"OS6" - a question or three

garyd9 wrote:

So far, I think the only really big problem (meaning I can't work around it) is the USB drive issue. One of the mechanisms we use for backing up involves portable USB hard drives, and that just doesn't seem to work at all. It's also unacceptable that the USB drive is automatically made into a share with all protocols and extremely liberal permissions. (Ideally, it wouldn't be shared at all unless I specifically configured it to be - its for backing up, not sharing!)

Following up on this, I decided to see what would happen if I plugged a USB HDD that wasn't partitioned or formatted into the USB port. I grabbed a USB drive that I used some time in the past for a windows backup (Win7's backup tool) and just deleted the partition from it using the windows disk management tool. I "ejected" it from windows and plugged it into the NAS. The result was... nothing. I plug it in, and absolutely nothing whatsoever happens.

Checking dmesg over ssh, I see the following:

usb 1-2: new high-speed USB device number 5 using ehci_hcd
scsi10 : usb-storage 1-2:1.0
input: Western Digital  External HDD     as /devices/pci0000:00/0000:00:1a.7/usb1/1-2/1-2:1.1/input/input5
generic-usb 0003:1058:0707.0005: input: USB HID v1.10 Device [Western Digital  External HDD    ] on usb-0000:00:1a.7-2/input1
scsi 10:0:0:0: Direct-Access     WD       PP III Studio II 0817 PQ: 0 ANSI: 0
sd 10:0:0:0: [sdf] 976773168 512-byte logical blocks: (500 GB/465 GiB)
sd 10:0:0:0: Attached scsi generic sg5 type 0
sd 10:0:0:0: [sdf] Write Protect is off
sd 10:0:0:0: [sdf] Mode Sense: 23 00 00 00
sd 10:0:0:0: [sdf] No Caching mode page found
sd 10:0:0:0: [sdf] Assuming drive cache: write through
sd 10:0:0:0: [sdf] No Caching mode page found
sd 10:0:0:0: [sdf] Assuming drive cache: write through
 sdf:
sd 10:0:0:0: [sdf] No Caching mode page found
sd 10:0:0:0: [sdf] Assuming drive cache: write through
sd 10:0:0:0: [sdf] Attached SCSI disk
root@NasPro:~#

So, at least the kernel see's the drive, even if the readynas stuff doesn't. (I get similar results, but with a different USB port number, if I plug it into a different USB port.)

Just to keep myself amused, I dropped into SSH on the NAS and created a partition on the drive via fdisk, type 83, full drive size. As soon as I wrote the table (w), I got an alert that an external device was connected. Here's where things start to get interesting. The next thing I knew, it appears that the NAS started doing something with the device. dmesg now has some more stuff in it:

ufsd: driver (8.6, Aug 20 2014 11:54:08, LBD=OFF, delalloc, acl, ioctl, ugm, rwm) loaded at ffffffffa0010000
NTFS (with native replay) support included
Hfs+/HfsX support included
optimized: speed
Build_for__Netgear_RN_R6_x86_k3.0.48_2012-11-26_U86_r193235_b3

ufsd: use builtin utf8 instead of kernel utf8
ufsd: sdf1 without journal

Now.. and this is really wild... notice above that I created a partition with type 83? In fact, here's the partition table:

root@NasPro:~# fdisk -l /dev/sdf

Disk /dev/sdf: 500.1 GB, 500107862016 bytes
81 heads, 63 sectors/track, 191411 cylinders, total 976773168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0xfc2d02b8

   Device Boot      Start         End      Blocks   Id  System
/dev/sdf1            2048   976773167   488385560   83  Linux

Well, after I wrote that table, it appears that the NAS mounted /dev/sdf1 with some kind of self-sensing driver that mounted, get this, the NTFS partition that was on the device before I deleted the partition table. (Kind of a neat treat, actually.) I guess this is possible in theory, but I have to be honest that I've never seen it done before. I wouldn't trust any of the data in the partition, but it still confused me to see it.

Oh, and the NAS mounted the filesystem that shouldn't exist, and automatically created a share for it with full access on both SMB and AFP :(

Okay, "eject" the drive from the web interface, and run mkfs.ext4 /dev/sdf1. Unplug the device and plug it back in. Device is mounted (again) and a share auto-created for it.. and I still can't backup to it directly. At least it was mounted as ext4 this time. ;) Backup jobs still don't work.

A question for @mdgm: Does backing up to the USB port devices work on a proper 516? The reason I'm asking is this is really a show stopper for me with the current firmware. However, it might only be a problem because of the hardware I'm running OS6 on. It simply might not be mapping the USB ports in the firmware to the proper device names in the kernel due to the hardware difference.