dtlokey Posted November 7, 2014 Share Posted November 7, 2014 I recently went through the process of trying to fix a gremlin that still seems to exist with my unraid server. I have no parity drive installed. the issue i originally had was with a HDD constantly erroring out when i'd try to write files to it. i did a smartctl long test and found no issues with it. swapped cables, etc got to the point where once i rebooted it wouldn't even allow me to add the drive back to the array... long story short i ended up swapping motherboards and changing out my sata controller card and bought a few new drives to replace the one i thought was bad (even though it worked fine for weeks hooked up to my windows 7 machine) fast forward to today. new board in, new controller card, new hdd along with old disk, no parity still i just wanted to get my server up and running again. everything worked fine for about 2 weeks but then a couple days ago on the new drive (which i precleared multiple times prior to adding it to the array) starts giving me a massive amount of errors and won't allow me to add new files to the new drive. (i had already added ~200GB prior to this issue to this drive) i'm at a loss i have the same issue as before on a new drive with new hardware. the only old remaining parts are memory, flash drive, PSU, other original HDD's, and processor. this issue is driving me a bit up the wall as my server was rock solid for years until about 5 months ago or so. I currently running Unraid 5.0 Asus M5A78LM LX Plus 4GB DDR 3 1333 Thermaltake TR2 600W PSU Vantec UGT-ST644R PCIe 4 channel sata raid host card AMD FX 6300 1x2TB Samsung 5x2TB WD Green 2x2TB WD Reds (both of these were giving me issues, now only one is) PSU has 6 sata connectors so the other two drives are running off molex to sata adapters i recently attempted to run a smart report on the WD red giving me issues now (drives only been in use for a few weeks now) and i get this message when attempting to run it root@Media:~# smartctl -a -A /dev/sdd smartctl 6.2 2013-07-26 r3841 [i686-linux-3.9.11p-unRAID] (local build) Copyright © 2002-13, Bruce Allen, Christian Franke, www.smartmontools.org === START OF INFORMATION SECTION === Vendor: /3:0:0:0 Product: scsiModePageOffset: response length too short, resp_len=47 offset=50 bd_len=46 >> Terminate command early due to bad response to IEC mode page A mandatory SMART command failed: exiting. To continue, add one or more '-T permissive' options. i've attached the sys log for when the errors occur please help me figure out what these issues stem from syslog.zip Quote Link to comment
dtlokey Posted November 10, 2014 Author Share Posted November 10, 2014 anyone have any ideas? Quote Link to comment
Frank1940 Posted November 10, 2014 Share Posted November 10, 2014 Read this: http://lime-technology.com/forum/index.php?topic=36065.0 An inspection of your cables will tell if this might be an issue. Quote Link to comment
dtlokey Posted November 10, 2014 Author Share Posted November 10, 2014 thanks, just checked and the cable i have does indeed have the bump for the drive giving me most of the issues. though it's not a latching style cable. i am running 9 drives total on a PSU rated at 600W but not on a single 12v rail, could that be causing these issues? that was going to be my next move if the logs didn't point to anything obvious Quote Link to comment
dtlokey Posted November 10, 2014 Author Share Posted November 10, 2014 i also just noticed that this is the only disk in my array that wont spin down with the others.... Quote Link to comment
Joe L. Posted November 10, 2014 Share Posted November 10, 2014 i am running 9 drives total on a PSU rated at 600W but not on a single 12v rail, could that be causing these issues? Yes, it could easily cause the issues you are having. Most multi-rail power supplies use only one of the rails to power all the hard disks, and they frequently share it with the motherboard as well. The other rails are used for power-hungry video cards used by gamers. Many users have reported issues when going over 6 or 7 drives on a multi-rail supply. Since your PS rails are rated at 20 Amps, and 24 Amps, but we have no way to know which is used for your disks, it might be why you did not have issues until you went to 9 disks. Joe L. Quote Link to comment
dtlokey Posted November 11, 2014 Author Share Posted November 11, 2014 i am running 9 drives total on a PSU rated at 600W but not on a single 12v rail, could that be causing these issues? Yes, it could easily cause the issues you are having. Most multi-rail power supplies use only one of the rails to power all the hard disks, and they frequently share it with the motherboard as well. The other rails are used for power-hungry video cards used by gamers. Many users have reported issues when going over 6 or 7 drives on a multi-rail supply. Since your PS rails are rated at 20 Amps, and 24 Amps, but we have no way to know which is used for your disks, it might be why you did not have issues until you went to 9 disks. Joe L. thanks, i'm going to grab a Corsair Cx 600M and go from there if it's any help here's the smart report (long version) that i just did once i receive the PSU i'll post results. smartl.txt Quote Link to comment
Frank1940 Posted November 12, 2014 Share Posted November 12, 2014 I am not a real expert on interpreting smart reports and I am only answering because no one has taken the time to respond. I suspect that the drive is good. I seem to recall that cabling is often the issue in these types of cases. You might want to replace the SATA cable to the drive in question just as a precautionary measure. (As an added benefit those cables are very cheap.) If there are still issues, you might also want to eliminate the SATA/RAID card by switching the drive to the unused port on the card. By the way, that card does seem to have some reliability issues. (14% of the newegg purchasers reported problems with operating irregularities during normal operation. The sample size was not large with only 21 total users reporting.) Also, have a good look at the power cable side of the equation. Don't daisy-chain multiple power splitters off of the same molex connector. Double check that all of the wires in any splitters are securely terminated in their connectors by pulling gently (but firmly) on each wire. Quote Link to comment
dtlokey Posted January 23, 2015 Author Share Posted January 23, 2015 Long time since i've posted this issue, got a new PSU and also wanted to give it a good amount of time and use before i posted my issue as resolved. turned out to be the PSU causing the issues i was having. i've not had one issue since now that i have a PSU with a single 12v rail and it's been aprox 2 months and change since going that route. seems that was the solution to my issue. Quote Link to comment
Recommended Posts
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.