Areca Contoller Configuration for unRAID


Recommended Posts

That ain't gonna happen, and it's not a driver issue as much as it is architecture.  Areca HW RAID cards can combine multiple drives to appear to the system as physically 1 pseudo-drive.  That's ALL the OS sees.  So /dev/sda may be 3 drives in HW RAID.  How do you query SMART from 3 drives at the same time?  You can't.  So HW RAID controllers have dedicated applications (and sometimes special SMART parameters) to talk directly to the controller and to get to the individual drives behind the pseudo-drives that are presented to the OS.  For example, maybe you want to report the highest of the temps of the 3 drives.

 

You can't match the /dev/sdX to the drives behind the controller without some custom scripting.  For example, maybe you want to report the highest of the temps of the 3 drives.  Someone else may want the average.  It's a kludge, but you can make it work by writing a shell around smartctl.

 

All that unRAID could do is let people plug in custom SMART "extra parameters" on a drive-by-drive basis, and save them with in a config file somewhere for unRAID to use when querying drive temps.

Link to comment

That ain't gonna happen, and it's not a driver issue as much as it is architecture.  Areca HW RAID cards can combine multiple drives to appear to the system as physically 1 pseudo-drive.  That's ALL the OS sees.  So /dev/sda may be 3 drives in HW RAID.  How do you query SMART from 3 drives at the same time?  You can't.  So HW RAID controllers have dedicated applications (and sometimes special SMART parameters) to talk directly to the controller and to get to the individual drives behind the pseudo-drives that are presented to the OS.  For example, maybe you want to report the highest of the temps of the 3 drives.

 

You can't match the /dev/sdX to the drives behind the controller without some custom scripting.  For example, maybe you want to report the highest of the temps of the 3 drives.  Someone else may want the average.  It's a kludge, but you can make it work by writing a shell around smartctl.

 

All that unRAID could do is let people plug in custom SMART "extra parameters" on a drive-by-drive basis, and save them with in a config file somewhere for unRAID to use when querying drive temps.

 

I have this working in unMenu. It now contains a script that you can add to the "go" file that figures out all of the mappings for Areca connected drives (pass through, JBOD, and even RAID drives) and creates a small file in the /tmp directory that contains that data. Then myMain has the logic to use that file to access and use the smart reports natively for the Areca connected drives (even component drives in RAID sets). This includes getting the spin status and temperature. The only thing that doesn't work is manually spinning down a drive - have not found a way to get Areca connected drives to do that with smartctl, and hdparm does not work with Areca connected drives.

 

I am working on trying to release this as an update. Just a lot going on in my life which is slowing me down.

Link to comment

[quote author=bjp999 link=topic=38487.msg403172#msg403172

I have this working in unMenu. It now contains a script that you can add to the "go" file that figures out all of the mappings for Areca connected drives (pass through, JBOD, and even RAID drives) and creates a small file in the /tmp directory that contains that data. Then myMain has the logic to use that file to access and use the smart reports natively for the Areca connected drives (even component drives in RAID sets). This includes getting the spin status and temperature. The only thing that doesn't work is manually spinning down a drive - have not found a way to get Areca connected drives to do that with smartctl, and hdparm does not work with Areca connected drives.

 

I do the same thing... I create an "areca.map" file in /tmp at boot, and subsequent scripts parse it for mappings.

 

I have not found a way to spin down a drive on the Areca either.

 

Link to comment

I seemed to have good luck setting spindown using the areca software built in and available via a web browser.

 

Am looking forward to getting tighter integration with unRaid gui and areca for temperature monitoring.

 

Presently you can temperature monitor the areca connected drives by the above mentioned browser method, but it's outside the neatly presented unRaid dash board.  And I can't really tell when they spindown as the temperature stays fixed at the last known temp, even if the drive is sleeping.  The only way I know they are sleeping is the power draw drops off on the APC dashboard.

 

Link to comment

Setup RAID0 volume to use for parity on the ARC-1231 today. 2x4TB HGST 7200RPM drives in RAID0 (thanks to previous post knew to rename volume without #). Couple reboots and panic attacks later figured out how to set the remaining disks on the controller to "pass-through" so unRaid could see them. Reassigned parity and now running parity sync. Interesting data bits on how unRaid sees the RAID0 volume:

 

Vendor: Areca

Product: ARC-1231-RAID00

Revision: R001

User Capacity: 8,001,573,355,520 bytes [8.00 TB]

Logical block size: 512 bytes

Rotation Rate: 10000 rpm

 

Liking that 10K RPM LOL. Says only 20 hours estimate to finish but I bet it picks up speed once it passes the 6TB mark and its all on it own in the sync. Also bought one of the new SMR 8TB Seagates (precleared for past 2 days) so once the sync is done we will be introducing the new big'un to the herd. :P Later on the 6TB the RAID0 replaced (both will replace 2TBs). So lots of syncs in the near future... then monthly parity check on the 1st. Going to get a workout for sure.

Link to comment
  • 2 weeks later...

I would like to setup a server using Unraid with a combination of an Areca 1261ML and onboard SATA ports with 24 disks in total, 16 on the Areca and the rest onboard. Would I set the Areca up in JBOD or RAID 0? Also would I need to do special things to the config as discussed in this thread to make it work with the Areca?

Link to comment

This would work fine.

 

If you want to combine the capacities of two disk to one larger logical drive, you can configure then as RAID0. This can be an effective way to create a larger parity disk from two equally sized smaller ones. You will want to disable drive size rounding. You could create multiple RAID0 pairs if you like.

 

If you create RAID sets, you can assign the disks that are not allocated to RAIDs as "pass through" disks. That will make then appear as drives in unRaid.

 

If you have no disks set up in RAIDs, you can select the JBOD option. That is equivalent to passing through every drive on the card, only much easier to do.

 

You should read this entire thread. There are important steps outlined to name your drives appropriately in unRaid.

 

Let me know if any questions or problems.

Link to comment
  • 2 weeks later...

The unassigned devices plugin now has been updated for Areca support.  It never showed the correct name for the attached devices.

 

Also the preclear plugin will be updated to match.

 

I still do my preclearing of drives via the motherboard ports since the smart support is still an issue on Areca.

 

I believe all we need are temperatures on the unRaid main page to close the book on full Areca support in unRaid 6. 

Link to comment
  • 2 weeks later...

I had problems with preclearing an 8TB Seagate recently that may have been related to being on the Areca card (didn't realize it at the time). Things like it would not finish the last bit; probably hung on the final SMART? This time around with 6TB Seagate I had problems getting Preclear to start from the plugin GUI. Once I moved it to a motherboard Sata III port it started up fine and ran without issue.

Link to comment

So if I want to pre clear a drive currently connected to my Areca, I should plug it into an onboard SATA port to pre clear it then?

 

It will IF you use bjp999 version of preclear_disk. It has a switch to bypass all those "Yes" prompts, and this capability is recognised by the plugin. What happen is that with the "official" version of preclear_disk, the plugin only jumps the first prompt, and preclearing from an Areca card will prompt you a second time about not retrieving SMART data.  Because of that, you will have a stuck "Type Yes" message with no means to circumvent that.

 

The SMART data gathering is important to verify if sector reallocation took place, or if any attributes of the disk are failing. Because of that, it's advised to preclear from a SATA port on the motherboard or on a LSI/Marvell card.

Link to comment
  • 6 months later...

I know It has been a couple of months since someone posted something here.

I'm absolutely new to unRAID, I bought a used Areca ARC-1680IX-24 and connected it to 6X2TB seagate in a Norco RPC-4220.

 

I had problems with smartmontools not getting any SMART readings from the drives.

 

I managed to track down this in the areca FAQ:

http://faq.areca.com.tw/index.php?sid=308038〈=en&action=artikel&cat=9&id=85&artlang=en

 

My controller had version 1.48 and after updating to the latest firmware (1.51) it works just fine.

 

I thought it's worth mentioning in order to help anyone who's having this problem,

and I can also confirm that this method works for 1680 controllers as well.

Link to comment

So where did you download and install smart tools? On your UnRaid server?

 

I didn't download it. smartmontools (smartctl) was already present in my unraid installation - version 6.19.

Also in order for it to recognize the drives it is recommended to update the drivesdb.h file

which contains the drives names and how to handle their SMART data.

My drives are of a new model (Seagate NAS drives), so I had to update the file in order for it to work.

Link to comment
  • 1 month later...

I'm using this card on unRaid 5.0.6 (thanks guys) and would like to upgrade to the latest stable unRaid 6.  Would the steps be:

 

1. run the 6.01 Upgrade Plugin for V5 and reboot

2. verify that 60-persistent-storage.rules is still at \\tower\flash

3. update the fresh ‘go’ file with the 4 lines of script before loading emhttp and reboot

 

Link to comment
  • 1 month later...

hey all so i've been away from the board for a while now and just happened on this thread..

i'm definitely not an expert

 

i have the areca ARC-1284Ml-24 card and i'm not sure that i have the issues you all are describing. its the only card ive used with unraid so i cant compare it with anything.. i've been using it after 6.0 Beta 12  when they added drivers for it. It seems to run without any problems and i have never had to do anything custom with it.. just put it in JBOD mode and slap drives in it.

 

here's a screen shot. on ver 6.1.9

 

 

Capture.JPG.261441aff000b199d8538f075cb654fb.JPG

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.