With electrical failures of course anything is possible, but RAM errors usually manifest with row or column errors (Or just column with ECC) not general voltages on the card.
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
ARECA Owner's Thread (SAS/SATA RAID Cards)
- Thread starter odditory
- Start date
That makes sense, if the voltage regulator(s) are failing it would be inconsistent right? Right now the card keeps going back and forth with over voltages and recovered. its really wierd.
Current status:
| CPU Temperature | 101 ºC |
| Controller Temp. | 33 ºC |
| 12V | 12.342 V |
| 5V | 5.134 V |
| 3.3V | 3.376 V |
| DDR-II +1.8V | 1.856 V |
| CPU +1.8V | 1.872 V |
| CPU +1.2V | 1.264 V |
| CPU +1.0V | 1.056 V |
| DDR-II +0.9V | 0.928 V |
| Battery Status | Not Installed |
| 2020-07-29 07:36:54 | Ctrl CPU 1.0V | Recovered | ||
| 2020-07-29 07:36:54 | Ctrl CPU 1.2V | Recovered | ||
| 2020-07-29 07:36:54 | Ctrl CPU 1.8V | Recovered | ||
| 2020-07-29 07:36:54 | Ctrl DDR-II 1.8V | Recovered | ||
| 2020-07-29 07:36:54 | Ctrl 3.3V | Recovered | ||
| 2020-07-29 07:36:54 | Ctrl 5V | Recovered | ||
| 2020-07-29 07:36:54 | Ctrl 12V | Recovered | ||
| 2020-07-29 07:36:54 | Ctrl CPU Temp. | Over Temp. | ||
| 2020-07-29 07:36:24 | Ctrl CPU 1.0V | Over Voltage | ||
| 2020-07-29 07:36:24 | Ctrl CPU 1.2V | Over Voltage | ||
| 2020-07-29 07:36:24 | Ctrl CPU 1.8V | Over Voltage | ||
| 2020-07-29 07:36:24 | Ctrl DDR-II 1.8V | Over Voltage | ||
| 2020-07-29 07:36:24 | Ctrl 3.3V | Over Voltage | ||
| 2020-07-29 07:36:24 | Ctrl 5V | Over Voltage | ||
| 2020-07-29 07:36:24 | Ctrl 12V | Over Voltage | ||
| 2020-07-29 07:36:24 | Ctrl CPU Temp. | Recovered | ||
| 2020-07-29 07:35:54 | Ctrl CPU 1.0V | Recovered | ||
| 2020-07-29 07:35:54 | Ctrl CPU 1.2V | Recovered | ||
| 2020-07-29 07:35:54 | Ctrl CPU 1.8V | Recovered | ||
| 2020-07-29 07:35:54 | Ctrl DDR-II 1.8V | Recovered | ||
| 2020-07-29 07:35:54 | Ctrl 3.3V | Recovered | ||
| 2020-07-29 07:35:54 | Ctrl 5V | Recovered | ||
| 2020-07-29 07:35:54 | Ctrl 12V | Recovered | ||
| 2020-07-29 07:35:54 | Ctrl CPU Temp. | Over Temp. | ||
| 2020-07-29 07:35:24 | Ctrl CPU 1.0V | Over Voltage | ||
| 2020-07-29 07:35:24 | Ctrl CPU 1.2V | Over Voltage | ||
| 2020-07-29 07:35:24 | Ctrl CPU 1.8V | Over Voltage | ||
| 2020-07-29 07:35:24 | Ctrl DDR-II 1.8V | Over Voltage | ||
| 2020-07-29 07:35:24 | Ctrl 3.3V | Over Voltage | ||
| 2020-07-29 07:35:24 | Ctrl 5V | Over Voltage | ||
| 2020-07-29 07:35:24 | Ctrl 12V | Over Voltage |
Thank you again for all your help.
Bill
Last edited:
Polynomial-C
n00b
- Joined
- May 5, 2020
- Messages
- 4
Hello,
first of all, thank you very much for this really valuable thread. I've read it from the beginning and learned a lot about my Areca controllers.
I'd like to reply to two requests that IMHO have not been answered thoroughly yet.
first of all, thank you very much for this really valuable thread. I've read it from the beginning and learned a lot about my Areca controllers.
I'd like to reply to two requests that IMHO have not been answered thoroughly yet.
- There were some inquries about ARC-1680 controllers having trouble with SSDs that only have a SATA link speed of 1.5G. I haven't seen a reply that pointed into the right direction to fix this issue so I'd like to provide some additional information about it. The problem is with the onboard SAS-Expander. The embedded firmware of that expander has a bug that makes most SSDs having only that slow 1.5G SATA link speed. There is a firmware upgrade available for the 1680's SAS-Expander but flashing it onto the expander chip is rather dangerous and can even brick the SAS-Expander and thus bricks the entire controller card.
- Regarding the "Over Voltage" issue: I had an ARC-1880ix-12 controller with that problem myself. After getting in touch with Areca support they told me to rule out faulty PCIe slots on the motherboard first and if that is not the cause for the over voltage, the controller card is broken and must be replaced.
Code:
CLI> hw info
The Hardware Monitor Information
=====================================================
[Controller H/W Monitor]
CPU Temperature : 4294967294 C
Controller Temp. : 4294967248 C
12V : 15.504 V
5V : 6.854 V
3.3V : 4.080 V
DDR-II +1.8V : 4.080 V
CPU +1.8V : 4.080 V
CPU +1.2V : 4.080 V
CPU +1.0V : 4.080 V
DDR-II +0.9V : 4.080 V
Battery Status : Not Installed
[Enclosure#1 : SAS RAID Adapter V1.0]
[Enclosure#2 : Areca ARC-8018-.01.06.0106]
=====================================================
GuiErrMsg<0x00>: Success.
Last edited:
You are dealing with a card that is coming up on 13 years old, about (or just before) the time that consumer SATA SSD's were hitting the market. If there are issues you are going to have to deal with them, upgrade to a newer card or use something on the HCL. The problems have to do with the Expander on the 12/12+ cards, similar to issues with the HP Expander had with 1.5/3 SATA negotiation. Areca is not going to put any effort into updating the firmware to deal with issues on a card so old, and technically any drive not on the HCL is suspect for full capacity/speed/usefulness when it comes to enterprise HBA's.
Polynomial-C
n00b
- Joined
- May 5, 2020
- Messages
- 4
Perhaps I didn't express myself good enough. All I wanted to tell is the reason for the 1680 cards with SAS-Expander setting the SATA link speed for (perhaps not even all, but mine were all affected) SSDs to 1.5G and not 3.0G.
And yes, Areca stopped supporting this card a long time ago. But the SAS-Expander's firmware upgrade I was talking about can be found on the official Areca FTP server. There's even a readme.txt file in the very same directory containing a changelog and there you can find the following entry:
There's also documentation on how to perform the upgrade but it's quite awkward to handle and like I said, you risk bricking the card.
More recent Areca cards seem to not have this issue so perhaps it's better to upgrade the card in case you still have 1680s in use. But I could not afford new Areca cards for a very long time and had to stick with my three ARC-1680ix-12 and was quite happy when the new SAS-Expander firmware finally allowed me to have 3.0G SATA link speed with my SSDs.
And yes, Areca stopped supporting this card a long time ago. But the SAS-Expander's firmware upgrade I was talking about can be found on the official Areca FTP server. There's even a readme.txt file in the very same directory containing a changelog and there you can find the following entry:
Code:
D)Set the initial Min. speed to 3.0G. for some 6G SATA HDD negotiate as 1.5G.More recent Areca cards seem to not have this issue so perhaps it's better to upgrade the card in case you still have 1680s in use. But I could not afford new Areca cards for a very long time and had to stick with my three ARC-1680ix-12 and was quite happy when the new SAS-Expander firmware finally allowed me to have 3.0G SATA link speed with my SSDs.
Last edited:
It's been a long time since I dealt with an -IX version, but IIRC the "awkward" portion is simply setting up the proper serial (COM port) connection to the expander command-line interface. After that, as with most command-line flashing procedures, it's not complicated but you do have to proceed carefully.
If you have the RJ-11 to DB-9 cable that came with the card, try using a cheap USB-to-serial adapter and see if you can establish a connection. Just try a 'safe' command. e.g. the SY (system info) to see if it works. If you can't establish a connection, one wrinkle might be that the USB adapter communicates using 5V signals while the expander expects 12V (the standard). In that case, the easiest might be a cheap $20 PCI-E to serial port adapter OR if you happen to have an old-ish motherboard that has a COM-port header on-board. After that, (AFAIR) it's as simple as two commands (erase flash, flash new firmware).
I have a server running Windows 10 Pro with an Areca 1882 controller with some RAID 60 volumes.
I have stood up a Hyper-V cluster and need to give the VMs hosted there access to the RAID 60 volumes via iSCSI.
I thought it would just be a matter of installing the iSCSI initiator service and then add the 2 RAID 60 volumes, but the initiator does not see the volumes. Perhaps it is only possible to create a target on windows server?
Any tips on how to create iSCSI targets for shares on an Areca controller on a Windows Pro box?
Thanks!
I have stood up a Hyper-V cluster and need to give the VMs hosted there access to the RAID 60 volumes via iSCSI.
I thought it would just be a matter of installing the iSCSI initiator service and then add the 2 RAID 60 volumes, but the initiator does not see the volumes. Perhaps it is only possible to create a target on windows server?
Any tips on how to create iSCSI targets for shares on an Areca controller on a Windows Pro box?
Thanks!
Last edited:
Win 10 desktop versions don't have native iSCSI target creation support, but you can use free third-party software to do so, e.g. https://www.starwindsoftware.com/starwind-virtual-san-free
I have an ARC-1883i with three RAID-5 arrays (4 drives per array, 4-8 TB each, via Intel RE2SV240 expander) AND an ARC-1880i (on another machine) with one directly connected 4-drive RAID-5 array (4TB). All arrays are 64-bit LBA with a 128k stripe size.
I need to do some maintenance and have a spare, old ARC-1680i. Will the ARC-1883/0 arrays (and the Intel expander) temporarily work using the 1680? I know that the latest 1680 firmware is from 2012 and may be lacking large drive support.
I need to do some maintenance and have a spare, old ARC-1680i. Will the ARC-1883/0 arrays (and the Intel expander) temporarily work using the 1680? I know that the latest 1680 firmware is from 2012 and may be lacking large drive support.
Hi Blue Fox,
I've been away from the storage world for a few years and looking to upgrade. Could you give a few examples of the higher-performing Broadcom cards these days (or preferably, their OEM flashable versions)? I assume that the better performance comes from the ARM processors and/or higher native channels on the newer SAS35xx/SAS39xx series?
And from the specs (PowerPC), the ARC-1884 series appears to be using the SAS3316?
Pretty much any newer Broadcom card will do it. 9460-16i/9560-16i for example, with the latter good for 13.7GB/s sequential and 3M IOPs, not to mention NVMe support. You can get cheaper OEM versions of those as well too (at least I've seen Dell versions of the 9460).
The Areca 1884 line is a bit newer, but they only released 2 models, and I've never come across one in person. As much as I like their products, I feel like their days are probably numbered unfortunately.
The Areca 1884 line is a bit newer, but they only released 2 models, and I've never come across one in person. As much as I like their products, I feel like their days are probably numbered unfortunately.
Polynomial-C
n00b
- Joined
- May 5, 2020
- Messages
- 4
Well, the documentation is solely based on Windows + Hyperterminal. So I had to figure out myself how to perform this with the right tools on Linux. I ended up bricking an ARC-1880ix-12 card because it requires different steps to flash the SAS-Expander than on older ARC-1680ix cards.
Sorry to hear about that. We were talking about the 1680 series -- it would seem natural for a different model with a different expander chip to have a different flashing procedure. Doesn't the 1880 have it's own Expander CLI manual explaining the steps?
It might have been better to ask here first if you were unfamiliar (I think I used minicom in Linux but there may be better options now). The Areca's are essentially enterprise hardware, and the basic serial communication steps would be familiar to anyone who regularly works with such hardware regardless of the OS used.
Please do contact Areca support about the borked 1880-ix-12 and see if they can help you out. Even if needs to be sent in for service it should be a simple fix that can probably be done at their US distributor (Tekram) instead of going back to Taiwan.
Polynomial-C
n00b
- Joined
- May 5, 2020
- Messages
- 4
No need to be sorry. In the end it was my own fault although I am under the impression that the documentation for upgrading a SAS-Expander firmware in ARC-1880ix cards is worse compared to the documentation from ARC-1680ix card series. Yes, I did read the documentation for the ARC-1880ix cards but some details were unclear to me so I tried to fill these with what I knew from my successful ARC-1680ix upgrades. And that was my huge mistake.
In case someone wants to know, I ended up using "cu" to communicate with the SAS-Expander from Linux and I used "sx" to transfer the firmware files to the SAS-Expander and at least with ARC-1680ix cards, that works very good.
The bricked card has already been replaced. I had to get in touch with one German Areca distributor (yes, I am from Germany
) and they sent the bricked card to Taiwan. I did not get back the same ARC-1880ix-12 card but a completely overhauled one (the serial number is different and the card has a single big heat sink instead of the two small ones that were on my bricked card). Overall I had to pay more for the "repair" than I had spent for the purchase of the card but well... it was my own fault so I had to learn that the hard way.Does the 9560 support tiering like a SAN does? So you would have your HDD tier and a SDD tier. You would just carve LUNs (volumes) out of it and specify the redundancy at the LUN level. The card then manages which blocks of files (not files) are on each tier to optimize performance. You don’t worry about RAID Level or anything when you set it up. You just add disks and create LUNs and it handles all of the other stuff for you. Will the 9650 do that?
I currently have an Areca 1882 12i with 24 8TB drivers in a RAID60 (2 x 12) and 24 6TB drives also in a RAID60 for 2 large volumes.
In addition to that, I'm running a 3 Node S2D cluster with each Node having 4x NMVe drives. While the VMs have very fast access to the Cluster Storage (seeing about 5,000 MB/s reads and 1,200 MB/s writes) S2D forces a 3-way mirror with 3 Nodes. So I'd love to setup tiered storage as explained above where the VMs would hit the fast NVMe storage.
I didn't even see this reply, Thank you @Polynomial-C, other than the obvious of using a pcie card in the system how I would I test it?
Everything seems to be running and its also inconsistent. most of the time it shows normal.
Within the same brand it should be ok. I was able to read the array from my 1220 with my new 1883. I did have some oddities with performance, that were resolved after formatting the drives and remaking the array later once all the data was fully backed up, never figured out why this was. I am guessing it was due to how the 1220 made arrays vs the 1883, but it did work and I didn't notice it outside of this test (note this image contains the old array from the 1220 on the left and a different new array on the right.:
Thanks for the response, I am not really worried if i had the same brand, my biggest concern is my card is dead, I can't do anything, windows will not boot with it in the system, a fresh install it works but then I go and install the driver and it just keeps saying "installing driver"
I can get ubuntu to at least boot but cant get the driver installed either. I also had already wiped the drive it was originally working on so I dont think I can see any logs.
Even tried reflashing the firmware but it just freezes the system.
cant use the rs232 either, I just dont get how a card can just die like that.
It sucks, but all hardware fails. Have you tried the card in another computer? It sounds like a hardware failure as you've tried different operating systems too. Of course you can give it another shot by just removing it and plugging it back in, sometimes a card can inexplicably can get just loose enough to present issues. I myself just had a random reboot a half hour ago on the server, and I am suspecting it may be my card because the web interface just shows a white page when going to half of the pages including the log.
I have tried it in another system, ugh, same issue so its definatly the card, guess its time to see what i can do to access my raid.
thank you again.
Snowknight26
Supreme [H]ardness
- Joined
- May 8, 2005
- Messages
- 4,434
Shouldn't be an issue.
I have pretty reliable underground electric service in my neighborhood, so I fearlessly started a week-long expansion from 6 to 7 each 14TB Western digital gold drives in RAID6 on an 1882ix-16. Without the battery backup module for the raid card connected, as it has given me problems at boot in the past. I don't have a backup UPS that powers my server for a few minutes supporting a clean shutdown of the Windows host PC anymore either, as those have been a headache in the past. Battery backups to my server PC haven't gone well.
During the expansion a car plowed into the big high-voltage pole a mile away causing my power to go out. When I turned the PC on after power had been restored, it resumed expanding the array as if nothing had happened. Once this happened and survived, I rebooted cleanly once more during the array expansion due to an urgent software problem. It all seems good now, and is scheduled to automatically check the array periodically so I'm sure it will notice if there are any errors. But I suspect the data are all good.
Obviously I don't recommend pulling the plug on your PC during a rebuild, and some might think it wise to be more rigorous about battery backup. The issues I've had were expensive APC brand UPS failing their electronics. Of course SLA batteries are a maintenance item, and I replaced these, but it is frequent, a chore and expensive. I think I'll wait for lithium UPS. My Areca battery backup module was keeping the PC from booting due to some error supposedly having to do with the battery RAM having backed-up data to write to the drives, but it couldn't. Crazy stuff.
Thanks for this big picture update. I don't know about the rest of you, but I don't think about replacing/upgrading my RAID card (1882ix-16) very often, this is my second card since 2004. It's a big undertaking involving a lot of $$ for the card and of course a new set of drives, and the practicality of running simultaneous dual arrays for the massive data transfer. Thankfully obsolescence isn't video card-like for RAID.
It's sad to hear about Areca vs. Broadcom, but it is good to be informed. I wouldn't be surprised if Broadcom was able to wield some custom ASIC silicon while Areca may have been programming off-the-shelf processors and FPGAs.
Help!
I just set up a new raid 5 in my system. I did it with my old raid 5 physically disconnected in case I did something stupid. I've initiated the raid, but now with the old raid connected again the new raid is showing, but the old one has gone. It's set the new raid up with the same ID.
edit: I changed the volume and raid set names, I've found where to change the CH/ID/Lun, but dont know which to change.
I just set up a new raid 5 in my system. I did it with my old raid 5 physically disconnected in case I did something stupid. I've initiated the raid, but now with the old raid connected again the new raid is showing, but the old one has gone. It's set the new raid up with the same ID.
edit: I changed the volume and raid set names, I've found where to change the CH/ID/Lun, but dont know which to change.
Last edited:
good to know! thanks. I've got a UPS (and a battery on my card), but I usually switch off my system during lightning storms (and unplug everything): I've lost 2 systems to lightning!
I have a problem with Areca 1880-i on Linux. I have a script that uses cli64 to monitor drive temperatures, polling every 3 minutes. This works fine, but after a few days, cli64 stops responding, as does archttpd. The web ui and telnet continue work from the management port.
I'm on the current version of both utilities (cli / archttpd). There is nothing unusual in the system logs. Everything seems fine.
I'm not sure when this started happening, but it seems like when I upgraded to ubuntu 20.04. Maybe its the new kernel or some services interaction on the system? I'm not sure how the utilties connect to the card, is there something in /proc I could look for?
I'm on the current version of both utilities (cli / archttpd). There is nothing unusual in the system logs. Everything seems fine.
I'm not sure when this started happening, but it seems like when I upgraded to ubuntu 20.04. Maybe its the new kernel or some services interaction on the system? I'm not sure how the utilties connect to the card, is there something in /proc I could look for?
Hi, wanted to see if someone can help me with my Areca 1880X RAID6 array which is in a degraded state after I accidentally turned off my Supermicro JBOD enclosure. When I turned the JBOD enclosure back on, a bunch of the disks in the array were marked FAILED and the array was marked FAILED. I manually activated each disk and rebooted and now, the array is accessible, but in a degraded state because two of the disks are marked FREE. If I remove these two disks and reinsert them, they come back on as belonging to a different Raidset with the same name. I tried rebooting, but after the reboot, instead of combining the two Raidsets into one, the two disks are marked free again. Also, even though I have a hot spare (only one hot spare, however), the rebuild process has not begun.
Ideally, I would like to combine the two Raidsets to get back a Raidset in NORMAL state, but being able to do a rebuild with the two "FREE" disks would be good too.
Any ideas help will be greatly appreciated.
Thank you.
Ideally, I would like to combine the two Raidsets to get back a Raidset in NORMAL state, but being able to do a rebuild with the two "FREE" disks would be good too.
Any ideas help will be greatly appreciated.
Thank you.
Hello Everyone!
I am happy owner of an ARC-1880, Firmware V1.49, for many years. I find myself needing to build a new (much larger) volume. I'm looking for 10TB or greater per disk. RAID6. Looking to grow to roughly 7 drives with global hot spare. I have several questions...
1. Is anyone aware if this card supports (or has issues with) newer high capacity disks?
2. Are there HDD brand/capacity recommendations or limitations?
3. I currently have 7x2TB drives, allowing space for 5 more drives in my current chassis. Is there a preferred method to building this larger capacity volume and migrating data over?
I am happy owner of an ARC-1880, Firmware V1.49, for many years. I find myself needing to build a new (much larger) volume. I'm looking for 10TB or greater per disk. RAID6. Looking to grow to roughly 7 drives with global hot spare. I have several questions...
1. Is anyone aware if this card supports (or has issues with) newer high capacity disks?
2. Are there HDD brand/capacity recommendations or limitations?
3. I currently have 7x2TB drives, allowing space for 5 more drives in my current chassis. Is there a preferred method to building this larger capacity volume and migrating data over?
I still use two ARC-8040 boxes which deploy the ARC-1880 card. One of them is 36TB (8x HGST HUS726060AL5210 ) and the other is 18TB (8x SEAGATE ST33000650SS)., i.e. both are using fairly contemporary high capacity HDD models and they work fine. You should check the HDD compatibility document for the ARC-1880 card and look for similar drives to the ones you want to use in terms of interfaces and features. The drives I'm using are not on the compatibility list literally, but they have been working fine for years. Good luck.