RocketRAID RAID5 issues


Recommended Posts

So I've been running a file server using a RocketRAID 2300 with 4 2TB WD Green drives in a RAID 5 for the past three years. A few weeks ago I started noticing some performance issues starting to happen: Slow transfer speeds, video freezing, etc. when accessing the data over my gigabit network.

 

I opened up the controller's web management software and saw that one of the drives was reporting 231 bad sectors found and repaired, while the rest still had zero. I tried running a "Verify" task on the array multiple times but each time it would eventually stop responding (the raid management software wouldn't even open). Eventually I would have to restart.

 

Newegg had a 2TB Green drive as a ShellShocker deal last week so I figured it would be a good time to replace the drive with the bad sectors, hoping it would solve my problem. I replaced the drive Tuesday night. When I booted up my system the RAID BIOS screen at startup listed the array as Critical and I added the new drive to the array and continued on to Windows. Once Windows was running I opened the RAID management software and saw that it had automatically started a Rebuild like it was supposed to.

 

The next day around noon (15+ hours into the rebuild) the software still said 0% complete. I tried to reopen the software and again it would not respond or open, just as it did before when I tried to run a verify task. It's now been three days with no response as Ive just been letting it run. It appears to not be doing anything right now, the most recent entry in the raid management log is the "starting rebuild task" entry from Tuesday night. I was able to backup all my data from the array beforehand so I'm not at risk of losing any data thankfully.

 

Could this be a bad controller card? I don't know what else to try. I'd hate to drop the cash on a new card if I wasn't sure.

Link to comment
Share on other sites

The best thing to do would be to call the manufacturer and have them troubleshoot it and point you in the right direction. 

 

IMO, the card is having problems rebuilding the array.  This is a common issue.  You may be instructed to wipe out the array and rebuild it from scratch, wiping out any data that you may have.  I am having this same exact issue with an old dell server where I have inserted the drive, it sees the drive, but it will not rebuild automatically.  I have been instructed to backup the server, delete the raid settings (this will delete the disk and all data on it), and then recreated the raid array.  This is something that we do not want to do as a company but have done in other situations without issue.

Link to comment
Share on other sites

You first clue was that your Verify did not complete and thats bad because replacing the drive can make things worse but what I would like to know is when was the last time you did a Verify of the RAID?


 


Running SpinRite would have been a better option before you replaced the drive and Verify completes.


 


The best way to test your RAID card if one of the ports is bad would be to remove your array and with the new drive connected do a test run as a single drive on all the ports.


Link to comment
Share on other sites

Its been over a year since the last Verify was done before this. I know I haven't done one since I moved to the new house as the server hasn't been running much since then. How often should they be done?

 

I deleted the old array and created a new one using the new drive. It's initializing now. I started it last night as a (Foreground) initialization and it got about 50% complete when the electricity went out early this morning and was out longer than my UPS could keep the server up for, so I was forced to shut it down partway through initialization. When I started it back up the initialization resumed only it was a Background initialization this time. Should it be ok still?

Link to comment
Share on other sites

Given the scale of 2TB disks in RAID 5 and 4 of them with 2TB of parity that if one goes you are running in a sense RAID 0 but by calculating the missing block so really once every two months for a Verify . When initialization is done and should be fine on resumed with your data put back on run a Verify then.


Link to comment
Share on other sites

Try booting and going into the Highpoint Bios and doing a rebuild from there.  Windows might be interfering with the rebuild process.  Right now you haven't lost any data, so I would begin copying the data off the server as a backup incase another drive begins to fail. 

 

On a side note:  I've used Highpoint cards before and I wasn't very impressed.  Areca, PERC, or LSI are the only cards I'd trust for critical data.  Also, RAID 5 isn't very efficient for a multiple terabyte array.  It was good back in the days of 146Gb SCSI drives, but the rebuild time for an array your size could be days.  I personally use RAID 6 or RAID10 if the data is critical and important.  If it's just Bluray rips and Tv Episodes... JBOD and deal with single drive failures when they happen. 

Link to comment
Share on other sites

Rebuild is finished and the array is working good so far, nice and fast like it was when it was new; still in the process or copying 3TB of data back onto it.

 

Noticed that the new WD Green I put in as a replacement consistently runs about 5 degrees cooler than the other three drives (also WD Greens but from 2010).

Link to comment
Share on other sites

This topic is now closed to further replies.