You might choose to try swapping the GPU, its power source, or its PCIe-extending paraphernalia where applicable. The most-recently-added GPU is likely causing your problems. It might be that the numbering order of the GPUs is the reverse of what you expect.Ĭontinue to iterate through this process, adding one GPU at a time, until an addition prevents you from finding a working graphical desktop anywhere. ![]() If not, try a reboot with the screen swapped to the second GPU. Boot the system, with the screen still plugged into the first GPU, and see if you get to a graphical desktop again. Next, shut down the rig again, and re-introduce your second GPU. In most cases, the system will boot and the desktop will be shown. In this case, the best course of action to identify this is try to isolate the problem GPU.īegin by powering on the rig with just a single GPU connected, having your screen plugged into this one GPU. Sometimes a minor hardware problem with a GPU means that although the system can boot, the graphical desktop will fail to display anywhere. In some cases this may help to give insight about which GPU gives home to the desktop, based on what ports exist on it. In the same terminal mentioned in the step above, you can also try to run the command ‘xrandr -d :0’. You can try your screen on each port of each GPU in turn until you find where the desktop is being shown. In most cases, there is nothing wrong with the system, and the graphical desktop has loaded somewhere. ❷ Iterate through all GPUs to find the active port: If there are errors in the log, they may be self-explanatory, allowing you to correct the problem, or you may choose to submit any errors you find to our support team, so they can help you solve the problem. You may use the up and down arrow keys and page up and page down keys to scroll through the output. Run the command: ‘journalctl -u gdm -n all -a -b’ to show the full display manager logs for this boot. This should automatically log you into the system as the ‘root’ user. Try either the key combination Ctrl+Alt+F2, or Alt+F2, to switch to an alternative command-line terminal. You may try the following troubleshooting steps:– During the first boot of CudoOS, we detect which of this hardware is present, and set up the necessary vendor-supplied driver software.Īfter this completes, the graphical desktop should usually show on the rig’s first supported GPU, even though the Linux kernel console messages up to this point may have successfully been displayed on a screen connected to any port. This is a Linux kernel console, and considering that it has continued to persist, then it’s likely the graphical desktop is being displayed on a different port to where your screen is currently plugged in.Īt present, CudoOS is supported on nVidia-based and AMD-based mining rigs. Sometimes when starting CudoOS, after the boot sequence completes, you may be left at a screen similar to the following: Successes have been reported with SD cards, and also with writing the IMG file directly onto a hard drive, but we do not directly support these scenarios. We only support USB flash drives and recommend you select a high-speed one, with a USB 3 interface, of size 16GB or greater. If you do not have a hard drive, use the CudoOS run-from-USB IMG file on your USB drive instead of the ISO installer. Using the ISO installer requires a hard drive to install onto. Otherwise, you can still use the Linux version of Cudo Miner, but you’ll need to set up your chosen Linux-based operating system and GPU drivers yourself. ![]() If you wish to confirm this is the problem, you can download an unmodified installer for Ubuntu 18.04 from their website and try it on your machine. Other configurations may work, but are not directly supported. You can try simplifying the installation environment by only having a single SATA hard disk drive in the machine, which has had its partition table wiped. If you have an unusual disk configuration, or there is existing data on your disk(s) in an unusual layout, the Ubuntu installer might not risk proceeding. ❷ The current layout or structure of the hard disk(s) is indeterminate, so installation was failsafe aborted.īecause the CudoOS installation from ISO runs unattended, the Ubuntu installer makes its best guess about where to install. Consider changing hardware to something that works with Linux kernel 4.15.0 if you wish to use CudoOS. The only version of Ubuntu supported by both nVidia and AMD at present is Ubuntu 18.04, and so it is what CudoOS has to be derived from. ❶ The disk controller in the machine is unsupported by Ubuntu 18.04 (Linux kernel 4.15.0). View our full CudoOS Installation video here CudoOS ISO installation problems: causes and potential solutions (No root file system is defined).
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |