Unfortunately no one can be told what fun_plug is - you have to see it for yourself.
You are not logged in.
Hi Folks,
DNS-323 with 1.03 (UK) firmware and 2 x Western Digital 160Gb SATA drives (WD1600YS) in RAID-1 config EXT2 filesystem connected to DSL-G624T router and accessed wirelessly by several XP Pro PCs.
Setup of the DNS was a painless task but subsequently we discovered anomalies in the document archive hosted on the DNS. Upgrading the firmware from 1.00 to 1.03 has resolved that anomolous behaviour. The drives were reformatted to EXT2 at the time of the firmware upgrade.
Another challenge I've encountered a couple of times now on logging into the administration console on the DNS is that it tells me that it needs to format (one of the drives) and then presumably re-synchronise the data across the drives. It's as though the DNS is responding to me having swapped one of the drives in the device which I haven't done.
Not seeing anything other than blue status LEDs on the DNS box which suggests to me that all is in order with both drives.
So to my question: has anyone else encountered this or something like this?
Thanks
Last edited by roastbeef (2007-06-22 15:44:17)
Offline
FYI - Just raised a ticket with D-Link support in relation to this issue. Ticket ref. 1135717. Will report back with outcome in due course.
Offline
Whilst I haven't experienced the problem you describe, when I first got my DNS-323, I did some fairly extensive testing of it's RAID1 capabilities and found - much to my dismay - that the drive failure sensing is not up to my requirements.
I believe it may be possible depending on the circumstances for the unit to detect a failed drive and not report it, as well as to have a drive fail and not detect it. In fact, I have never seen the front panel LEDs go amber, even with an usable drive installed in the unit.
Starting with a pair of identical 80 GB drives in RAID1, I have deliberately pulled one drive - both power on, and power off. Prior to firmware 1.03 there was no indication of a failed drive and if the same drive was re-installed, the unit was apparently completely unaware that it had ever been removed. If a new (or clean) drive was installed, the unit would at the netx login of the web interface indicate that it needed to format and resync the drive. After upgrading to firmware 1.03, the unit will detect a drive as having been removed and send an email, if this has been configured.
I'd like to point out here that RAID1 worked in that at no time during these tests was data lost.
I have subsequently installed drives that report SMART failures, drives known to have bad sectors and drives that could not be detected by the BIOS in a desktop PC - at no time has the DNS-323 ever indicated, either by email or amber front panel LEDs that there was a problem with any of these disks, the drives with SMART failures & bad sectors were formatted and I was allowed to use them, however, the drives that were not detected by the PC BIOS were, to some extent, detected by the DNS-323 in that the corresponding drive LED would flash blue continuously.
Initially I had hoped to recommend the DNS-323 to a couple of small business clients, but have refrained from doing so, out of an abundance of caution - an inexperienced user with dialup internet could/would be left unaware that a drive had failed, until the second one went. My DNS-323 is used for disk to disk backup under the assumption that I will not bso unfortunate as to lose the disk in my desktop at the same time as I lost a disk in the DNS-323.
Last edited by fordem (2007-06-22 15:52:04)
Offline
Thanks for your response on this fordem. I hadn't performed any testing of the RAID1 capabilities of the device until the problem described above forced me to. Call it blind faith, if you like ) The price I've paid for that blind faith is, so far, just a heart stopping moment, when this issue first happened. Thankfully I was still able to download the contents of the DNS to the local drive on my desktop prior to commencing the format/resynch.
It was a great relief to see the resynch complete and all my data still present and intact but, understandably, it concerns me that this seems to be a persistent issue.
Offline
Since you mention blind faith - a word of warning, and this is not a criticism of the DNS-323, it is a fact of life.
If you value your data - BACK IT UP - RAID1 is NOT a form of backup.
Offline
The data is valued and is, of course, backed up routinely to an external device.
The key benefit of the DNS device is mostly in the fact that it's a more fault tolerant storage device than a shared drive in a desktop, for example. In buying the DNS the assumption I've made is that the RAID1 mirroring works effectively, thereby protecting me from individual drive failure, and that the device is, in itself, fundamentally reliable. Anything else it delivers is, as I see it, a bonus.
We shall see how the DNS fairs in both measurements. The RAID1 certainly seems to be working effectively based on my first-hand experience.
Offline
Thought I'd take a moment to comment on the support response I've had from D-Link so far...
Firstly, here's the detail I added to the e-ticket via the D-Link web site on 22nd June...
" Hi There,
Please will you help me with this challenge?
Environment details: DNS-323 with 1.03 (UK) firmware and 2 x Western Digital 160Gb SATA drives (WD1600YS) in RAID-1 config EXT2 filesystem connected to DSL-G624T router and accessed wirelessly by several XP Pro PCs.
Setup of the DNS was a painless task but subsequently I discovered anomalies in the document archive hosted on the DNS. Upgrading the firmware from 1.00 (I think) to 1.03 and reformatting the drives with EXT2 filesystem has resolved that anomolous behaviour.
Another challenge (the one I need help with) I've encountered a couple of times now on logging into the administration console on the DNS via one of the client PCs (using IE6) is that it tells me that it needs to format one of the drives and then presumably re-synchronise the data onto the newly formatted drive. This appears to go through to successful completion. It's as though the DNS is responding to me having swapped one of the drives in the device which I haven't done.
I'm not seeing anything other than normal blue status LEDs on the front of the DNS box which suggests to me that all is in order with both drives. i.e. no faults.
So to my question: what is the cause of this? And, more importantly how can we resolve this issue?
Thanks"
And here's the response I received from an unnamed representative on 25th...
"Were you having this issue before the firmware upgrade ?"
I replied to this question promptly but have had nothing since.
To say that I'm disappointed in the single threaded response is an understatement.
Clearly a big mismatch between my expectations of D-Link support and what they actually deliver in terms of customer service :-(
Offline
fordem wrote:
Since you mention blind faith - a word of warning, and this is not a criticism of the DNS-323, it is a fact of life.
If you value your data - BACK IT UP - RAID1 is NOT a form of backup.
What if the DNS-323 is used as a backup server? I'd imagine the RAID1 would be helpful in making sure your disk image backups are safe.
Offline