I want to say for people, using the zstdmt (vs pzstd as I can’t get it to build on 32 bit) the imaging is MUCH faster even at -19.
It’s compressing at about the same rate, more or less, but it’s much faster than the 35 minutes I originally had tested against. Using zstdmt on the same image with all cores in use took the 35 minute time and got it all done in 2 minutes 58 seconds.
@Wayne-Workman I’m already looking at it.
Because of what you reported, I’m working on a hopefully better approach to getting the dumped out file.
In as simple a word, No.
The plan, however, is to give the user the option to choose default of pigz/gzip or zstd, but PartImage will not have any “default” usage medium. This is because PartImage has not seen any developments and still considers NTFS “beta” implementation. While it “works” it’s not actively developed (from what I’ve seen) and has become quite antiquated.
Partclone will, for now, be the “goto” for capturing images. This is regardless of the compression medium as the compression system is totally independent of the partclone process.
I do plan to add the ability to let the user decide the “default” compression type, but as for the “default compression size” I think 6 is still viable. While it’s not the ‘mid level’ for the zstd compression mechanism, trying to “default” compression numbers between multiple compression types would become very messy to manage. The only reason the “default” was added in the past was because it was unsettable at the time and there had only been one compression choice.
@Sebastian-Roth I’ve looked into it and corrected it on my side. I have not pushed it so I can try to fix the db backup issue.
Default capture method is now in the “FOG Configuration Page->FOG Settings->General Settings category.”
@Wayne-Workman Update and you too can now try it all out.
I updated the init’s and we are no longer having udev startup. It also comes with 0.2.88 version of Partclone!
I’ve added these items:
FOG_TASKING_ADV_SHUTDOWN_ENABLED
FOG_TASKING_ADV_WOL_ENABLED
FOG_TASKING_ADV_DEBUG_ENABLED
With schema update 252 of the working branch of fog. They will do what you were hoping for here now.
@andyroo54 I’ve added error checking for getting the $_FILES[‘blah’][‘error’]
The problem, in the case of too large a file chosen, is it completely blows away ALL $_POST/$_FILES variables as blank. This is why it’s reporting the “blank name”.
Unfortunately there’s not much I can do to fix that.
I have also added the suggested 3g limit instead of 100 megs. Of course, it won’t change this if your values have been set to other the the 8m (post_max_size) and 2m (upload_max_filesize).
@jj-fullmer and I had a remote session and the cause of the resize issue as described was found.
The issue he was seeing was the resize allowed a partition to be sized larger than the boundary of the disk. This only occurred, as far as I could tell, in the case of 4k disks again, but I’ve added code to correct for this in ANY case of the chunk size passed in.
Just thought I’d give an update to the “resize” issue as it was described and I’ve released RC-16 in an attempt to ensure this is corrected for.
Confirmed, found, fixed and pushed (Notice I didn’t say tested cause well I don’t have anything to test with. sorry).
Apparently with 0.2.88 (maybe earlier but before that for us was 0.2.69) they added a “note” feature
The picture shows the (note:) line now which shows the Storage location and the Image name.
Just a guess but this drive is likely based on 512 sectors expecting to be started at 63 but the image is in 2048 which makes the drive not understand how to use the drive (why Windows showed it as inactive)
Might I ask what you’re using this for? This wasn’t added because it caused other problems (when I tested it).
Adding it is simple enough sure, but what other impacts does it have?
I’ve found the issue as described and it is now fixed in the latest.
This already exists in the form of a notification gui. It will give the user 60 seconds before any action happens. It also allows the user to select a delay/postpone.
@coco65 I’ve edited the code and removed the .mo files. They will only be generated if installlang variable is set.
I can’t really remove the language menu item though. So not having the .mo files will make any language selected automatically defaulted to english.
https://www.microsoft.com/en-us/Useterms/OEM/Windows/10/UseTerms_OEM_Windows_10_English.htm
In case anybody’s overly concerned. Here’s the OEM EULA. Notice, it states that you are allowed to transfer the license to another user so long as it’s with the device and the software is installed and the product key is given with it. There are no rules on what you can/cannot document. That’s like saying you can’t pull the serial number from the bios.
Just trying to get people to calm down. We know the device and software are owned by the person. We know how to obtain the information. We don’t know how the user intends to use it after it’s been stored, but that’s out of our hands. We cannot control what the admins/users do with the key once it’s know, but that’s out of our control to begin with. There’s plenty of ways for users to get that information to begin with, I don’t see anywhere in the EULA where it states we cannot store a copy of it for ourselves.