Error trying to restore GPT partition tables
-
Hy
I reset the server. I installed Ubuntu and then the latest Fog update. Everything worked correctly I even managed to capture and deploy an image of a MacMini without worries. When I put on the IMac I capture the image again on Fog. Then at the time of the deployment I found the same error message with a command line to test “sgdisk -gel /images/CO515/D1.mbr /dev/sda”
So I tried this command on the server and there is the drama. After restarting the machine this is the message I had
I tried everything and could not retrieve the data from the server. So I reinstall everything and do all the operations again. Installing Ubuntu, Fog, Capturing and deploying the MacMini is ok. Same operation for the IMac I had the same error message I retype the same command line on the server and I have my server that had the same worries. I do not think this command line is the right one. Do you have any other solution?
I may have a complementary information to add you there is on this IMac a DualBoot with windows. The hard disk of the IMac is partitioned into 2 a MAC image and a PC image. There is no DualBoot on the MacMini
Bonjour
J’ai remis à zéro le serveur. J’ai installer Ubuntu puis la dernière mise à jour de Fog. Tout fonctionnais correctement j’ai même réussi à capturer et déployer une image d’un MacMini sans soucis. Quand je me suis mis sur l’IMac j’ai capturer à nouveau l’image sur Fog. Puis au moment du déploiement j’ai retrouvé le même message d’erreur avec une ligne de commande à tester “sgdisk -gel /images/CO515/D1.mbr /dev/sda”
J’ai donc essayé cette commande sur le serveur et là c’est le drame. Après avoir redémarrer la machine voici le message que j’avais
J’ai tout essayé et impossible de récupérer les données du serveur. J’ai donc tout réinstaller et refais toutes les opérations. Installation de Ubuntu, Fog, Capturer et déployer le MacMini c’est ok. Même opération pour l’IMac j’ai eu le même message d’erreur j’ai retaper la même ligne de commande sur le serveur et j’ai mon serveur qui a eu le même soucis. Je ne pense pas que cette ligne de commande soit la bonne. Avez-vous d’autre solution ?
J’ai peut être une information complémentaire à vous ajouter il y a sur cette IMac un DualBoot avec windows. Le disque dur de l’IMac est partitionné en 2 une image MAC et une image PC. Il n’y a pas de DualBoot sur le MacMini
-
@melvinpaz Would you mind installing FOG 1.3.3 official?
There was a bug introduced that had impacted restoring GPT partitions that I’m pretty sure is fixed in the 1.3.3 version of FOG.
-
Hey @Tom-Elliott
I have reformed my server with the update of Fog in 1.3.4
I Capture the image from my MAC and by deploying I have the error message again except that now I have one more line that tells me about the kernel.
Below is my kernel version. Should I test with different kernels?
Thank you for your help to the community.
Bonjour @Tom-Elliott
J’ai fais reformater mon serveur avec la mise à jour de Fog en 1.3.4
J’ai télécharger l’image de mon MAC et en déployant j’ai à nouveau le message d’erreur sauf que maintenant j’ai une ligne en plus qui me parle du kernel.
Ci-dessous voici ma version de kernel. Dois-je faire le test avec différent kernel ?
Merci de votre aide la communauté.
-
Hey @Tom-Elliott
I allow you to restart because I have the same error message with a more recent revision on a MAC.
Do you have an idea ?
@+
-
@melvinpaz Can I ask you to update to the latest 1.4.0RCx release? There were several disk issues resolved in 1.4.0RC1. The developers are getting ready to release 1.4.0 (stable) very soon. I’m sure they would want to ensure your issue is resolved before the next stable release is made.
-
@george1421 That she is how git to retrieve the version 1.4.0 RCx ?
-
@melvinpaz If you already have git installed, you should be able to just change to the dev-branch and rerun the installer.
All things git and upgrading can be found:
https://wiki.fogproject.org/wiki/index.php?title=Upgrade_to_trunk
-
Hey @Tom-Elliott et @george1421
The problem persists despite the update with the revision RC 1.4.0 RC8
I captured the image again and tried to deploy it without success
Do you have another idea?
@+
-
The problem persists despite the update with the revision RC 1.4.0 RC9
I captured the image again and tried to deploy it without success
Do you have another idea?
@+
-
@melvinpaz While it is appreciated, you don’t need to continue posting a picture of everything, only if the error itself is different.
Can you please reschedule this system as a “debug” task? I am suspecting the error is the e in the command and by being in debug we can test to find out exactly what’s wrong.
-
I’ve made a very small change to the init’s based on what I suspect is the problem. Mind updating to RC 9.1 to see if it helps? No you do not need to post another picture, just let us know if it fails in the same manner (more or less).
-
Hey @Tom-Elliott
I thank you for the update but the problem remains the same. Are there any restrictions on a minimum space in the partition that you want to deploy?
Let me explain :
I captured an image with 20 GB of available space. I get my capture but maybe the 20GB space on this partition is not enough to perform the deployment.I am currently reformatting the hard drive of the deployment machine and reinstalling the system without putting any contents in the hard drive. I will perform the capture and then the deployment. If it works, it takes a minimum amount of space available in the image in order to realize the deployment.
If you have this information about a percentage or quantity in GB of available space that should be left on the partition to deploy it would be cool to communicate it or share the link of this information because I did not find it.
We do not tell you it may not be enough but still thanks the community for the work done for this really practical tool
@+
-
@melvinpaz If the system was captured in resizable mode, the smallest size disk that can be used should be the “used” space, but this is not a perfect science unfortunately. Rather than capture as non-resizable, maybe try for resizable? (Unless there’s a specific reason not to).
When dealing with non-resizable images, the disk sizes must EXACTLY match one another. So if the reference machine was captured with a disk of 500GB and the system being deployed to is 250, you will run into problems. This would be regardless of the “used space” sizes.
-
Hey @Tom-Elliott
I confirm that now everything works correctly. The error message was on one of the 2 partitions on the hard drive was too full. I will refine the settings and make you a return on the minimum size of available space to deploy.
I confirm that I capture and deploy on the same machine and therefore even hard drive. So no space problem available.
We can pass the inviter in resolve
As soon as I have the information on minimum available space I will add them in this topic@+
