl/sbin/mknic-nbi --help
Enjoy.
Steven,
You guys have done an incredible job with these two projects, keep it up!
Thanks again for the quick replies.
Andrew.