Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[DementedBuddha]

Hi rws8258,

I tried your approach a few times without success (note that my setup procedure was perhaps different than some peoples as I have twonky installed, which requires a different approach ).  A usual first strategy is to redo from scratch so long as it doesn't take too much time.  I kept getting the problem so I investigated further.  I'm really curious as to exactly why the LEDs are blinking purple/pink and why dlink support staff can't answer this question.  There was one theory that this may be a blend of blue and amber and that's why they'd never seen it.  I'm no color expert so I couldn't say, but it sounds possible I suppose.

I did some experiments and I'm pretty sure the root filesystem getting filled up can cause havoc with your system, including all of the symptoms mentioned.  Installing Twonky is a non-standard thing to do so you kind of expect to come across problems like this and have to troubleshoot them.  I fixed the issue by putting the Twonky datafiles elsewhere and everything seems very stable now.

I'm curious if anyone has had the pink/purple LED problem with the standard setup on the DNS-323.  Is there anyone on this thread that's had the problem without installing any "other" software?

[ECF]

I have actually seen it before a few time but simply rebooting the unit resolved the LED issue. It seemed to just happen out of nowhere and remember that the unit has not been rebooted for quite some time. I do not know the cause of these incidents.

[rws8258]

Well, I assumed the reason for the pink or purple (or, more correctly, magenta; and since I've also heard here about white) was simply the result of "sequencing" the input(s) to the LED, thereby giving the "appearance" of a different color. In the case of RGB LEDs, blue gives blue, red gives red, and green gives green. But, when you sequence two or more of these input signals (at 25 ms, or faster), the naked eye "sees"  yellow, cyan, magenta, or white.

So, since we know what red means, and we know what blue means... Well, there ya' go.

[jdawg]

For what it's worth, I just started getting the pink lights as well.. running F/W 1.5 with 2x 300 GB Maxtors in JBOD.  I think this is a firmware issue - the only thing I've changed recently is the firmware, otherwise, things have been running great - large files, used for backup with ViceVersa pro.  Again, no issues until recently.  Rebooting solves the issue for a day or two.

[rws8258]

BTW... No sarcasm intended; it's just that I haven't bothered searching around for a detailed schematic, etc.

Even so, we know this much... The circuit that drives the HDD Activity LEDs is capable of "sending" pink to the LED(s), otherwise we wouldn't see it. Is this a bug? I don't think so.

So, the question... brings us back to Matt's original post... Are there any similarities in the problems being reported when pink is displayed? Or, what conditions must be present to cause pink?

[sylaan]

I received pink once when twonky filled up the root partition so I think I that is one thing that triggers it.

However, after setting up the right symlinks, twonky never filled up the root partition again and I still got the pink lights. It happened once when I wanted to added my music folder to twonky. I have many thousands of MP3's there and twonky just choked I think, started trashing the hard drives in the NAS and I had to kill the process eventually. That's when the pink light appeared even though root filesystem was not full. The NAS was fully functional though. A reboot took care of the pink lights.

[jdawg]

I did a quick "restore to factory settings" and my DNS-323 seems to be working again.    I must have changed something in the configuration - I've been seeing a lot of posts on root directories for uPnP/iTunes functions.  I was playing around with this right before I started receiving the error, so this is a high probability root cause.  Just remember, if you do this, it also resets the default logon info (admin, blank password.)

[hsimp]

I had the Pink/Purple light on one drive only.  Testing showed that the partitions were the problem.  we repartitioned and formatted the drive.  I'll be reinstalling it later.

[AKT]

I am having the same problem with my DLink 323... I have two Seagate 500GB drives which never exhibited the problem... Then I upgraded the software a couple time and also changed internet service provider (mention this only because Email does not work any more).... I use the drives for a period of time then the LEDs go Pink/Purple and the drives are not accessible from the network... I would then have to reboot the system and it would work well for a period of time (not sure how long... long enough to transfer a few hundred MB).
I hope DLink would post what is the meaning of the Pink/Purple LEDs... are they the same as the amber described in the manual.

[fordem]

BTW... No sarcasm intended; it's just that I haven't bothered searching around for a detailed schematic, etc.

Even so, we know this much... The circuit that drives the HDD Activity LEDs is capable of "sending" pink to the LED(s), otherwise we wouldn't see it. Is this a bug? I don't think so.

So, the question... brings us back to Matt's original post... Are there any similarities in the problems being reported when pink is displayed? Or, what conditions must be present to cause pink?

The intent (as documented) behind the logic that drives the LEDs is to "send" either blue or amber - pink is not programmed, but results if both blue & amber are displayed together.

Is that the result of an undocumented, undesired logic state - ie - a bug?  That's actually the definition of a bug.

AKT - no - the pink/purple LEDs is definitely NOT the same as the amber described in the manual.

[rws8258]

Is that the result of an undocumented, undesired logic state - ie - a bug?  That's actually the definition of a bug.

Uh, Maybe I need to clarify my point... From what I've read here, it seems that when "pink" is being displayed, the hardware is functioning perfectly, because it is doing exactly what it is being told to do; and so, it's not a hardware bug.

HOWEVER, when the firmware/software tells the hardware to display red+blue (i.e., at the same time), you get magenta (pink, purple, not amber). This is a programming error, or "bug injection"... So, ultimately, as you said, it ends up being a "bug" to you and me, the consumer, the end user. And no, we don't like it (nor should we) because it's not documented and interferes with normal operation of the LEDs.

More importantly, this "bug" has been around long enough for D-Link to have done something about it. They need to take this bug and turn it around into a usable function, or defeat the conditions that could cause the event... without disabling normal operation of the LEDs, of course.

[fordem]

Could you run that by me again - what is a hardware bug?

At this point - I don't think anyone knows definitvely what causes the pink/purple/white LED indication, making it impossible to state that it is NOT or can NOT be caused by a hardware failure.  I myself am not certain whether pink/purple/white are just personal interpretations of the same color displayed, or actually three different displays where the duty cycle of the signal driving the LEDs is different.

An inability to simulate or replicate the failure mode causes an inability to locate the problem - from anecdotal reports that I have seen, some of the failures have occured with combinations of specific disks & firmware, others have resulted from the use of unsanctioned, third party 'hacks', and you want D-Link to do something about it - they have a responsibility to deal with the first of this situations, but none for the second.

You hacked, you broke it - you fix it.

For the record, the one time I have seen a pink/purple/white LED indication it was triggered by removing a disk, which is a simulation of a hardware (ie. a disk) failure - how does that fit in with your theory?

[unnameable]

This problem occurred to my DNS-323 this morning at 04:12AM.
Yesterday at 11:00 PM I lauched a complete mirror from the DNS-323 to an other NAS using the "lftp" client. During the nigth I observed that both HD's blue led became pink while the network blue led continued to blink. This morning at 09:00 AM, the DNS-323 was unacessible by samba, but accessible by telnet. The lftp task was alive, then just killed it. In the e-mail I received this at 04:12 AM:
------------------------------
Left Hard Drive Has Failed
Right Hard Drive Has Failed

Sincerely,
Your DNS-323
-------------------------------
In the target NAS were files copied until 09:XX when I killed the lftp task.
In the status page of the DNS-323 the information about the installed HD "PHYSICAL DISK INFO: " was: no drive appears (blank table) and I remember that all other informations ok also:

Volume Name:    Volume_1
Volume Type:    RAID 1
Sync Time Remaining:    Completed"

but I'm not sure about :

"Total Drive(s):"

"Total Hard Drive Capacity:"
"Used Space:"
"Unused Space:"

I don't remember. In my mind these rows wasn't in the status page, but I'm not sure....

After a reboot the HD's led was blue.

During 02:34 AM to 04:42 AM the DNS-323 was copying 5 dvd iso images (debian-502a-i386-DVD-X.iso).


This seems a software problem.... It's strange that something happen in both drives in the same instant, without a tunderstorm and without breaking the tranfer.    

When I have time I try to check if in any of the tranfered dvd image is valid or corrupted.

[neuwirth]

I have been experiencing the same problem.  Firmware at 1.07 level.  I have two identical Western Digital 750GB drives
installed. 

1 WDC WD5000AACS-00ZUB WD-WCASU0852725 500 G 
2 WDC WD5000AACS-00ZUB WD-WCASU0877102 500 G

  When the lights turn Magenta (the combination of blue and red), I get a nasty write protected, talk to your system administrator message.  With Windows Explorer I can display the directories, but not the files themselves.  I can log into the DNS323 firmware and do a reset of the unit and everything is fine again.  The drives are configured in RAID 1, so all my data is at risk if both drives really fail at the same time.  I installed the drives when I purchased the DNS323 over a year ago.  At that time I configured the storage as two separate drives.  A couple of months ago I upgraded the firmware to 1.07 and in the process backed up everything and reformatted the drives for RAID1.  Since doing this the 'magenta light condition
has appeared twice.  After reading the numerous posts about this problem, I believe another firmware upgrade is needed to fix the problem because something in the firmware senses a false drive failure.

This item in the forum has been around for quite some time.  I see different levels of firmware, different brands of disk drives and no real response from D-Link.  The common thread seems to be the RAID-1 configuration.  I never had the problem before using that option.

[jtrevill]

Sorry, I know this is an old thread, but I've probably recreated the issued again as well... 

I have been trying to get my Upnp working on the 323 with my XBox 360.  Saw numerous posts about not being able to see the 323 with the XBox - with suggestions of restoring to default, and enabling Upnp AV Server.  I did this - and can now see the 323 with the Xbox.  However, if I try to "refresh", I get the above described lights - Everytime. 

I had been running 1.07, but after getting the weird colored lights, went back to 1.06 with the same result.  The status of the refresh is only up to around 30% or so, however, shows succesful when I noticed the lights are "weird". 

Here are my specs - and I'm not using any hacks at all:



Total Drive(s):  2
--------------------------------------------------------------------------------
Volume Name:  Volume_1
Volume Type:  RAID 0
Total Hard Drive Capacity:  980806 MB
Used Space:  176471 MB
Unused Space:  804335 MB






Slot Vendor Model Serial Number Size

--------------------------------------------------------------------------------
 
1 WDC WD5000AAKS-65TMA WD-WCAPW1080569 500 G 
2 WDC WD5000AAKS-00TMA WD-WCAPW1540045 500 G



Any thoughts?

Thanks!

Jason