May 10, 201214 yr unRAID pro system. 5b14. BIOSTAR A55MH / AMD 3400 / OCZ 650 PS / 4gb ram. Ok I had a system up and spinning with 4 WD-EARS drives that I pulled from the last of my WHSv1 servers. I did have 5 but after one of them was borked enough that after I got things moved off and put them in the unRAID box it would not recognize it, tried to run WDDiags on it and it failed in seconds.. of off to WD for an RMA. So while waiting for that, put the other 4 on preclear (all 4 running at once, 3 passed to beat the server up too... ) preclears ran fine (couple of them had 2 or 4 reallocated sectors, but for drives that had been spinning 24/7 for about 16 months, that's to be expected I guess). Still waiting for the RMA to come back, so I moved over 6tb from the WHSv1 (still had 4 drives in it) with TerraCopy, took forever, but everything moved over with no errors. Now the RMA arrives, sent a 2tb EARS, got a 2tb EARX with no jumper (non of the others had them either, those were put in action before WHS would handle AF drives, so all unjumpered)... put it in the box, connected to motherboard port.. 3 passes of pre-clear and NOT accessing any files on the box while it does it. Days pass.... (grin) the Preclear finishes on the new drive, all good. So I stop the array, assign it, start it up, all looks good. Start to copy another 2tb or so worth to that drive (gonna get that thing empty eventually). Goes fine for about... 20 minutes. Then it slows down.. WAY DOWN.. see errors in the log about stripe errors, then the whole array freezes up and unresponsive. TerraCopy gets 'network name unavailable' and hangs up. Finally hit the reset button and comes up... it starts a parity check.. getting HUNDREDS OF ERRORS (note: the ONLY drive I was copying data to was the new one, sde.) I stop it and shut down to check things out. Start checking cables on the new drive. Power looks good, SATA cable on the drive is ok, I tug on the m'board end of the cable (locking cables) gently and (locking cables)... the cabble come out in my hand. The cable, NOT THE CONNECTOR! Oops. ok get that out of the board socket, get a fresh cable (tugging on the connectors first to make sure .. all good). Restart the array.. and it starts the parity check again... starts getting errors in the same places? Hmmm ok what ever happened may have messed up the area it was writing on the drive Stop the parity check. Start up BeyondCompare on the data on the other 3 drives (I copy no move till I"m sure all is good) doing the CRC compare. Everyfile compares exactly. No issues. Boot from a DOS stick. Ran WDiags. No errors. Restarted the system, it starts the parity check again, stated getting errors again. Stop it and run smarts on all the drive. No errors, same as since the preclear on them. Ran a LONG SMART on the new drive. Again, no errors. I deleted off the files from the new drive I had started to copy to and run a NOCORRECT Parity check. AGAIN HUNDRED OF ERRORS. But I let it run. 6 hours later. NOW it says. Parity is VALID, Last checked on Wed May 9 05:24:08 2012 MDT (yesterday), finding 934 errors. Huh? Ok I'm starting to copy to the new drive again, so far it's gone WAY past where it was before. No errors in the log. but it seems to be going slow (about 12-20mb/s). Questions. 1) Most likely the problems were a stinky cable, but how did it pass the pre-clear with a bad cable? 2) Parity is VALID? with 934 errors? Yet BeyondCompare shows all files from before match. 3) Should I have dropped that drive out of the array, reformatted it and put it back before copying to it? the current log (the last parity check showing errors) is available, the others are not. Any words of wisdom?
May 10, 201214 yr Author no, 0 errors on all drives, just the comment about parity being valid.. with 934 errors.
May 10, 201214 yr It sounds like you confirmed the data is OK and the disks are appearing healthy so run a correcting check to fix the errors.
May 10, 201214 yr Author Parity Correct then, ok. I'm gonna finish reloading the new drive (that SHOULD rewrite the sectors that may have gotten borked before with the funky cable (still trying to figure how it passed the pre_clear). any thoughts on the really slow speed? I'm only getting 17-22mb/s on the TerraCopy write and the same on the Verify. I may be paranoid and run the BeyondCompare again after I copy the rest of the drive over. Actually I think I'll run another parity check (no correct) after the data is copied.. then BeyondCompare, then the Parity Correct.
May 10, 201214 yr Author and this is the current eth0 info (as of a few minutes ago, realtek nic). Been restarted and copied about 110gb over with 5 dropped packets, still only writing at 16gb/sec. NIC info (from ethtool) Settings for eth0: Supported ports: [ TP MII ] Supported link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Supports auto-negotiation: Yes Advertised link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Advertised pause frame use: Symmetric Receive-only Advertised auto-negotiation: Yes Link partner advertised link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Full Link partner advertised pause frame use: Symmetric Receive-only Link partner advertised auto-negotiation: Yes Speed: 1000Mb/s Duplex: Full Port: MII PHYAD: 0 Transceiver: internal Auto-negotiation: on Supports Wake-on: pumbg Wake-on: g Current message level: 0x00000033 (51) Link detected: yes NIC driver info (from ethtool -i) driver: r8169 version: 2.3LK-NAPI firmware-version: N/A bus-info: 0000:02:00.0 Ethernet config info (from ifconfig) eth0 Link encap:Ethernet HWaddr 00:30:67:e5:e4:ed inet addr:192.168.1.22 Bcast:192.168.1.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:90597346 errors:0 dropped:5 overruns:0 frame:0 TX packets:74361872 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:1689776922 (1.5 GiB) TX bytes:2958192311 (2.7 GiB) Interrupt:42 Base address:0x2000
May 10, 201214 yr Author Well that's the only anomaly I've seen on the network since changing the SATA cable. 5 packets out of > 250gb transferred.
Archived
This topic is now archived and is closed to further replies.