Installer live in action

Here is a little life sign from OMV. The newest video shows you the installer live in action. After a short installation phase OMV can be used out-of-the-box.

Note, OMV uses the whole harddisk selected during the installation for the system OS. So if you’ll use a 1 TB volume for that you’ll waste 999 GiB of this disk for nothing. That’s fact, nothing will be changed in that way. No stupid OS/data partition solution as in FreeNAS will be supported, instead use a small HDD (maybe an older one) or better a SSD/DOM/whatever. Currently the OS requires about ~250MB in summary, but this may increase in future.

74 thoughts on “Installer live in action

  1. I disagree on flash (compact or usb)as boot device. I’ve found them to be far less reliable then a ssd. Count me in for ssd and if you like flash, you go do that. JFYI, you can get a good model ssd for like 69 dollars now in the US. This space issue thing is moronic. I think some of you are on wrong place. It will be nice to have some extra space to add more debian packages post install. Just reading these posts it makes me see why Volker almost quit the project. So many demanding everything under the sun and then the idiot posts, omfg. And I’ll say it. I know others may have been thinking it. J Smith you are a frickin tool.

  2. USB Flash drive is the way to go. It will load faster and is less likely to fail due to no moving parts, less noise. There really isn’t a downside I can see here.

    Volker: again thanks so much for all your efforts and I can’t wait to see this finished.

  3. I like the idea of a usb stick, however i also understand the headaches that causes on developing a seperate install routine. On a side note im planning a new storage server for when OMV is released.. in the interest of saving cost i was planning on using internal SATA multipliers combined with md raid …. does anyone have any experience in this arena with multipliers?

  4. I think I speak for all of us when I say, I can’t wait to test OMV. On a side note, I’m wondering how OMV will act in a VM on an Ubuntu server. I’m assuming we’ll be able to allocate ~500mb for the OS on the main drive, and then mount other dedicated drives for data storage. Even though you probably won’t support running OMV in a VM permanently, I think you should keep the possibility in mind.

    Keep up the great work!

    • For your info, FreeNAS was also developed by myself, so i think i can say that.

      Perhaps you find this stupid, but to me, its more stupid waisting space.

      As said, if the user is using a 1TB disk for the root OS this is true, but also a stupid decision by the user. If you do not like this design decision of OMV you are free to use any other product.

    • This is a NAS, not a standard OS. If the waste of space is an issue, you could just install it to a USB stick and boot from that.

    • I agree with Volker and Antworten – install to something else, i.e. USB stick of some variety. Keep your OS off the data drives. Save those precious IDE/SATA ports for something useful.

  5. I run my ESXi server with a small SSD. The server runs awesome with that as the operating system drive, addtional disks added for the datastores. I think a small SSD is the way to go with OMV. As for filesystem I’m hoping for ext4 but would be fine with ext3. Thanks for the update Volker. Was pretty quiet there for a bit and we missed you. Hope you are having a great summer. : )

    • Dsds = ‘Deutschland sucht den Superstar’ 😉

      Hab jetzt mal Debian stable auf dem alten 800MHz Thin Client installiert und siehe da, die Samba Transferrate ist VIEL besser als unter FreeNAS, HD-Video geht jetzt. Allerdings vermisse ich schon ein bißchen die einfache Web-Administration von FrreNAS. 🙂
      Mir ist noch ein Grund eingefallen, wieso es OMV ‘braucht’; falls so eine alte Kiste den Geist aufgibt kann man die Festplatte einfach in ein externes Gehäuse knallen und über einen Linux-Rechner leicht darauf zugreifen, ist nicht so leicht bei einem FreeBSD Dateisystem, wie es FreeNAS benutzt…

  6. Den richtigen Medienserver zu finden hatte ich mir auch leichter vorgestellt, hab jetzt auf meinem Athlon64 Ubuntu Server PS3MediaServer am Laufen, funktioniert ganz gut, ist aber weil auf Java basierend nichts fuer allzu schwachbrüstige Maschinen. Allerdings kann mein WDTV-Live praktisch alles spielen und braucht kein Transkodieren und kann auch auf Samba Shares zugreifen, also braucht der nicht wirklich einen MediaServer.
    Warum ich mich jetzt noch fuer OpenMediaVault interessiere? 🙂 Ich hab noch einen alten Thin-Client unter FreeNAS laufen, aber leider ist unter Samba die Datentransfer-Rate zu gering, um HD-Videos abzuspielen (nur 1-1,5MByte/s). Fuppes bringt da auch nichts, obwohl über FTP 6-7MByte/s drin sind. Da Samba ja unter FreeBSD bekanntermassen nicht so gut läuft hoffe ich, daß OMV da besser ist.
    Also weiterhin frohes Schaffen!
    Grüsse, signorRossi.

  7. Bin mit meinem FreeNAS System am kämpfen um den richtigen UPNP-Server zu finden. Keine Installation funktioniert out-of-the-box (mediatomb, ushare). Mein Wunschserver ist Twonkymedia, aber das tu ich mir auf FreeBSD nicht an.
    Jetzt überleg’ ich mir echt zu warten, bis OMV verfügbar ist (auf Linux ist das alles kein Problem).

    PS: OMV ist vom Konzept her ein Volltreffer (FreeNAS-Replacement auf Linux-Basis!)
    Ich wünsch dir viel Energie!!

    • Leider gibt es unter Linux auch keinen gescheiten UPNP server out-of-the-box im Debian repository. Man muss also immer selbst Hand anlegen, oder gibts da mittlerweile was. Coherence gibts as deb Paket, aber ich hab das ganze sieht ziemlich übel/kompliziert von der Konfiguration her aus. Hab mich aber auch nicht mehr sehr damit beschäftigt da ich kein UPNP verwende, mir reicht DAAP. Von daher soll die Community das machen wenn sie sowas braucht, muss ja nicht alles selber machen.

  8. How about liveCD and web-ui based installer?
    This might have security problem but this is good for display-less devices.

    • No WebUI installer will be available, the Debian installer will be used only. LiveCD will be supported as far as i did not see any problems, but i do not actively maintain this, its only a rubbish of the live-helper framework used to create the ISO image.

  9. Sehr geil!
    Da es ja ein vollständiges Debian System ist, denke ich, ist es auch möglich, eigene Dienste zu installieren, oder?

    • Deshalb mach ich mir ja die ganze Mühe 🙂 Man kann alles installieren was man will, es gibt dann halt keine WebUI integration, aber das ist ja nicht umbedingt schlimm, es gibt ja noch die gute alte Konsole.

  10. Volker, will you be implementing some tweak so you can install omv on USB/CF card? so the writes will not break the USB/CF card?

  11. Thank you for the update Volker. I check on the progress of OpenMediaVault regularly and just wanted to let you know that i appreciate your fine work. My Nas machine is chugging along waiting patiently for that special day.

    I have been following you without mentioning it as I am sure a lot of people are. Here is a great reminder:

    Every person that has an issue with a product will complain about it whilst out of the ones that are happy with it, only a fraction of a percent will write to the manufacturer and comment on it.

    This is the same as for every positive reply you get, there are hundreds of positive ones that don’t write. The negative comments you get are from every single person that is negative.

    Take that into consideration, remind yourself that there are a SHITLOAD of people out there that admire what you are doing and keep up the good work!

    On another note, what are your thought on PulsarOS as it seems to be your main competitor with apporxiamtely the same features, modular design etc?

    • On another note, what are your thought on PulsarOS as it seems to be your main competitor with apporxiamtely the same features, modular design etc?

      There is no competition 🙂 Nevertheless, what shall i say, progress seems to be more slower than in OMV. But, don’t call me selfish, OMV will make the race.

      • Yup, you seem more qualified which gives your os an advantage in my eyes. However, I can’t help to compare them. There really arent that different from what I can notice, sure the measures to reach features might be different but the end results seem pretty similar. Just a few questions though:

        Will you be implementing any kind of btrfs or similar filesystem, ie “next gen” filesystem? i know that you will be using the older debian buthow hard wwould you think that it would be to implement once btrfs is fully finished? I know that you might intend for others to add the feature but since I have little knowledge in the area i was wondering if you could give us your opinion, is it possible or even remotely likely that we might se such a feature later on?

        What would you say are the advantages/disadvantages of running the os loaded into ram?

        Thx again, remember, behind every negative comment there is a person. Behind every positive comment there are 100 since they tend not to speak up as easily -a wellkown principle.

        • You can be certain to finish yours ahead of theirs btw since they are basing theirs on newer unstable debian. This means stabilityissues until it is done but more features once it is finished I guess.

          Really hope for someone to add btrfs at some point.

          Looking forward to trying OMV once it is released (even a beta would be nice :))

        • What would you say are the advantages/disadvantages of running the os loaded into ram?

          On of the great disadvantages is that it is a little bit more difficult to add plugin support. One advantage is that it does not require a spinning disc which results in a quiet PC.
          OMV has a Live mode but this is only a side-effect of the framework used to build the install images and is not supported or maintained by myself. I think i will remove it in the final product to get not bothered by useless questions and bug reports. OMV requires a full installation on a HDD/SSD/USB and does not run in a RAM disk.

          Currently i’m working on upgrading to Debian Squeeze but the framework (live-helper) has some bugs till now so i think it will take some time. Also some packages in Squeeze break the build process, e.g. mdadm (see
          With Squeeze BTRFS support can be included into OMV, but right at the moment it does not replace mdadm because BTRFS does not support RAID5/6 right at the moment. I’m also not sure about the stability and quality which is required by a data store. So EXT3/4 or JFS/XFS is the better choice at the moment in my eyes.
          Give BTRFS some years to grow and get the quality all the other filesystems have right now. Having a f**king cool and hyped filesystem which destroyes your data is not what you want.

          • That is kind of my question. In “some years” when debian squeeze and hopefully btrfs is fully stable, do you think it will be hard to impement btrfs. Even perhaps not fully stable until sid is released, what are the chances of succesfully adding it to your lenny without having to do so many bad solutions which might compromise the integrity of such a, by then, nice filesystem?

            I will most likely not care much about the ramissue since I will have some 6ish drives nearby, so sound ain’t an issue. Most likely a cf-to-ide drive anyways to save sataports. How intense will the read/write be to the osdisk?

            Running os in ram is not really a guarantee of low osdisk usage, just reads and not writes, so will have to test this with both of your os’ eventually

            Yay, I got one of the silent ones to speak up, kinda proving my point!

            • Implementing BTRFS will not be more difficult than implementing the rest of OMV. But the seemless integration into the WebUI and the existing infrastructure will be more difficult. The best would be to replace the mdadm backend completely with BTRFS, but i do not think that the users will not be happy about that.

              How intense will the read/write be to the osdisk?

              Not more as a usual installation. Maybe the RRD tools will increase it, but i did not measure it till now.

  13. I can’t believe it, volker is back 😉

    good to hear something. i really have had fears about projekt. But all seems to be ok 😛

    Good work volker

  14. Could this mean it is somewhere near a release state ? I’m looking forward to install it in my home server.
    //thoughts about using flattr on your homepage (at least when it’s released ?

Leave a Comment