NAS Server on Mede8er stops working randomly

[talib]

same here .. a bit different tho ..
my share always disappears on startup of the MED500x
i have to turn nas off / on again and it works ..
got core dump files on my hdd tho that come from the smbd process

and another thing i noticed .. after some reboots my internal hdd shows as sdf1 already .. i'm pretty sure the 1st time i started my box it was sda1 .. 2nd time sdb1 .. 3rd time sde1 and now it's sdf1 .. courious behavior .. but i guess it has nothing to do with the samba/NAS troubles

and a 3rd thing .. also nothing to do with the NAS problem .. but not really worth a new thread ..
the local date on the box always resets on reboot / standby and such .. and you have to set it
via date -s to get accurate dating on files .. else it's somewhat in feb.2000 .. would be nice to grab the time
from the net on boot if possible

i hope i'm not mixing up things too much in this post since some things are cout of the thread contex a bit 

[Insomniac]

and another thing i noticed .. after some reboots my internal hdd shows as sdf1 already .. i'm pretty sure the 1st time i started my box it was sda1 .. 2nd time sdb1 .. 3rd time sde1 and now it's sdf1 .. courious behavior .. but i guess it has nothing to do with the samba/NAS troubles

I have noted some strange behaviour whenever I boot with USB devices plugged in - the USBs are picked up first and are assigned earlier drive letters and the internal HDD is mounted afterwards. I'm of the opinion that this mixes up the assigned drive letters.
Are you booting with USBs plugged in?

Just to point out that the new generation of firmware based on the latest Realtek SDK appears to address this issue of erratic drive mounting.
The negative is that this will have an impact on backward compatibility and a decision was made to hold back on inclusion of new SDK until an assessment could be made on the implications of this change going forward.

the local date on the box always resets on reboot / standby and such .. and you have to set it
via date -s to get accurate dating on files .. else it's somewhat in feb.2000 .. would be nice to grab the time
from the net on boot if possible

We're currently testing time server lookups and it appears to be working.

[talib]

I have noted some strange behaviour whenever I boot with USB devices plugged in - the USBs are picked up first and are assigned earlier drive letters and the internal HDD is mounted afterwards. I'm of the opinion that this mixes up the assigned drive letters.
Are you booting with USBs plugged in?

nope no usb device plugged in .. and it keeps changing

Code:

mount |grep sd
/dev/scsi/host0/bus0/target0/lun0/part1 on /tmp/usbmounts/sdg1 type ufsd (rw,nodiratime)

executing mount shows also some strange mounts to the 3rd partition of my internal hdd (the 2nd partition is type Linux Swap) and yes i formated the hdd with the internal format function:

Code:

/dev/scsi/host0/bus0/target0/lun0/part3 on /tmp/hdd/root type ext3 (ro)
/dev/scsi/host0/bus0/target0/lun0/part3 on /tmp/hdd/root type ext3 (ro)
/dev/scsi/host0/bus0/target0/lun0/part3 on /tmp/hdd/root type ext3 (ro)
/dev/scsi/host0/bus0/target0/lun0/part3 on /tmp/hdd/root type ext3 (rw)
/dev/scsi/host0/bus0/target0/lun0/part3 on /tmp/hdd/root type ext3 (rw)
/dev/scsi/host0/bus0/target0/lun0/part3 on /tmp/hdd/root type ext3 (ro)
/dev/scsi/host0/bus0/target0/lun0/part3 on /tmp/hdd/root type ext3 (ro)
/dev/scsi/host0/bus0/target0/lun0/part3 on /tmp/hdd/root type ext3 (rw)
/dev/scsi/host0/bus0/target0/lun0/part3 on /tmp/hdd/root type ext3 (rw)

so i rebooted again to have a look at dmesg after the booting procedure
executing dmesg after boot shows:

Code:

EXT3-fs error (device sdh3): ext3_find_entry: reading directory #2 offset 0
Aborting journal on device sdh3.
......
.......
This is the current time: PowerUP_Done!!!
sata1: dev 0 cfg 49:2f00 82:746b 83:7f61 84:4123 85:7469 86:bc41 87:4123 88:007f
sata1: dev 0 ATA, max UDMA/133, 2930277168 sectors: lba48
  Vendor: ATA       Model: WDC WD15EADS-00P  Rev: 01.0
  Type:   Direct-Access                      ANSI SCSI revision: 05
SCSI device sdi: 2930277168 512-byte hdwr sectors (1500302 MB)
SCSI device sdi: drive cache: write back
SCSI device sdi: 2930277168 512-byte hdwr sectors (1500302 MB)
SCSI device sdi: drive cache: write back
 /dev/scsi/host0/bus0/target0/lun0: p1 p2 p3
data[0] 0x0 signature[0] 0xbabeface
data[1] 0x0 signature[1] 0xdeadbeef
Attached scsi disk sdi at scsi0, channel 0, id 0, lun 0
Attached scsi generic sg0 at scsi0, channel 0, id 0, lun 0,  type 0
mars_sata_work_fn(1038) hot plug end

still no other usb devices connected got no clue why it mounts the ext3 partition 8 times and strange that it changed to sdi after reboot

i also forgot .. i'm running the latest firmware (1.2.2a)


on the ext3 partition there is another samba folder similar to /usr/local/package/samba
just a guess .. but since i had ext3 errors in dmesg

Code:

EXT3-fs error (device sdh3): ext3_find_entry: reading directory #2 offset 0
Aborting journal on device sdh3.
scsi0 (0:0): rejecting I/O to dead device
printk: 359 messages suppressed.
Buffer I/O error on device sdh3, logical block 270
lost page write due to I/O error on sdh3
scsi0 (0:0): rejecting I/O to dead device
Buffer I/O error on device sdh3, logical block 1
lost page write due to I/O error on sdh3

i guess this might cause my samba core dump on startup of the box .. if it's started from there at all

*edit*
are there any plans to make the firmware opensource or on giving out a toolchain ? would be nice to be able to compile some goodies to play arround with

[phoenix73]

are there any plans to make the firmware opensource or on giving out a toolchain ? would be nice to be able to compile some goodies to play arround with

A big +1

I'm also experienced on Linux and cross compile, it will be fun to add some goodies to this little Linux device.
It's important, it will bring more functions to med500x and a greater user community

[talib]

after cleaning up the ext3 partition and formating it with ext.3 again .. the 1st boot looked good apart from samba again didn't work out of the box .. the usual core dump happened

apart from that my hdd is sda1 again and the ext3 is mounted only once and mounted rw
it also contains the samba and script folders again

so it should have something to to with that mysterious ext.3 partition and it's contents

[engelkie]

I recently upgraded to 1.2.2a...

Now after while the NAS goes down and I have to go and turn it off and on again to get it working. I didn't have this problem for a while now, but after the upgrade I have to do it at least once a day.

The other night I also received a funny message when I tried to turn the NAS on again: "HDD init failed" or something like that, displayed in big green letters on the screen. Even though I could still play movies from the internal drive. I had to unplug the Mede8er from the wall to fix it.

[utketser]

There are some SAMBA issues being picked up with by 64bit Win7 OS users - not so with 32bit users.
Internal tests have picked up issues with gigabit switches with respect to HD streaming.

Can you give out more details about this?
I have 64bit Win7 HP & also TPLink Gbit switch.

wbr,
Toomas

[talib]

so i got it working for now .. no more mountpath changing (sda1-sdb1-sdc1 and so on) and no more doubled mounts after reboots

even samba is working out of the box ... without crashing on boot

just remove the swap and ext3 partitions if you formated the hdd with the mede8er format option and it's fine
(just remove them .. don't resize the ntfs partition so you'll be able to add them later on without loosing data or such)

i guess some scripts mixed up partitions on boot somehow so you got this strange behaviors and samba core dumps

i'm fine with this for now (since i have nothing to run from the ext3 partition wich would need swapping and i don't wanna loose warranty by modifying the mtd blocks)
let's hope the next firmware will fix these things

[apotheker]

Hi,I also have some problems with the mede8ers network.I tried to copy files from my pc to mede8er using the mede8er and I have no problems,but when I use the NAS on option and copy files from my pc everything seems ok but the next day the files I transfered this way were no longer there,all empty .

[engelkie]

but the next day the files I transfered this way were no longer there,all empty .

Also had this problem... Lost about a 100GB of movies this way, because I was so confident, I moved them 

It happens when the NAS crashes on the Mede8er while you are copying... Seems that Windows keeps on copying.

[juwo1811]

I'd like to report that I'm also having issues with NAS..
my Mede8er  work in  NAS Mode, W7 64 Bit Home Premium, Gigabit  LAN. The same problem, at the moment I have no connection from my  workstation to mede8er. Reboot of my Mede8er and Workstation will also take no effect. I have the new Firmware.  Fix IP.  Yesterday it was no problem copy movie from my workstation to mede8er, I have no Idea, Ping is o.k.

[lblank]

Same problem here. Firmware version 1.2.2.a. After starting XP for copying files (32 bit) no connection can be made to the mede8er. Even if the workgroup is found the mede8er cannot be found. \\mede8er or \\<ipadress> gives the error 
"Network path cannot be found".
Pinging the mede8er is possible
After switching Nas OFF and On again everything is working fine. I use a fixed IP adres on a wired network.
Any solutions?

[Insomniac]

Hi guys,

The fixed IP on wired network would be my recommendation too.
I would have to advise against using the Windows Explorer shell drag and drop to move files around - especially if one moves one movie, then another while the first is copying, then another again, etc - this causes network collisions, bringing the network traffic to a standstill and can crash SAMBA.
I would rather recommend using a file manager which manages the file transfer process to queue file transfer so that only one file is transferred at a time. I use SuperCopier2, but there are several others too.

The new firmware includes an improved SDK and apparently much work on SAMBA - this is expected to be ready within about the next 2 weeks.

If you could in the meantime detail exact steps you have taken to duplicate these issues it would assists us greatly in testing this new firmware to be sure that the issue is resolved.
I must apologise for the issues you are experiencing and ask that you bear with us for the next two weeks while we test these improvements.

Your detailed feedback would be appreciated - please use the board sticky as a guide to providing such issue logs.

[lblank]

After some testing i think the problem is in the netbios name resolution.
Because after turning off and on Nas everything is working (for a while). It looks like the samba server is not reacting to netbios broadcasts.
Maybe the next link is helpful for your engineers to solve the problem. http://technet.microsoft.com/en-us/library/bb727013.aspx.

Goodluck and greetings from Holland.

[Scrooge]

I can report similar behavior.

Using Windows XP, Vox ADSL-Phone Router and the Mede8er WiFi dongle.

I switch the Mede8er into Standby(Red light) when not in use. When switching back on a day or two later the share is not available even though ping still works.

Unplugging the WiFi dongle and re inserting does not work. Switching Off and then On again the NAS in the setup brings it up immediately. Using static IP.