ARECA Owner's Thread (SAS/SATA RAID Cards)

Hello all. I have 2 raid cards - an ARC1220 and an ARC1210. I'm not using the 8 port version much anymore, as I've moved to a seperate Unraid box.

I am still using the 4 port in my desktop though. I'm just about out of space; it's currently using 4x 750gb drives.

This black friday, I picked up a pair of 3TB seagates from Newegg. Specifically, the 7200.14 ST3000DM001 . I was wondering, is there anything special I need to do to format these drives to build an array with them? These will be the first Advanced Format drives I'll be adding to an Areca card, and also the first to break the 2tb barrier as well.
 
Those drives work fine in hardware RAID - myself and several people I know are running quantities of that drive on Areca controllers with no problem. As for > 2TB support just make sure you choose "64Bit LBA" when creating the Raidset, DO NOT choose 4k - thats a legacy option and has nothing to do with 4K sectors.

I'd also highly recommend running a full read/write test of the individual disks before creating an array with them, so you aren't leaving it to the array build process to discover a bad drive which means lost time and having to start it over, its also not the most comprehensive test.

Use either badblocks in PartedMagic, or HDTune Pro's "Erase" tab with the Verify option ticked, or any number of third party or vendor supplied tools that can perform a surface scan. Then using a tool like HDSentinel, CrystalDiskInfo or HDTune's "Health" tab, examine the SMART stats before and after the full disk scan to make sure no sectors were silently reallocated during the scan - if so return the drive.
 
Last edited:
odditory said:
Those drives work fine in hardware RAID - myself and several people I know are running quantities of that drive on Areca controllers with no problem. As for > 2TB support just make sure you choose "64Bit LBA" when creating the Raidset, DO NOT choose 4k - thats a legacy option and has nothing to do with 4K sectors.

I'd also highly recommend running a full read/write test of the individual disks before creating an array with them, so you aren't leaving it to the array build process to discover a bad drive which means lost time and having to start it over, its also not the most comprehensive test.

Use either badblocks in PartedMagic, or HDTune Pro's "Erase" tab with the Verify option ticked, or any number of third party or vendor supplied tools that can perform a surface scan. Then using a tool like HDSentinel, CrystalDiskInfo or HDTune's "Health" tab, examine the SMART stats before and after the full disk scan to make sure no sectors were silently reallocated during the scan - if so return the drive.
Click to expand...

Thank you much, that was incredibly helpful and specific :)
 
OK, I have followed your directions. I installed a disk as directed, checked SMART with HD Sentinel, did a surface scan with verify in HDtune, Pro, and when done; rechecked SMART with HD Sentinel. I then repeated those steps for the second disk.

I think everything OK but I'm not too good at reading SMART; but I'd like to get another set of eyeballs on it before I start getting everything going on this - these look good to you?

Disk 1 Disk 2

Thanks again for the assistance. :D
 
Yeah those are fine, HDSentinel formats SMART data in a way thats unnecessarily hard to read and interpret, but if you look under "Reallocated sectors count" under the Data column you notice "0". The green checkmarks it places next to certain stats are the more critical stats and the main ones it weighs to assess an overall health status.

I also see a Load/Unload Cycle Count of only 7 which is good because it means you haven't changed the "Time To Hdd Low Power Idle" from the default "Disabled" in Areca's management GUI. You want to leave that disabled since it keeps APM disabled on every drive connected to the controller (meaning the heads aren't constantly parking). I mention it because sometimes people play around with the HDD power mgmt settings trying to reduce power consumption but not understanding what each setting does completely - and the only one of any consequence in that regard is Time To Spin Down Idle HDD, which I usually set to 40 or 60 (minutes).
 
Last edited:
I was indeed planning on leave those alone (except for the spin down, which is indeed usually 60 min but I turned it off while messing with these new drives). The way I see it, when the raid is on, I want it cranking. I have an unraid box for lower power consumption (and less throughput, but that's OK).

You have helped me enormously and I really am grateful. Thank you again.
 
Apologies if I'm hallucinating but I thought I read earlier in this thread Areca cards had abnormal performance degradation where multiple volumes were accessed simultaneously... is this still the case with the 1882 series?
 
Hey, how does one get the status of a RAID1 mirror on a 1210 card from within an ESXi server setup?

I'm using a 1210 with a mirror for VM storage. I'd like to be able to keep tabs on the status of the drives on that card. But their health values don't appear to show up inside the vSphere client. Anyone know if a 1210 can be monitored in this manner?
 
Hello,

I am new to this forum so maybe this is a silly question, but I was not really able to find it anywhere. I have Areca ARC-1882i card and two Hitachi 15k600 drives (SAS, connected via cable directly to the card). I created a raidset and a volumeset, all good. Then I formatted the new drives and copied the data I wanted to have there. My setup looks like this:

CLI> rsf info
# Name Disks TotalCap FreeCap MinDiskCap State
===============================================================================
1 PassThroughDisk 1 2000.4GB 0.0GB 2000.4GB Normal
2 PassThroughDisk 1 2000.4GB 0.0GB 2000.4GB Normal
3 raidset1 2 1200.0GB 0.0GB 600.0GB Normal
===============================================================================
GuiErrMsg<0x00>: Success.

CLI> vsf info
# Name Raid Name Level Capacity Ch/Id/Lun State
===============================================================================
1 ST2000DL003-9VT1 PassThroughDisk PassThr 2000.4GB 00/00/00 Normal
2 ST2000DL003-9VT1 PassThroughDisk PassThr 2000.4GB 00/00/01 Normal
3 volset1 raidset1 Raid0 1200.0GB 00/00/02 Normal
===============================================================================
GuiErrMsg<0x00>: Success.

CLI>
Click to expand...

While I was doing this, the drives were actually on a cardboard box. Then I used the "offline raid set" function to spin the drives down to be able to disconnect them and mount them in a box properly. After reconnecting I had incomplete array and after forcing array to start, it marked one of drives as failed and that was it. I was unable to restart the server as there are some calculations running so I repeated the whole process, deleted volumeset and raidset and started pretty much from the scratch. I also lost the data on the drives but that was a minor issue as the drives are backup and I just re-run the backup again. The data loss was my bad (i guess) because I Ok-ed volumie initialization in windows, probably shouldn't have done that. But still, rebuilding all the arrays and volumes seems tedious (and what happens if i no longer remember parameters of the array, e.g. stripe?).

My question is: how is one supposed to disconnect drives from a raidset and connect them later without needing to reboot the system and without losing the data, of course.

Thanks ...
 
If anyone is using an Areca RAID controller with either Windows 8 or WIndows Server 2012 you should go up to the Areca website and download the brand new (as of today) v24 WIndows drivers asap. I've been working with Erich Chen @ Areca to diagnose a very weird and nasty bug which was causing volumes to lock up under heavy IO, and Erich seems to have fixed the issue in the new driver. Would be interested to hear feedback from others on whether this does indeed fix any issues you may have been having with Areca on the latest versions of Windows.
 
Excellent heads up, I have an alert configured for changes to their driver page but somehow it didnt trigger. Talk about long overdue, the last update to the x64 win driver was october '11. But in fairness Areca tends to "just work" whereas LSI seems to be issuing a new firmware/driver combo every other week and tend to break as much as they fix in each iteration.

I haven't noticed freezes under heavy I/O load, but I have noticed quirks which I think I'd narrowed down to the previous storport driver, having to do with accessing VSS snapshots with a volume that was spundown. Will see if I can reproduce with this newer driver. (X9SCM-iiF + 1882i PCIe3.0 + HP Expander + Server 2012)
 
Hi, I'm looking to upgrade the cache memory on my 1882i-12 card from the standard 1GB to 4GB. After some googling, I think what I need is a single rank x8 or x16 DIMM. However, I can't seem to find such a beast. Anyone have a part number I can use that's not the Areca part since that is over $125 and other ECC DIMM of same size a $30-40? TIA!
 
I have an Areca 1222 with 6 drives running in a RAID 6 (5+1 hot spare). It was running great for a while but then the entire server stared to just freeze (requires hard reset of power/reset button). It seemed to freeze anytime I started hitting the drives hard (installing something, for example). Now it is extremely rare I can even get Windows Server to boot at all (even into the Repair Console (but it DOES boot from CD)).

The Areca card "boots" just fine and I can get into all of the settings and everything looks normal. There are no errors or warnings in the System Events and I've run "Check Volume Set" several times without any errors. When I continue past the Areca screen, Windows then tries to launch. Right now it freezes almost instantly after it starts loading. The HDD activity lights will also "freeze" in whatever state they were in. Usually all lights but 1 are on (and it always seems random as to which light will be out). When this happens the Areca Web-based Browser Configuration also stop responding. I've tried moving the card to a different PCI-E slot with identical results. I've physically moved the HDDs around (there are 8 drive bays all with different power and SATA connections) and it doesn't change anything.

Just for added information the serverboard/case is a SuperMicro SuperServer 6026T-NTR+ with a single Intel Xeon E5645. It does have a backplane that the SAS-to-SATA cables from the Areca card plug into.

At this point I'm just not sure what to check next. Does it sound like the Areca card? Could it be the mainboard? The backplane? Should I try to bypass the backplane and go directly into the card? Try a single SATA drive plugged directly into the mainboard? I ask because the way everything is set up will make any physical change difficult so I'd like to try to most informative method first. Should I try something else? I think both the bare server and the areca card are still under limited warranty so I (hopefully) should be able to RMA them if there is something wrong.

Oh, one other random note: When I was working on it earlier tonight, after moving around the card, the first few boots gave me a screen with nothing but:
RN50 DDR2 2A 266M/166E 32MB
Click to expand...
Which I am assuming is referring to the onboard video but it stopped when I re-seated the Areca card. This could mean nothing but it started me thinking that it may not be the raid card and may be the mainboard.

Thanks.

EDIT: I should probably also mention that after it initially started happening I did a lot of repairing in Windows Server (rollbacks/restores, repair install and now the current OS is a clean install on a brand new RAID/Volume Set), all having the same problems. Also, the firmware on the Areca card is up-to-date as are the drivers on Windows.
 
Last edited:
Hammer! said:
Hi, I'm looking to upgrade the cache memory on my 1882i-12 card from the standard 1GB to 4GB. After some googling, I think what I need is a single rank x8 or x16 DIMM. However, I can't seem to find such a beast. Anyone have a part number I can use that's not the Areca part since that is over $125 and other ECC DIMM of same size a $30-40? TIA!
Click to expand...

You're probably not going to find a better price than $125 for the 4GB ECC registered DIMM

Some older topics (not specific to 1882i)

http://www.techreaction.net/forums/...-4gb-memory-areca-1880-series-raid-cards.html
http://hardforum.com/showthread.php?t=1360900
 
I've noticed occasional bugcheck BSODs on my 1680i based backup server after upgrading to Server 2012- here's hoping that the new driver does the trick.
 
Thanks. The old 1880 series uses DDR2 memory, but the newer 1882 series used DDR3. Thanks. I found something and will report back if it works.

Hammer!
 
toadthetoad said:
I have an Areca 1222...
Click to expand...

After many, many more hours of testing I have determined that it is definitely my Areca ARC1222 that is crashing. I guess I get to see how good their customer support is now, has anyone had to RMA anything through them before?
 
Can anyone recommend a better source for ARC-1880i cards where you can choose the style of cable you get? Tekram wants $30 to ship one and that seems a bit on the high side to me.
 
Hi guys,

I am using an Areca 1880i
I have a Raid6 volume with 2Tb Sata3 disks on ports 3,4,5,6,7,8

In the next two days I'll buy a HP SAS Expander,
The thing I would like to know is how to change the ports without destroying the Raid?
Since I am going to change from internal Ports into the Expander, should I do something special or the Areca does everything Automaticaly recognizing the disks even that they change Ports?
 
areca controller doesn't care on which ports or order raid member disks appear, as long as they are all present. However best practice is to keep them in sequential order in case you ever needed to recreate the array in recovery mode (no init)
 
odditory said:
areca controller doesn't care on which ports or order raid member disks appear, as long as they are all present. However best practice is to keep them in sequential order in case you ever needed to recreate the array in recovery mode (no init)
Click to expand...

Thanks for your reply.

That means i can shutdown Server, unplug Disks from Internal Areca Ports 3-8 and connect them on Port 1-6 in the HP Expander right?
 
That site doesn't say out of stock because they aren't stocking anything, they're simply reselling. When an order is placed with them they submit a PO to Tekram (US Distro) and it gets drop-shipped.

My advice is either stalk the Newegg openbox items for Areca (search Areca -> Open Box items), check out ebay, or if buying new is paramount then Jus is a trusted Areca reseller, I bought my most recent 1882i PCIe3 from him a couple months ago. He's been on this and other storage forums helping people with RAID arrays for years, and has been known to hook people up with [H] pricing.. Ordering from a reseller instead of Newegg generally ensures you're getting the freshest stock since its coming from the distro.

If you're after latest/greatest then get an 1882i PCIe3.0 and before placing an order with someone, confirm/specify you're in fact getting the PCIe3.0 version.
 
odditory said:
That site doesn't say out of stock because they aren't stocking anything, they're simply reselling. When an order is placed with them they submit a PO to Tekram (US Distro) and it gets drop-shipped.
....
Click to expand...

Thanks for the advice odditory! :)
 
toadthetoad said:
After many, many more hours of testing I have determined that it is definitely my Areca ARC1222 that is crashing. I guess I get to see how good their customer support is now, has anyone had to RMA anything through them before?
Click to expand...

Common 1222 Problem. I have RMA'd over 350 1222's with areca that suffered from this problem. They usually took 2-3 weeks when I RMA'd them in 20-50 card batches.
 
odditory said:
That site doesn't say out of stock because they aren't stocking anything, they're simply reselling. When an order is placed with them they submit a PO to Tekram (US Distro) and it gets drop-shipped.
Click to expand...
Seems pretty dumb of them.
 
If it's listed as out of stock, people won't order. Who wants a back order of unknown length? That they'll have to [try to] cancel? If they drop ship and don't carry stock, they shouldn't say not in stock -- they should just say the item is shipped in two or three days -- whatever their supplier-lead turnaround is.
 
I tried a couple different DIMM's and neither worked. So I just bit the bullet and ordered the real thing from ArecaDirect. Also figured my data is worth much more than the $100 I was going to save....
 
You must log in or register to reply here.
Back
Top