ARGOSY - HD363N - Network Storage

rokitrik:

I don't have that problem, but the biggest I have used so far is a
Segate 80 Gb. The only thing I can think of to try is to check the
jumpers on the drive. According to reports, some drive brands work
best with Master set. Some work best with Cable Select set. I don't
think it is supposed to work at all with Slave set. Whether it is
currently Master or Cable Select, try the other setting and see if it
helps. I had my Segate 80 set as Master.

Please advise if the jumper settings help. I was planning to get a
Segate 160 myself to put in my NAS. It'd be a bummer if it turned out
to be incompatable.

Thanks Segbert

The diskdrive is a segate with the folowing charateristics:

Barracuda 7200.7

Model Number:ST3160021A
Capacity:160 GB
Speed:7200 rpm
Seek time:11.5 ms avg
Interface:Ultra ATA/100

The jumper setting was "cable select"
I switching to "master or single drive" but it did not solve the issue.


I was wondering if i had to put in on slave mode since it is not is a
primary drive. The NAS OS is part of the ROM, so the disk is basically
a slave. I will test this tonight

But the funny part is that when I go back to an old firmware version
(i.e. NetHDD004-0317) this issue is gone. But I get a lot of other bugs
back...

Coouple of questions:
- What are your jumbersettings with the segate 80GB drive?
- did you switched cables form the NAS to the drive. I read some where
that it solved another is issue by replacing the capel with a 80-40
wire instead of a 40-40
- Do you know if there are hidden files created on the drive, while
formatting the drive using the build in format. Because it does create
a public folder. so it could also create more files/dis.

thansk
ReinJan

I've used a Seagate 160GB and it worked fine...until I took the drive
out, put it in another computer to restore some data, and that
computer fried the drive.

Anyways, I then put a Samsung 160GB in it, and it works fine. Seeing
how Tritton sells them with 160GB drives, any 160GB should work--I
don't know what brand they use though.

I have the Segate 80 in my Linux box now, but when it first arrived, I
just made sure it was jumpered to Master and popped it into the NAS. I
used the cable that came with the NAS. When you format it in the NAS,
it creates the Public directory at the root level of the drive. It
also creates two files at the root level, whose name I don't
remember. I assume these retain the NAS settings.

When you format in the NAS, it only takes a few seconds, so it could
not be really formatting the entire drive. Perhaps the boot block, or
file structure is corrupted on it. Maybe it was formerly a Linux ext
drive and confuses the NAS format. You could try putting it in a
Windows PC and using something like Partition Magic, or some other
such utility to check and really reformat the entire drive.

Good luck

Problem: Formatting a QUANTUM 20 MB (a few years old), tried
formatting with FAT32 before use, also removing partition info,
nothing worked.

Solution: Used a newer 120 GB Western Digital drive, removed the
Partition info, put into the HD363N, told it to format and the format
was almost instant.

Noise: Some say the drive is "noisy" -- it's the fan. Could probably
reduce that noise with a resistor in series with the fan.

Sleep Timer - Fan on/off: Some mentioned that the fan never goes off
even after the sleep timer expires (5 or 10 minute timer you set to
spins down the drive) - maybe that was the case at one time, but the
firmware 1013 stops the fan when it stops the drive.

SetEOF Problem: I saw one problem reporti not recognizing the SETEOF -
that's a programming issue that has no bearing on using the drive as
file storage and retrieval.

I have had no trouble editing files that reside on the drive, and they
have to open and close the files and seem to do so without problems.

------------

So, bottom line, use an up-to-date drive to avoid formatting
problems.

I like the ability to set the FTP port being used, which means I can
have an FTP server listening on port 21 (my Buffalo-Tech Terrabyte
device) and on port 2100 (where I have this HD363N drive).

Hal

SetEOF Problem: I saw one problem reporti not recognizing the SETEOF
- that's a programming issue that has no bearing on using the drive as
file storage and retrieval.

"Programming" issue? Indeed so, and one in which the firmware fails
to properly implement a fundamental SMB operation. If you use the
device as a file server with anything but trivial client software it
is also one that'll bite you sooner or later.

Those who can get by treating the NAS as some sort of dumb
file-by-file backup device will probably never experience it. If you
try to use it as an actual file server though you're just a heartbeat
away from a failure. Any software that updates a file in place has
the potential to trip over this bug. Programs that perform file
updating using simulated "tape to tape" operations creating a whole
new file on each update will be immune. Reports suggest Win9x
versions of NotePad work that way.

I would not be surprised to see file archivers or other more
sophisticated backup utilities fail for this reason.

This is a real bug, and seriously limits the use of this product
outside hobbyist scenarios. I know it has kept my units benched
instead of being deployed. If they fail with NT4/Win2K/XP NotePad I
can't begin to trust them in the wild.

That said, in other ways the product performs quite well. I have
about 1200 hours of continuous operation on one unit without a
restart.

As dilettante wrote ""Programming" issue? Indeed so, and one in which
the firmware fails to properly implement a fundamental SMB operation.
If you use the device as a file server with anything but trivial
client software it is also one that'll bite you sooner or later. "

dilettante also wrote "This is a real bug, and seriously limits the
use of this product outside hobbyist scenarios. I know it has kept my
units benched instead of being deployed. If they fail with
NT4/Win2K/XP NotePad I can't begin to trust them in the wild. "

Perhaps an example of the failure using notepad would be useful.

Clearly nobody wants a hidden failure lurking that will cause lost
data. I've experienced network errors with bad NIC cards as well,
which might point to some other hardware issue somewhere, not
necessarily the firmware in the unit itself. The jury may still be
out ??

So far, having saved numerous files, edited files, downloaded files,
etc., to the drive, and using it as an FTP server for a while, it
seems to work fine. Also, it has a relatively unique ability to allow
both Anonymous FTP and user account/password FTP. For the price, it's
got a lot of features.

If the SetEOF issue is indeed narrowed down to the firmware, and not
some other potential hardware issue, it would be nice to have a way
to duplicate the problem and get some other users to verify it. So
far, I think (and I might be wrong) the only person reporting this
"bug" is dilettante. As they say, one Robin doesn't make it spring.
But two or more might be indicative of something

Hal

I have independently replicated dilletante's SET EOF bug.

When using W98se and notpad I cannot replicate the problem.
When using Linux and KWrite I cannot replicate the problem
When using W2k and notepad, I CAN SOLIDLY replicate the problem.
I have not tried XP.

When I get a chance, I will boot up a Live Linux CD in the W2k machine
and see if it replicates using Linux through the same hardware.

HalRogers,
Please see [b:3de617d56d]dilettante[/b:3de617d56d]'s post on Sun Sep
11, 2005, in this thread, in which he clearly describes how to
reproduce the end-of-file error.

In general:
I've continued to test for this issue with other editors. To date,
the only other editor that can produce the same result is Open
Office's Writer. If I edit a text file, not in Open Office format
but just text, the same way as is suggested in Notepad, I get the
same result. But, if I edit a text file with other editors, I am not
able to produced the same outcome ... the files look fine.

To be best of knowledge, this issue was reported to both Argosy and
Tritton, and subsequently confirmed by both, before the 6-1013
firmware was released. Thus, I believe they have not yet found the
root cause.

segbert,
When you run Linux, perhaps you could run Open Office's Writer to edit
the text file (remember to save the file in text format, not Open
Office's file formats) and see if the error occurs. Maybe you should
do it running Open Office in Windows first, just so you know you can
create the error with Open Office.
At this exact momment, I'm not sure if I was running Open Office 1.5
or 2.0 when I tested ... but, I'm guessing it was 1.5.