Re: USB key && NTFS

El día jueves, enero 18, 2024 a las 02:43:23 +1100, Ian Smith escribió:

> On 17 January 2024 5:16:27 pm AEDT, Matthias Apitz <guru@unixarea.de> wrote:
>  > El día martes, enero 16, 2024 a las 05:01:54 -0700, Gary Aitken
>  > escribió:
> 
>  > > As pointed out by Alexander, it's actually exFAT, not NTFS.  Since
>  > > you've already reformatted it, you're ok, but if you want to NOT
>  > > reformat a new one in the future, use mount.exfat.  manpage is
>  > > man mount.exfat-fuse
> 
>  > I haven't reformatted the key. I compiled the port
>  > sysutils/fusefs-exfat
>  > on my poudriere server, installed it and could mount the key fine.
> 
> Another advantage of exFAT is that you can read and write it natively from Android phones with an 'OTG' cable.

I do not have any Android phones, only a Linux phone and the USB key
mounts fine there on plug-in.

>  > I used f3write (from the ports) to check the capacity by writing
>  > 117 files of 1 GByte and to check the write performance which is very
>  > poor.
> 
> From your first post:
> 
>  > Jan 16 17:50:52 c720-1400094 kernel: da0: 40.000MB/s transfers
>  > Jan 16 17:50:52 c720-1400094 kernel: da0: 120000MB (245760000 512 byte sectors)

I checked older /var/log/messages and other external USB disks have
400.000MB/s. So it is not the port of the laptop.

I run on my Linux phone a lsusb command which shows:

$ lsusb -v -d 058f:6387
Bus 003 Device 002: ID 058f:6387 Alcor Micro Corp. Flash Drive
Device Descriptor:
  bLength                18
  bDescriptorType         1
  bcdUSB               2.10
  bDeviceClass            0
  bDeviceSubClass         0
  bDeviceProtocol         0
  bMaxPacketSize0        64
  idVendor           0x058f Alcor Micro Corp.
  idProduct          0x6387 Flash Drive
  bcdDevice            0.02
  iManufacturer           1 Generic
  iProduct                2 Mass Storage
  iSerial                 3 A430786F
  ...

I interpret 'bcdUSB  2.10' as an indicator of USB 2.10. The key is new
and its wrapping says USB 3.0.

> 
> So either the memstick is only USB2 and/or your port is only USB2; USB3 is ~10 times faster.
> 
>  > [guru@c720-1400094 ~]$ mkdir   /mnt/f3
>  > [guru@c720-1400094 ~]$ f3write /mnt/f3
>  > ..
>  > Creating file 115.h2w ... OK!
>  > Creating file 116.h2w ... OK!
>  > Creating file 117.h2w ... OK!
>  > Creating file 118.h2w ... Write failure: Input/output error
>  > 
>  > WARNING:
>  > The write error above may be due to your memory card overheating
>  > under constant, maximum write rate. You can test this hypothesis
>  > touching your memory card. If it is hot, you can try f3write
>  > again, once your card has cooled down, using parameter
>  > --max-write-rate=2048
>  > to limit the maximum write rate to 2MB/s, or another suitable rate.
>  > 
>  > Free space: 0.00 Byte
>  > Average writing speed: 9.27 MB/s
> 
> Looks like it just ran out of space; maybe you haven't accounted for space for a) MBR plus, b) the FAT and c) directories? (see dumpexfat below, and gpart).

Yes, it run out of space by intention. The tool f3write writes as much
as chunks of 1 GB until ENOSPC to check how much data really could be
written on the key (a tool against fraude).

Thanks

	matthias

-- 
Matthias Apitz, ✉ guru@unixarea.de, http://www.unixarea.de/ +49-176-38902045
Public GnuPG key: http://www.unixarea.de/key.pub

I am not at war with Russia.  Я не воюю с Россией.
Ich bin nicht im Krieg mit Russland.