[SOLVED] New Build issues with Read Errors and a missing disk


Recommended Posts

I have a new build that has been giving me problems from the beginning.  I finally switched over to a Dell H310 SAS HBA w/ LSI 9211-8i in IT Mode and have changed/re-seated cables a few times per the other posts I have read here. I am getting a huge amount of read errors.  I am using a Rosewill RSV-L4312 4U case and am leaning towards the bays being bad.  To diagnose this I was going to bypass the hot swap bays and hook the drives up directly to the HBA.  Any other ideas? Bad memory, hardware too old?

 

I have attached the diagnostics files. I hope that is enough to diagnose my issue.

 

 

tower-diagnostics-20200115-2110.zip

Edited by Capt_Rekt
Link to comment
2 hours ago, johnnie.black said:

That's a good idea, also test with a different PSU if possible.

Since these are shucked WD drives I have ordered some monoprice molex to sata power adapters, should be here tomorrow so I can test tomorrow night. The PSU I am used is pulled from a working system, as was the MB/CPU/ram. The PSU is a Senty XPP725-HS (725w), I am assuming this would be enough power to run the setup. I am open to a larger PSU if needed. 

Edited by Capt_Rekt
Link to comment
5 minutes ago, johnnie.black said:

Some of those drives can have issues if supplied with 3.3v, google "wd sata 3.3v"

I have seen that, ie the fix with the kapton tape.  I figured that since my chassis was powered only via the molex connectors I wouldn't have an issue with the 3.3v. I can always order the tape as well. 

Link to comment
51 minutes ago, johnnie.black said:

some molex powered backlplanes still have 3.3v on SATA.

I forget which exact ones I have, but I know mine do. they have multiple SATA and 4 pin power points, and even using only the 4 pin they still kept my 8TB WD from spinning up. The circuit boards in them evidently do some power conditioning instead of just passing the supply straight through.

 

They are fairly old now though.

Link to comment
36 minutes ago, jonathanm said:

I forget which exact ones I have, but I know mine do. they have multiple SATA and 4 pin power points, and even using only the 4 pin they still kept my 8TB WD from spinning up. The circuit boards in them evidently do some power conditioning instead of just passing the supply straight through.

 

They are fairly old now though.

I ordered some kapton tape and some new power adapters, they should both arrive by tomorrow.  I also sent a message to Rosewill support asking if the backplane provides the 3.3v power. I just hope that all of this troubleshooting hasn't damaged any of my drives. All the SMART tests I have run have come back passed.  

Link to comment

Just wanted to follow up with this for anyone else who has has the same issue. After replacing cables (no solution), bypassing the Rosewill backplane (no solution) and finally changing the PSU (solved) my read errors are gone. I was able to run the read check and then rebuild the parity over the weekend. The power supply I was using was not able to support my build. Since the Rosewill backplane requires 2 4-pin molex connectors per drive cage so 6 total.  The Senty XPP725-HS (725w) just wasn't cutting it. That PSU only had 3 4-pin molex connectors, so I had to split them to power the cages.   I ordered and installed and EVGA SuperNOVA 850 G+ fully modular PSU, it came with cables to power up to 6 molex devices. I have two cages powered. Tonight I will attempt to power all three and install a cache drive. Thanks for all the help!

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.