Beiträge von afremont

    I guess my point was that the odroid will actually transfer over the Ethernet at almost saturation speed.according to iperf. It will also do a good job of maximizing the transfer from a usb3 to SATA adapter, assuming the adapter is up to the job -- mine isn't when it comes to a good SSD. At any rate it, I believe that it massively outperforms a pi at being a headless server.

    Nope. Those ODROIDs that have USB3 implement also Gigabit Ethernet via USB3 (doesn't matter, just slightly higher CPU utilization compared to 'real' GbE). But the only good ODROID choices currently are HC1 and HC2 anyway (C1/C2 have all USB ports behind an internal USB hub and XU4 suffers from USB3 receptacle problems -- see my signature)


    Sent from my Moto G (4) using Tapatalk

    BTW, if you are planning to use this as a headless media server with external USB hard drive, I recommend looking at the odroid boxes. They have USB 3 and real gigabit ethernet interfaces.


    Sent from my Moto G (4) using Tapatalk

    All kinds of things can result in this. What software did you use to create your bootable flash card? It used to be that you needed specific boot images and kernels for each model of the pi. It seems that no longer applies to raspbian images from raspberrypi.org, but it might still apply to OMV. Some pi models don't like some brands/types of flash media. I'm just offering this up as food for thought.


    Sent from my Moto G (4) using Tapatalk

    RAID mirroring is what is, a potential life saver. It may be twice as likely to suffer a drive failure, but two drives failing simultaneously is pretty unlikely. I've seen it keep someone running after a drive failure more than once. It was a simple repair and didn't stop business on the spot. Well worth it in my opinion. You should still back up critical data.


    Sent from my Moto G (4) using Tapatalk

    cat /proc/filesystems and see if it's listed. If not, you might be out of luck, unless you can manually load a module for it. If it can't automatically identify it, you might be able to specify some extra arguments to the mount command. -t ifs perhaps.


    Sent from my Moto G (4) using Tapatalk

    Adding a drive to a raid 5 array means rewriting absolutely everything on the existing drives. I know you wanted the added space, but you really beat the original drives up pretty hard. Might want to think about adding a hot spare or two now.


    As for broken bones, I've had a few. I'm going on 56 now and 15 years ago I had a low speed motorcycle accident. I broke my finger, clavicle (sternum end, pretty rare), a pile of ribs and crushed my T7 vertebrae. Soft tissue injuries included a bone deep cut above my eye and nearly the complete severing of my ear and many more pulls and tears. Fortunately a plastic surgeon was available to fix my ear and they assigned a "sports doctor" to rehab me. That guy is completely merciless and has no soul, but I'm glad now that he made me "walk it off" and "work it out". If you met me, you'd never know any of that happened. But hey, I walked away and was going to ride my bike home, but it wasn't up to it. Adrenaline is like that.


    Sent from my Moto G (4) using Tapatalk

    That looks like a good way to do it. The rock 64 makes a good small NAS too. It has gigabit Ethernet and a USB 3 port. It's not quite that fast, but still very speedy using a decent usb3 SATA adapter.


    Sent from my Moto G (4) using Tapatalk

    I use this software on a raspberry pi, it transferres all files at 11MB/S, which is slow according to other posts on this forum. <X Pleasen can you help

    That doesn't sound very slow to me with that platform. Unless I'm mistaken, it only has USB2 and a 100 megabit per second network jack. Considering the Ethernet port is hanging on a USB bus, that's still maxing out the network interface. A Rock 64 would do a lot better since it has a real gigabit Ethernet and USB3.


    Sent from my Moto G (4) using Tapatalk

    I don't know about sqlite, but MySQL keeps its own user name and password information in a table named mysql. If the "regular" passwd command changed the one for MySQL, then that seems like an extension to the command. You usually have to run an update user query on the mysql table to change the password.


    Sent from my Moto G (4) using Tapatalk

    Some things to check:
    Are the computers in a "home group"?
    Did you set the workgroup name in omv to match what the Windows computers are set to?
    Did you give it all 30-40 minutes to allow the network browsers to sync up. Windows is kind of odd with the master browser and syncing the browse lists up. There are so-called "elections" to determine which box is going to be the master and computers may wait 20 minutes or so between broadcasting their presence on the network.



    Sent from my Moto G (4) using Tapatalk

    Well, this is also the symptom when the card is finally worn out (then the controller should set the card to read-only so that data can still be accessed but further data loss will be prevented). And with some of the counterfeit/fake SD cards it's the same once they exceed their real capacity.


    This (SD card dying) will happen to every SD card on this earth at some time unfortunately.

    Coincidentally, it marked it read only after erasing it. If I could get a datasheet on the specific flash chip, I'd see if there's any special/proprietary commands that I could send it to unlock it. I know a lot of people have had trouble with RPi corruption, but I don't know if it "ruined" their cards since I never had that happen to me, yet. ;)


    I don't think I've used this card very much, but I really can't be sure. smartctl won't access it because of the USB interface on the adapter. I have a new nuc7 that has a microSD socket. I'll boot it on Linux and see if I can get smartctl to tell me anything. I would like to see the statistical data, assuming it has SMART capability. I have no idea if any SD card has that though.


    It's possible that I bumped it when it was being manipulated by windows. I'm no fan of Windows, but I must confess that it generates a fair amount of business for me. Been a big Linux fan since the very early days of slackware and having to download floppy images.


    I'm inclined to believe that something went wrong during a critical time, rather than it just happened to be worn out. Anything is possible, but it sure is an odd coincidence that this was my first use of etcher (beta).


    Is there a specific place on the forum that i could discuss my slowish performance with iperf3 and hdparm? I'm getting about 250Mb/second on the Ethernet and a little less than 200MB/second using a USB3 to SATA adapter and a new Crucial 275GB SSD. Maybe I should take this up on the Pine64 forums since it's more Linux kernel related than OMV related.


    BTW, dd shows the contents to NOT be erased. I would expect it to be all 1s if it had actually been erased.


    Sent from my Moto G (4) using Tapatalk

    Just remembered that I fooled myself recently with an SD card to TF card adapter where the tiny read-only switch was set... but I would assume you already checked this?

    Oh yes, I've even tried two different adapters - one SD full size to micro SD and the other is a USB3 to micro SD. Somehow, it's internally in a read only state.


    Sent from my Moto G (4) using Tapatalk

    Since when does Windows deal with removable media correctly? AFAIK Windows still can only access the first partition of an SD card and if it can't cope with the filesystem there then silly error messages are spitten out.


    What about trying out 'SD card formatter' as already suggested here in this thread?

    I tried sdformatter and it didn't work either. I even downloaded the latest version. diskpart can't fix it either. Now to see what I can do with Linux.


    Sent from my Moto G (4) using Tapatalk

    I don't doubt that windows had something to do with this, but it is an interesting coincidence that it happened when trying to use etcher. Windows claims it is write protected, and diskpart acts like it can fix it, but doesn't. It shows up with an attribute of "Current Read-only State: Yes" followed by "Read-only : No".


    It's quite strange. I'll see if I can get Linux to do anything with it.


    Sent from my Moto G (4) using Tapatalk

    Interestingly enough, yesterday I ended up with a "current read only" bricked SD card when trying to burn a debian image to it for my rock 64 using etcher. Burned hundreds of cards before and never had this happen using win32diskimager. I'm still trying to unbrick it, hopefully one of the tools above will do it.


    Ended up here because I'm a new OMV user on the rock 64. It seems to be working aside from some iperf3 and hdparm -t performance issues. I'm using the version listed in the etcher beta I downloaded. No response needed on the performance things. If I can't live with them, I'll start a new thread. I just wanted to put my experience out there with killing an SD card.


    Thanks for OMV, it's a really nice package.


    Sent from my Moto G (4) using Tapatalk