Bom dia Steven, Conforme os testes realizados, confirmo que o comportamento não afeta o desempenho do servidor – em ambos os cenários, o sistema performou dentro do esperado. No entanto, observei um padrão diferente no uso do cache: enquanto no servidor ele utiliza toda a memória livre, no meu desktop esse comportamento não se repete. Apesar disso, como não há impacto na performance, entendo que seja uma particularidade da configuração ou do workload do servidor. Se houver qualquer preocupação adicional...
My C; drive is running out of room. Since this is the boot drive, it needs to have certain sectors in certain locations. I have a new hard drive ready. The question is do I partition the drive before or after I transfer the old C; drive over using clonezilla.
Cool. Thanks for sharing that.
So when was the performance affected? When the image was saved? How did you run Clonezilla SE? The client is using NFS to save the image? Or you use the samba service from Proxmox?
Please send up the log file and take a look(/var/tmp/clonezilla.log). After failure, check the nvme hard drive through the shell(lsblk).
Hi Steven, Unfortunately I missed your answer too :-( Yes, need this feature for already existing file systems and partitions. Anyway I can tell you that the problem of randomly assigned block devices at startup, for me was solved a few months ago from release 3.2.1-18 and boot parameter "ocs_1_cpu_udev": Added a mechanism to mitigate the random order of block devices. It can be enabled by adding the boot parameter "ocs_1_cpu_udev", i.e., when udev is started in initramfs, only 1 CPU is online. After...
Hi, I use clonezilla live to deploy win 10 images on computers. It works fine, but on a PC (HP laptop) with an NVME SSD, partition selection fails. It sees the disk when selecting the image target, the process starts and then I get an ‘unable to find target partition sda1’ error. The list of partitions is displayed below and the SSD's name is indeed "sda" Do you have any idea how to help me solve this problem? Thank you!
Recently, there was a need to translate the system between hard drives of different capacities, and after several days of research and testing, it was finally implemented. Based on the spirit of sharing, we will now record and share the following: prototype Win11 system standard installation, 4 partitions, 1:100M,EFI ; 2: Windows reserves partitions; 3: Windows system partition; 4: Windows recovery partition. After backup, needs to be restored to another computer, and the hard drive size may be between...
I was careless, my oversight was to remove - r when modifying the restore parameters. Adding - r will solve the problem caused. /usr/sbin/ocs-sr -batch -r -iefi -k -scr -p true restoredisk $img $disk
After restoring the Windows 11 system using the following script (proportionally enlarging the partition by - K1), it was found that the hard disk capacity in the resource manager was different from the actual capacity in the disk manager after entering the system normally. Have any engineers encountered this situation before, or do you have any solutions? /usr/sbin/ocs-sr -g auto -e1 auto -e2 -batch -k1 -scr -p true restoredisk $img $disk
OK
Hi chen dragon, thank you for your advice, unfortunately I opted for a file and folder level copying which was really inefficient. I'll keep you advice in mind for experiment in the future. Thanks.
Thanks i had the same issue cloning using DISM and same after cloning with clonezilla and i could not understand why.
Can you provide the script? I didn't get it running.
Can you provide the script? I didn't get it running.
-K0 Use the partition table from the image -K Do NoT create a partition table on the target disk -k1 Create partition table proportionally -K2 Enter command line prompt to create partition manually later -j0 Use dd to create partition (NOT OK if logical drives exist) After partitioning a 20T hard drive into 8k sectors,try -k to see the test results.
Good evening! I'm experiencing an issue with Clonezilla Server. I've created a VM on Proxmox and installed Clonezilla Server to perform network-based cloning. However, every time I restore an image via PXE, the system uses nearly all available RAM for disk caching. While this behavior is normal in Linux (due to page cache usage), it's affecting performance in environments with limited memory. I would appreciate guidance on how to optimize memory usage and limit this caching during the restore pr...
Hello all, I'm having issue getting CloneZilla to boot. I've tried the Stable Zip and the Plucky Iso build. I've included a photo of the error. But it says, Failed to open \EFI\BOOT\mmx.64.efi - Not Found failed to load image N(?): Not Found Failed to start MokManager: Not Found Something has gone Seriously wrong: import_mok_state() failed: Not Found. I'm not sure what the heck is going on. I looked in the path mentioned in the error and mmx64.efi doesn't exist there, even with invisibles showing....
Hi Steven, I reattempted the disk to disk cloning process again, but this time I picked beginner mode and -k0, the result was my windows is able recognize the partition as NTFS, maybe something is wrong with the -k1 option that make changes to the partition table? Also I encounter another issue, my original drive's NTFS' cluster size is the default 4k which means the maximum partition size is 16TB so I am unable to fully utilize my new 20 TB drive, ideally the new drive should have a 8k+ cluster...
Hi Steve, I reattempted the disk to disk cloning process again, but this time I picked beginner mode and -k0, the result was my windows is able recognize the partition as NTFS, maybe something is wrong with the -k1 option that make changes to the partition table? Also I encounter another issue, my original drive's NTFS' cluster size is the default 4k which means the maximum partition size is 16TB so I am unable to fully utilize my new 20 TB drive, ideally the new drive should have a 8k+ cluster size,...
Here’s the breakdown of events. These were the settings on both source and target workstations. SECURE BOOT: On TPM: On CLONEZILLA: Image fails to boot on target workstation (Can’t find OS) SECURE BOOT: Off TPM: On CLONEZILLA: Image fails to boot on target workstation (Can’t find OS) SECURE BOOT: Off TPM: Off CLONEZILLA: Image boots w/o issue on target workstation (Finds OS) I haven't tried leaving Secure Boot on with TPM off to see if TPM is the issue or if it requires both to be off.
No idea. It looks normal to me from the log file. Are you able to mount the cloned partition on Clonezilla live command line prompt like this? sudo mount /dev/sdb2 /mnt
No idea actually. Here we actually can turn Secure Boot and TPM on, and the cloning ran smoothly. In your case, did you try to turn both Secure Boot and TPM off?
OK, please keep us posted.
So maybe you can provide the file /var/log/clonezilla.log next time. It will be easier for us to find some clues.
I have a 10 TB data drive (D: Basic Data Partition NTFS) that contains specific user files such as documents, desktop, and other media (not Windows OS and other system files that is in the C: drive), I wish to migrate the entirety of this older disk to a new larger disk drive of 20 TB, I followed the documentation for disk cloning and select beginner mode and -k1, the result is a new disk drive (D: Basic Data Partition RAW) that is recognized by windows as RAW I checked the logs, but there doesn't...
Here are some more details. I installed WIN11 with TPM & Secure Boot On. Once completed, I went into the BIOS and turned off TPM and Secure Boot. Booted into Clonezilla. Then I made a “SAVEDISK” image. I then went to the target workstation and turned off TPM & Secure Boot. Booted into Clonezilla. Then I did a “RESTOREDISK” to that machine. The above steps finally worked but prior to that, I only shutoff Secure Boot but not TPM. When only shutting off Secure Boot, I would get the following error....
Thank You for the reply. My very old (like me) camera does not do video. I thought about getting a friend with a cellular telephone to capture the error but I did something else first. I reloaded all the "amd" files that I had access to. ALL other programs work flawlessly. I reinstalled the two stable clonezillas on two separate memory sticks (a lot of work). The end result is that I can no longer reproduce the error messages. The boot process still stalls after utils/win64/syslinux64.exe - 100%...
I can't take photos. Clonezilla finally cloned the p1 after having created it as an empty one on the destination with cfdisk before. I've read absolutely everywhere that this was managed by Clonezilla... Another little glitch: I've tried only with the p1 by now, so only checked this one in the dialog, accepted partition integrity check after that, and it checked the three ones however...
We did an improvement about the text font size. Now it can be automatically detected before starting Clonezilla menu. As for the menu for keyboard and language, since it has to show multi languages, we use jfbterm. Its mechanism is different from the pure console mode, so we have to find another way to deal with the font size in the future. Anyhow, the font size in the console should be easier to handle: Please give testing Clonezilla live 3.2.3-11 or 20250713-* a try: https://clonezilla.org/dow...
Yes, I had computers that failed in broadcast disk-restore over LAN. I "played safe", re-starting the whole restore process from zero but a very good alternative could have been to let all the other computers to finish the current partition (sda2) and after that to do a "partition restore" (not "disk restore") starting with the next partition (sda3) and excluding the computer that failed (this computer can be restored individually/separately). I'm sharing all these information hoping that they will...
Returning to the initial post of mr. Frank, my test revealed that BT cloning works very slow when the "slices" are sent by the server but works very fast when computers that have been started later are catching up - so they receive "slices" from multiple seeders/servers. I think the process should be equally fast when the "slices" are sent only from the initial server.
on reflection this line of questioning is probably better suited for the DRBL forum; posted here; https://sourceforge.net/p/drbl/discussion/Help/thread/d981bc792e/ please let me know if I need to wrap this one up and how I should? thanks
Basically yes, you can use the green command to start the multicast service on the server. Once it's done, you can boot the clients to start deployment. However, to automatically schedule that on the server, you need to run "at" command or "crontab". You can google to find that.
Oh, I got an idea to implement that. Will test and if it's OK, the mechanism to set the proper font size will be added in the near future.
Steven, Do you mean when I use multicast and a green command script will schedule the job ?
Thanks for your help. Actually "efitextmode 0" will only affect the grub menu. However, the text font size you mentioned is actually in GNU/Linux system. Hence that's another story. So far we do not have a good mechanism to detect the best solution of the monitor then use the proper font size in GNU/Linux console mode automatically. Therefore you have to try and get the best one. If you know a mechanism to do that automatically, please share that, We will try to implement that in the future. Tha...
Those steps are still in the early stages so GNU/Linux has not take over the system. Hence it's not easy to electronically capture the data. However, you can take a video. That should be good enough for us to debug for you. Thanks.
I tried Clonezilla live 3.2.3-10 and it works in my son's computer. But we noticed, that all text-size related options [that we tested] produced the same text size, it seems like alternative 2 in my opening post. I am not sure that you want that result. Edit 1: Selecting 'Large font' makes the font too big, so that several of the dialogue boxes will not be within the screen. All the other size choices will make the text the same size. (However, while selecting language, the text size will be the...
I tried Clonezilla live 3.2.3-10 and it works in my son's computer. But we noticed, that all text-size related options [that we tested] produced the same text size, it seems like alternative 2 in my opening post. I am not sure that you want that result. Edit 1: Selecting 'Large font' makes the font too big, so that several of the dialogue boxes will not be within the screen. All the other size choices will make the text the same size. (However, while selecting language, the text size will be the...
I tried Clonezilla live 3.2.3-10 and it works in my son's computer. But we noticed, that all text-size related options [that we tested] produced the same text size, it seems like alternative 2 in my opening post. I am not sure that you want that result. Edit: Selecting 'Large font' makes the font too big, so that several of the dialogue boxes will not be within the screen. All the other size choices will make the text the same size. (However, while selecting language, the text size will be the same...
Clonezilla live 3.2.3-10 in my son's computer 1-clonezilla-grub-menu_in_Dell-7080-with-3840x2160-monitor.jpg 1-for-default-and-other-font-selections_in_Dell-7080-with-3840x2160-monitor.jpg 1-for-specifically-1024x768-resolution_in_Dell-7080-with-3840x2160-monitor.jpg 1-large-font-after-setting-language_in_Dell-7080-with-3840x2160-monitor.jpg Clonezilla live 3.2.3-10 in my Dell Precision M4800 with a 1920x1080 screen 2-for-default-and-other-font-selections_in_Dell-M4800.jpg 2-large-font-after-setting-language_in_Dell-M4800.jpg...
Clonezilla live 3.2.3-10 in my son's computer 1-clonezilla-grub-menu_in_Dell-7080-with-3840x2160-monitor.jpg 1-for-default-and-other-font-selections_in_Dell-7080-with-3840x2160-monitor.jpg 1-for-specifically-1024x768-resolution_in_Dell-7080-with-3840x2160-monitor.jpg 1-large-font-after-setting-language_in_Dell-7080-with-3840x2160-monitor.jpg Clonezilla live 3.2.3-10 in my Dell Precision M4800 with a 1920x1080 screen 2-for-default-and-other-font-selections_in_Dell-M4800.jpg 2-large-font-after-setting-language_in_Dell-M4800.jpg...
Thank You for the reply. It was appreciated. My original post said, "got a lot of error messages - unfortunately too fast to copy". A subsequent post said, "took a few pictures - none good enough for public consumption". The images were blurry. I took the parts out that I could read and posted them. I can probably reproduce the error messages if you can advise me how to electronically capture the data. This is above my knowledge level. Thank You p-bea
Could you please take some photos about that then share? Thanks.
I tried Clonezilla live 3.2.3-10 and it works in my son's computer. But we noticed, that all text-size related options [that we tested] produced the same text size, it seems like alternative 2 in my opening post. I am not sure that you want that result.
oh, appears that the 'php' file got truncated; <?php $mac = $_GET['mac']; $pool = range(ip2long('x.211'), ip2long('x.220')); // File for storing leases $lease_file = '/var/www/leases.json'; $leases = file_exists($lease_file) ? json_decode(file_get_contents($lease_file), true) : []; // Assign new if not leased if (!isset($leases[$mac])) { $used_ips = array_values($leases); foreach ($pool as $ip_long) { $ip = long2ip($ip_long); if (!in_array($ip, $used_ips)) { $leases[$mac] = $ip; break; } } file_put_contents($lease_file,...
Hi there, Thanks for reading. on the back of the drbl and clonezilla integration I would like to accomplish a linux boot using drbl and clonezilla which does not run a 'dhcpdiscover' and instead uses an assigned address from a 'boot.php' embeded script. essentially my client should boot into ipxe, run a php script and then boot 'clonezilla' or 'distro-blah' (I did have the clonezilla menu and such working but fiddled and lost the config) essentially the php script assigned 'ip' should be used to...
Please take some photos about the error messages on the screen then share them. It's easy for us to debug. Thanks.
If you compress the image using gzip/pzip (-z1/-z1p), please try to use zstd (-z9p).
As I mentioned, ATM there is no any schedule function for Clonezilla server or lite server. If you really want to use the fixed number of clients, just stick to unicast, broadcast or multicast. However, for the server side, you have to have your own script to start that.
Next time please take some photos about the error messages on the screen then share them. It would be easier for us to debug. Thanks.
Please give testing Clonezilla live >= 3.2.3-10 or 20250712-* a try: https://clonezilla.org/ Your feedback has been included. Please let us know the results if you test it. Thanks.
Unable to find target partition nvme0n1p1
Bonjour, Merci pour votre réponse. Je travaille dans un lycée et je fait la restaurations dans différentes salles informatiques. J'avais effectivement pensé à un problème de câble réseau dans la première salle et de switch. j'ai changé le switch et les câbles des PC concernés mais j'ai eu le même problème. Le problème est aussi apparu dans les autres salles ayant ce modèle de PC Dell (optiplex AIO 5260) Quel autre protocole est ce que je peux utiliser et comment ? Je ne vais pas répondre tout de...
Steven, I am searching online but I don't see any information using clonezilla for schedule a restore.
Thank You for the reply. It was appreciated. Unfortunately prior to contacting you I downloaded every version(?) I could and I tried them. All resulted in the same. The program stopped in the same place. The only thing I did not try was loading your program from another computer at our place to see whether the amdgpu components were the fault. Not yet anyhow! Thank You p-bea
" still can't get work after three machine was restoring and on lite server shows me No MTA is running. Then I have to click Y twice and type /tmp/ocs-live-feed-img-2025-07-08-15-19. " -> Please take some photos about the messages on the screen then post them.
I followed all your suggestion and I still can't get work after three machine was restoring and on lite server shows me No MTA is running. Then I have to click Y twice and type /tmp/ocs-live-feed-img-2025-07-08-15-19. then whatever machines were in PXE would continue to restore it. what should I do to let it keep 3 machines restoring after others finished
Steven, I know it but I want to schedule the next machines after other machines are done. ? which method do I need to do ?
Thanks for sharing that. Yes, different mechanisms have different goals. Some people would like slow and steady, and some would like to complete asap. For multicast or broadcast, if one of the clients fails, then all the restoring might hang... Some people like to just start the BT server and boot the clients anytime... Hence choose the proper mechanism to meet your own need. Thanks.
If you are talking BT mode, well, once the Clonezilla lite server is ready in BT mode, just boot your clients via PXE. No matter how many clients you boot, they will be restored.
Thanks for your suggestion. We will check if this will have any side effects.
On GNU/Linux, to make a USB flash stick boot on a uEFI machine, it's very easy. You actually do not have to use Ether. Just have a vFAT partition, and unzip the Clonezilla live zip file on it: https://clonezilla.org/liveusb.php#linux-setup-uefi
Hiya, 20GB/min verify speed.... very nice indeed :o) Some feedback on the ISO... Using Ubuntu 22.04.5LTS, I could not get "Etcher-linux-x64-2.1.2.zip" to create a usable USB stick at all with the "clonezilla-live-3.2.3-6-amd64.iso" In the end I switched over to my Windows 10 and used Rufus 4.6. Rufus reported that the ISO file had two ways of being put onto a USB stick, either as a ISO copy or a Direct Image. Only the ISO copy method worked. Cheers.
Next time you can try alternative release of Clonezilla live, i.e., Ubuntu-based one: https://clonezilla.org/downloads.php
Actually I have no idea since I do not use Fedora Linux often. I suggest you ask this on Fedora forum.
"Even when making the image with secure boot off, it fails to load up on another machine even with that workstations secure boot off. " -> What were the error messages on the screen? Please take some photos and share that with us.
"since this issue affects only some drive (or perhaps drives / chipset combination, there is not much data about it), it has been a safe option not to enable it by default." -> Yes, this is what we thought... The Linux kernel should deal with the cache mechanism. However, maybe it's buggy for some types of hardware. Then you can enable it with "-edio". This I believe it's a safer way.
I believe the issue is the network. Try to use different protocol, or change your network cable....
The "bt zone" folder, with the "sliced" partitions is derived from the initial image of 270GB, compressed with z9p. The total partitions, UNCOMPRESSED, sould exceed 400GB: 65,9GB (Ubuntu / - that is root and system folders)+39,8GB (Ubuntu /home) +324,7GB (Windows 11) = 430,4GB Edit: The "bt zone" folder has 431,4GB.
The "bt zone" folder, with the "sliced" partitions is derived from the initial image of 270GB, compressed with z9p. The total partitions, UNCOMPRESSED, sould exceed 400GB (411GB if I recall correctly): 65,9GB (Ubuntu / - that is root and system folders)+39,8GB (Ubuntu /home) +324,7GB (Windows 11) = 430,4GB
I will let the pictures speak for themselves, below! I'm sorry but the final answer is that BT method CANNOT compete with Broadcast method! In short words we see over 8 hours with BT method versus 1,5 hours with Broadcast method, over gigabit LAN. BT method is very fast ONLY when computers that have joined later are catching up with the others! Otherwise it is very slow comparing to Broadcast. This is a Google Drive folder with all the pictures: https://tinyurl.com/clonezilla2025-07 Please have a...
I will let the pictures speak for themselves, below! I'm sorry but the final answer is that BT method CANNOT compete with Broadcast method! In short words we see over 8 hours with BT method versus 1,5 hours with Broadcast method, over gigabit LAN. BT method is very fast ONLY when computers that have joined later are catching up with the others! Otherwise it is very slow comparint to Broadcast. This is a Google Drive folder with all the pictures: https://tinyurl.com/clonezilla2025-07 Please have a...
I will let the pictures speak for themselves, below! I'm sorry but the final answer is that BT method CANNOT compete with Broadcast method! In short words we see over 8 hours with BT method versus 1,5 hours with Broadcast method, over gigabit LAN. BT method is very fast ONLY when computers that have joined later are catching up with the others! Otherwise it is very slow comparint to Broadcast. This is a Google Drive folder with all the pictures: https://tinyurl.com/clonezilla2025-07 Please have a...