Print

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

[EFighter]

I have 2 1.5TB WD Green Caviar drives in the NAS.

Got an email that one drive failed.  Ordered a new one(1.5TB) from new egg.  Stuck the drive in, formatted it and it began to auto rebuild.  it started by saying 288 min.  no matter how long i leave the drive in the second bay it never finished. i left it in there trying to sync for 4 days.

I removed the second drive and ran a boot util on it, the drive checks out just fine, so i reformatted the drive and tried again. only to get the same result.  it has been saying the same thing.  stuck on 260-320 min just depending on the refresh, but it never goes down.

Do i have any options as far as getting the rebuild done?

I read this http://forums.dlink.com/index.php?topic=46506.0 and i have to say that i didnt follow those directions.  i did the reformat while the good drive was still in the NAS.

Also when the NAS promted me it asked me about reconfiguring the Raid-1 array.  there was a check box.  Should that be checked or not?

Any help is appreciated.

Oh and on a side note so you dont jump all over me, all of my data is safe in the cloud i just would rather rebuild the raid-1 array and move on with my life.  Thanks

[fordem]

To the best of my knowledge, a recovery from a failed drive should require no more than replacement of the failed drive, a login and a click 'OK" (although that might be continue), followed by a reboot - no need to tick any check boxes.

On the first reboot after you replace the failed drive, when you login, the system should prompt you to format the new drive and you have to click OK or continue before it will format.  After the format you reboot and it should automatically resync.

As far as the instructions in that link go - personally - I wouldn't follow them - no swapping of disks is required - and yes, I do have hands on experience with rebuilding the arrays on a DNS-323, both from numerous lab tests and at least one actual failure.

Here is D-Link's procedure for failed drives in a RAID1 array.

If that doesn't work for you and you have a backup, just wipe both drives and create a new array and restore.

[EFighter]

is there a way for me to just tell the NAS, "Stop trying to sync and re-configure the array"?  Normally in the raid area of the NAS there is a button to do that but while it is syncing that button is replaced by the amount of time necessary to rebuild the array.

[EFighter]

Also another question.

While i am in here messing with things should i go a head and upgrade the firmware?  any known issues on OSX for upgrading?

[rogerwolf]

I just ran into the same issue myself (under Firmware 1.08, though). Clicked the 'rebuild now' button, let it run for some time, but the time to complete just reports between 240-350 minutes everytime, but it doesn't seem to go down after several hours.

However, after using the telnet interface into the box (available after installing fun-plug), I was able to issue the following command:
cat /proc/mdstat

That shows the rebuild process and includes a percentage (and also the same time-to-complete value as the web-interface is showing).

The percentage is growing slowly but steadily - but the time-to-complete value is just erratic. Based on the percentage completed and the real time passed, I'm now estimating that the rebuild process will take close to 40 hours (for my 1.4 TB volume); not the 264 minutes (or something) that is indicated.

Conclusion: my rebuild is happening, it's progressing, it will (hopefully) get there. The reported time remaining is so inaccurate that it's best ignored.

Roger

[fordem]

Just so you're aware of it - read/write access is given priority over rebuild, and this can be the reason why the rebuild estimates vary significantly