OCZ Agility 3 drives keeps on dropping

[Asvin]

My main operation system drive is of Samsung Spinpoint drive which is 1TB in size. As a gaming drives in a Raid 0 config I bought a new OCZ Agility 3 drive, I already tired my hands with the OS drive but it was too unreliable so therefore decided to go for OCZ drive but seems like even this drive not proper for my system. The OCZ drives keep on dropping without fail randomly. Not just the OCZ drive but also the other defaults Samsung drive keeps on dropping in multiple configuration. There is no doubt that both the drives are update with their respective latest firmware updates. I have noticed many a times they playing with the Raid, for a moment they are considered in Raid and the very next moment they are not and when i try to load or use it single then sometimes it loads and sometimes it doesn’t. I thought the issue was due to the cable but after I changed the cable same behavior was noted with both the drives. Initially I thought that only the OCZ drive is causing this problem so therefore I declared it as a default one but then I noticed the same behavior in other one too. I am using a Gigabyte board. Any help or suggestion?

[Denholm]

Just need some more detailed information about the necessary things, can you let me know whether your drive is running on a Intel chip with RAID set in BIOS. Most probably the issue should be due to the cable but as you have already mentioned that the cable is nowhere causing any such problem and everything is fine from the cable’s side so therefore I can predict that yours is bit complicated one. If possible let me know your CPU-z details and speedfan or anything that could probably help me in knowing your temperature and voltage. As according to my knowledge I guess if ever they get defected intel software somehow force a rebuilt when they are visible and even if nothing is wrong the problem is showed in the normal RAID.

[Dannai7]

Yeah you are right I have seen it in my machine, I have it sometimes happening on a RAID 5x5 running. Whenever I use to run an overclock thinking that everything is fine and alright my disk fails and after that even if I reset it to stock I still encounter a failed disk. The only reason I am letting you know about this is because I guess things are going wrong with the users who are running 2x00k's. I have already researched a lot on overclocking the gigabyte z68 boards but then however I ended up having attention to the low loads and also having a stable overclock. I myself have currently two Vertex 3s running onto my gigabyte z68 very much fine and well but let me tell you that none of them in on RAID. And as far as the cable is concerned I need not have to worry about them as it is going to be the intel Raid thing. Anyways the only thing now I wish to tell you is that your drivers are completely OK and you need not have to worry about it, even if you connect them to my machine they are going to work well. Other than this you have really a reason to worry about the bug that has hit on you machine and causing this problem.

[Ataa]

OK, so if you get the solution for this problem even I will be out of the same. Yes, you guessed it right. Even I am having the same problem and I am not you out with a perfect solution. Don’t know what exactly is causing this problem. At least in my case the problem should have been existed in the cable but I guess I was not so fortunate to solve my problem so easily. I too tried changing the cable which as expected didn’t show any difference but then other this I also tried two different thing:

• I checked all my drives on the Marvell 6Gb controller separately testing them, and raid configuration
• Also tried them on a standard controller in both a standalone, and raid configuration (in both IDE and AHCI config)

Both the tricks failed and ultimately I landed at no man’s land.

[Faiyaz9]

Hey I think if you try to stick your default Samsung and your DVD on the 2 MArvell lines then might be you can get some help. What exactly happens is that it somehow cleans the things a little bit and then helps you to simplifying troubleshoot. For my own system I have 2 Vertex on the intel and also I have other two drives on the Marvel and let me tell you that I haven’t tried a RAID plus other drives on the Intel SATA yet. I think you can also try checking the your board’s i.e. gigabyte notes on their BIOS releases, I am sure there might be at least a release with a line particularly for the OCZ drives.

For getting the Raid back you can simply enable the screen for the intel Raid boot ROM for just showing up the screen, after that you can press Alt+T. when you seen the ROM boot screen for the RAID just have a look at it before you try to get into the windows. I will somehow insist you on one rebuilt.

[Asvin]

Currently I am not overclocked. Yeah but I tried it bit once with the processor and ram speed the first time when I built my system but due to some reason I turned it back down to the default position fearing that I will lose something important. I performed the overclocking with the help of the gigabyte software. Anyways my temperature have always been at control as I have been using a huge cooler and also very well vented fan case.

Another thing I would like to tell is that I have been using the V11 version of BIOS which is now I guess the revision based the noted extended OCZ SSD support, now I have decided not to go with the Beta. Surely go with one of the suggestion and currently the best suited one is to go with using the SSDs on Marvell Controller. Just to get a solution to keep them from dropping down always. Also I am very much familiar with the intel screen which has been referred above to Ctrl+I, it actually states that the drive has been failed on the screen.

[ErikaGeek]

Also I am very much familiar with the intel screen which has been referred above to Ctrl+I, it actually states that the drive has been failed on the screen.

Are you sure it shows that the drive is failed or it just can’t show you up the drive on the screen? If it shows that the drive is failed and it can still show on the screen then a simple rebuilt can clear the issue I guess but if it can’t see it then surely you will have to go with next total Clear CMOS then load optimized defaults, power down, reboot and then see whether it sees it or not. If you do it by cable down or power out the back it would do a clear corrupt CMOS, this is the only thing that I could think of currently. Look if the drive is failed that doesn’t mean that you can’t see it, so please don’t combine these two thing together. You can also go for switching the cables. However it sounds as a RAID issue which happens just because of the missing TLER bit and because of the drives falling off a RAID due to the lack of time.

[fflygirl]

If I would have been in your place instead of breaking the head in getting into the root cause and discussing it so much I would have simply tested the drive if I would have another machine. Simply sitting around the machine and then opening the side and plug in the failed drive into the SATA line can show you up the actual status of the drive. If you can probably see the drive then the other system is having some trouble and if not then the drive is faulty. I think the intel version somehow doesn’t allows a rebuilt without losing the data. In my case to the drive use to reappear after powering down but then it seems to be losing down one of them as the time passes. During such problems and issues its better to call up the tech support they are the best one to help you out. When I called up the tech support they suggested me testing down the drives separately, I however had one drive with the O/S installed but then started thinking that those two drives don’t just like to be on the same controller together. Also what I did was benchmarked the drive on both the marvell and intel controller and from this I found that the intel controller was getting lot better speeds.

[phunsuk]

I too encountered BSOD on my main drive, the incident occur when I was trying to transfer larger number of huge files. I was also trying to perform some activity in the background anyways I just believe that this issue has nothing to do with these other background activities. With the help of other controller I simply added the other SSD on to it and it somehow worked. Initially when I was having other drives installed in my machine they use to somehow get disapper, however I am not sure whether it will do the same on a separate controller but that doesn’t matter to me as I am not supposed to use the drive at its half performance. I am really willing to try this drive on the same controller but worried with the fact that this controller has a bad history of making the drives disappear in the BIOS. I have a regular SATA drive which is already being connected to the port. However the system got stuck while loading the operating system when I tried to start it up but I failed to see whether this time my BIOS recognized the drive or not as I was busy with some other activity and within that period of time the screen went off within few seconds.