Print

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

[fordem]

All right I think this is severe enough that it warrants a separate thread.

The DNS-323 with fw 1.05 appears to have disk failure detection issues and may or may not send an email alert when a disk fails.

Basic information - Revision A1 hardware, firmware 1.05, configuration reset to a factory default and then the following changes made - static ip address set, system name set, iTunes and uPnP server disabled, all e-mail alerts enabled with a 24 hour interval set on the status e-mail alert.

I initially started with a pair of 80GB Maxtors in a standard volume configuration and backed up 60+ GB to Volume_1 and another 30+ to Volume_2.  I then powered the unit down and removed Volume_2 (left bay).

The unit gave no indication of a failed disk, no amber LED, no email alert.

I next installed a "clean" as in no partitions 80GB Maxtor and rebooted the unit and logged in, and allowed it to format the second drive and create a RAID1 pair - this went well.

Several hours later I hot unplugged the right bay drive to simulate a failure - it took the unit several munites to send me an email alert and even longer for it to give me a "LED" alert - (I now know what the amber LED looks like), but the status page still does not show that there is any problem, even after several refreshes .

After a reboot the status page shows the array as degraded, but the right LED remains off.

[fordem]

The "failed" drive was replaced with a "clean" drive and the unit allowed to format at it.

I noticed (and was somewhat concerned) that the format successful message reported the volume name as Volume_2, but on a reboot it proceeded to sync the drive as expected, so this last would appear to be cosmetic.

Attention does need to be paid to drive failure alerts on standard volumes as well as RAID volumes - admittedly it is easier to detect a failed standard volume since the data just disappears, something which should not happen with RAID1

[fordem]

For the next test in the sequence I powered the unit down, removed a drive - the right this time - and then powered it up.

The result was no amber LED and no email alert, the status page did however indicate a degraded status.  What this means is that is a user has a drive fail whilst the system is off he will not now about it if he is not observant enough to notice that a drive LED was not lit.