Created attachment 1899 [details] screenshot Wrong information of images. (1) wrong image file paths old: /buildarea1/build/toasterauto/build/tmp/deploy/images/qemux86core-image-minimal-qemux86-20140402052043.rootfs.ext3 new: /buildarea1/build/toasterauto/build/tmp/deploy/images/qemux86/core-image-minimal-qemux86-20140402052043.rootfs.ext3 (2) wrong ext3 image size. It should be 12M, not 64M. log: ysun0@p2:/buildarea1/build/toasterauto/build/tmp/deploy/images/qemux86$ du core-image-minimal-qemux86-20140402052043.rootfs.ext3 -sh 12M core-image-minimal-qemux86-20140402052043.rootfs.ext3
Passing to the Toaster backend.
(In reply to comment #0) > Created attachment 1899 [details] > screenshot > > Wrong information of images. > (1) wrong image file paths > old: > /buildarea1/build/toasterauto/build/tmp/deploy/images/qemux86core-image- > minimal-qemux86-20140402052043.rootfs.ext3 > new: > /buildarea1/build/toasterauto/build/tmp/deploy/images/qemux86/core-image- > minimal-qemux86-20140402052043.rootfs.ext3 The "new" name is the correct one: the "old" one was missing a '/' between the qemux86 and core-image-minimal (see bug 6090) Also, as part of the fix for bug 6091, only the file name should be shown (core-image-minimal-qemux86-20140402052043.rootfs.ext3).
> > (2) wrong ext3 image size. > It should be 12M, not 64M. > log: > ysun0@p2:/buildarea1/build/toasterauto/build/tmp/deploy/images/qemux86$ du > core-image-minimal-qemux86-20140402052043.rootfs.ext3 -sh > 12M core-image-minimal-qemux86-20140402052043.rootfs.ext3 But, I do see a difference in the rootfs file sizes. I have a core-image-minimal build that created ext3 and tar.bz2 files. This is the size information: For ext3 Toaster: 64.0 MB Size on disk: 67.1 MB For tar.bz2 Toaster: 2.3 MB Size on disk: 2.4 MB Passing this onto Cristiana and moving to 1.7
I've tested this bug, but I think that the size is implemented correct. @Yuan Sun: du shows disk usage, but the user should know the entire space ( more info http://serverfault.com/questions/290088/what-is-the-difference-between-du-h-and-ls-lh). So, it is better to test it using ls -la. The user needs to know the entire size, because this is the one used when you deploy the image on usb. @Belen: in the database you can find something like {\"file_name\": \"/media/ssd1/build/tmp/deploy/images/qemux86/core-image-minimal-qemux86-20140506091818.rootfs.ext3\", \"target\": 2, \"file_size\": 67108864} 67108864 is in bytes. To obtain MB, you have to divide it 2 times by 1024. So you obtain 64MB, which is correct. I will mark this bug as Not a bug, and if something is not OK please reopen it.
(In reply to comment #4) > I've tested this bug, but I think that the size is implemented correct. > > @Yuan Sun: du shows disk usage, but the user should know the entire space ( > more info > http://serverfault.com/questions/290088/what-is-the-difference-between-du-h- > and-ls-lh). So, it is better to test it using ls -la. The user needs to know > the entire size, because this is the one used when you deploy the image on > usb. Thanks, Cristiana. This should be explained in the interface using our help information bubbles. I will take this one. > > @Belen: in the database you can find something like {\"file_name\": > \"/media/ssd1/build/tmp/deploy/images/qemux86/core-image-minimal-qemux86- > 20140506091818.rootfs.ext3\", \"target\": 2, \"file_size\": 67108864} > > 67108864 is in bytes. To obtain MB, you have to divide it 2 times by 1024. > So you obtain 64MB, which is correct. > > I will mark this bug as Not a bug, and if something is not OK please reopen > it.
(In reply to comment #5) > (In reply to comment #4) > > I've tested this bug, but I think that the size is implemented correct. > > > > @Yuan Sun: du shows disk usage, but the user should know the entire space ( > > more info > > http://serverfault.com/questions/290088/what-is-the-difference-between-du-h- > > and-ls-lh). So, it is better to test it using ls -la. The user needs to know > > the entire size, because this is the one used when you deploy the image on > > usb. > > Thanks, Cristiana. This should be explained in the interface using our help > information bubbles. > > I will take this one. > On second thoughts, this might be overkill. Marking as 'not a bug'