OMV 3 did not pick up on actual size of SDcard in RaspberryPi

  • Installed OMV 3 beta image on microSDcard (128GB, SanDisk, UHS-I), booted it with RaspBerryPi 3B, did apt-get update, upgrade, dist-upgrade and then ended up with an up to date system; Erasmus 3.0.24


    In OMV web admin under Storage > Physical disks
    it actually says the disk is 119.08GB so that should be OK.
    However, under File systems this is what it shows:


    and I can't resize or do anything (gui doesn't respond to actions, only mounting works). Rather disappointing, but hey, it's a beta. I'm inclined to do a RPi debian install image, and then maybe try OMV on top of it, because this didn't happen with a debian install on the Pi (had it on this Pi before).

    • Offizieller Beitrag

    Chill out. I wrote a script to resize the third partition but obviously something didn't work. The normal OMV installer uses the entire OS drive for OS purposes and you would have a lot more work to do to use it for data.... Plus OMV 3.x is still beta as you mentioned. Is a beta release allowed to have issues?


    What is the output of cat /etc/rc.local

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    omv-extras.org plugins source code and issue tracker - github - changelogs


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!



  • So, I'm guessing there needs to be something else in there? Yes, I know it's a beta, and I first installed v2 Stable, but it is running such old wheezy stuff I can't have the OS do what I want, separate from OMV I mean. Plus I got an error when I tried to update the btsync repo (I really want OMV's multi-instance syncthing on this Pi).

    • Offizieller Beitrag

    raspi-config isn't installed on the image.


    The script ran because the entry in rc.local to execute it was gone. The resize script remove that entry from rc.local.


    Just execute the script - resize_mmcp3.sh - that is still in the /root/ directory. Let me know what kind of output you get. It might need to be tweaked for OMV 3. I didn't test it because it kills my SD cards due to the number of times it would run.

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    omv-extras.org plugins source code and issue tracker - github - changelogs


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!

  • Had already tried fdisk and raspi-config (I always first run apt-get install bash whereis bind-tools mc raspi-config ll to get me some crucial shell stuff),
    in raspi-config Expanding filesystem didn't work, saying


    "mmcblk0p2 is not the last partition. Don't know how to expand"


    and here's the output of the resize script


    I'm betting it has to do with the stupid UHS-I on the microSDcard. Never had any luck with UHS-I cards for raspberry pi systems.

    • Offizieller Beitrag

    Pretty sure bash and whereis are already installed on the image and I really wouldn't install raspi-config...


    And you are probably right, that card is most likely causing the issue. I don't have any cards anywhere near that size to test but it looks like it isn't supported well. Why not boot gparted-live on the system you wrote the sd card with and resize it?

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    omv-extras.org plugins source code and issue tracker - github - changelogs


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!

  • Why not boot gparted-live on the system you wrote the sd card with and resize it?


    Thought about that. I actually formatted it using SDformatter on Windows 7, then made it FAT32 using this tool and then used the recommended Win32imagewriting tool to write the OMV img to the sdcard.
    (By the way, strangely I do recall that the OMV 2 did show the full size in its filesystem..)
    Running a gparted live is a bit of a gamble on the laptop with the sdcard slot, not sure if it even sees the sdcard then, but I guess it's my only way out here.


    I went from the incorrect compatibility list for raspberry pi 3 B, which stated that 128 GB UHS 1 microSDXC cards are fine. I doubt that they are, it's the third one that has issues for me.

    • Offizieller Beitrag

    Why did you use sdformatter and make it FAT32? All you need to do is use win32diskimager. It writes everything including boot record, partitions, etc.


    If gparted-live isn't new enough, then use systemrescuecd 4.7.3 and select the alternate kernel in the boot menu. It should work on just about any hardware. Once it boots, type startx and then gparted is a icon on the bottom left. I recommend using linux live usb creator to write systemrescuecd to a usb stick. Don't worry when it says it can't match the exact version of systemrescuecd.

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    omv-extras.org plugins source code and issue tracker - github - changelogs


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!

  • Why did you use sdformatter and make it FAT32? All you need to do is use win32diskimager. It writes everything including boot record, partitions, etc.

    Yes, I would expect that too, and did that the first try. It left a weird unknown partition at the start of the sdcard, could not remove it, or re-write it using Win32diskimager. The errors lead me to a manual on RPi forums telling me to pre-format to FAT32 to get rid of that useless storage waste at the start of the SDXC card. Which is probably also why it's now not working as expected. These are strange cards, obviously, and need special treatment.


    I've used gparted live-disks/sticks about weekly for the past 10 years at work, but thanks anyway. :)

    • Offizieller Beitrag

    Strange... Other than wireless adapters and winmodems, I have yet to find something systemrescuecd doesn't work with since it based on bleeding-edge gentoo. But oh well.

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    omv-extras.org plugins source code and issue tracker - github - changelogs


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!

  • I'm running into the same reported issues when using OMV 2.2.5 (Stone Burner) on a Raspberry PI 2. After firing the script manually, I'm getting the following output:



    Anyting what I can do here to manually to rectify the problem of removing the /dev/mmcblk0p3 and extend the FS /dev/mmcblk0p1 ??

    • Offizieller Beitrag

    Anyting what I can do here to manually to rectify the problem of removing the /dev/mmcblk0p3 and extend the FS /dev/mmcblk0p1 ??

    The script doesn't remove the third partition and extend the OS partition. It extends the third partition to use the rest of the SD card since I create an image that will work on SD cards 4GB and bigger.


    The script should only be run after first installing. If you make changes, you don't want to use the script.

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    omv-extras.org plugins source code and issue tracker - github - changelogs


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!

    • Offizieller Beitrag

    Odd that you would increase the size of the first partition... You can't use that for data storage from the web interface. It will also shorten the life of the card if you use it for other things.

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    omv-extras.org plugins source code and issue tracker - github - changelogs


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!

  • OK, I went ahead and flashed the OMV 3.0.24 (Erasmus) image on a 32GB SD card.


    After the 1st run, it seems that a 3rd partition is being created with the "resize_mmcp3.sh" script, however it seems that it doesn't use the entire remainig disk space of the SD card, despite an additional reboot of the box.


    Here is the output of my system queries on the shell prompt:

    It only shows 58M instead of 25G pertaining to the partition size :(


    So I went ahead and fired the following commands to delete the 3rd partition and recreate it again not using:


    Now, If you pay attention when recreating the 3rd partition, it seems that "fdisk is using some default (and incorrect!) values that do not make senss, and I have to key in the correct values for the starting sector.


    Once, I have rebooted the box, I run the command "resize2fs /dev/mmcblk0p3" and the mounted 3rd partition shows its correct size:


    Unless I'm mistaken, to me it seems a bug with the "fdisk" command...

    • Offizieller Beitrag

    fdisk is updated in OMV 3.x. So, I probably need to update my script. I disable it when testing images because it is a lot of wear & tear on my SD cards.

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    omv-extras.org plugins source code and issue tracker - github - changelogs


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!

  • Unless it wasn't evident from the output quoted in my previous post, instead of taking the default values printed during the "fdisk create partition" prompt, it seems to me to be fine if parsing out the real values from the "fdisk" command output, and then input them within the script.


    Other people have similar challenges as described in the "Resized SD" article...

    • Offizieller Beitrag

    I guess I don't know what you are trying to say. Have you looked at the script? The only time it takes a default is when it is selecting the ending sector which should be fine.

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    omv-extras.org plugins source code and issue tracker - github - changelogs


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!

  • OK, let me do a walkthrough again from the "fdisk" output in my original posting by stripping the irrelevant info and walk step by step what I've been able to observe:


    When firing "fidsk" on the shell prompt, and am printing the disk info, I'm getting:


    Disk /dev/mmcblk0: 28.8 GiB, 30908350464 bytes, 60367872 sectors
    ...
    Device Boot Start End Sectors Size Id Type
    /dev/mmcblk0p1 8192 122879 114688 56M c W95 FAT32 (LBA)
    /dev/mmcblk0p2 122880 7028735 6905856 3.3G 83 Linux
    /dev/mmcblk0p3 7028736 7159807 131072 64M 83 Linux


    I deteted the 3rd partition in my list (last line above), and recreate it:


    First sector (2048-60367871, default 2048): 7028736
    Last sector, +sectors or +size{K,M,G,T,P} (7028736-60367871, default 60367871):
    Created a new partition 3 of type 'Linux' and of size 25.4 GiB.


    And here the situation when recreating the 3rd partition:

    • Instead of accepting the default value 2048 for the 1st sector, I have to type 7028736.
    • For the last sector, I keep the default proposed value.

    I'll check your script to which extent it takes into account the correct value for the 1st sector. However, confirming as input the default value for the 1st sector will destroy the contents of the SD card, and the device will not be able to boot up (that's what happened to me).

Jetzt mitmachen!

Sie haben noch kein Benutzerkonto auf unserer Seite? Registrieren Sie sich kostenlos und nehmen Sie an unserer Community teil!