March 7, 201313 yr So disk14 went offline a few days ago, I finally got a replacement drive, stuck it in, and hit rebuild, and now drive 13 is having a cow. The syslog as you can imagine is insanely large at this point... just containing this Mar 6 18:30:01 Tower kernel: md: disk13 read error Mar 6 18:30:01 Tower kernel: handle_stripe read error: 124409736/13, count: 1 Mar 6 18:30:01 Tower kernel: md: disk13 read error Mar 6 18:30:01 Tower kernel: handle_stripe read error: 124409744/13, count: 1 Mar 6 18:30:01 Tower kernel: md: disk13 read error Mar 6 18:30:01 Tower kernel: handle_stripe read error: 124409752/13, count: 1 Mar 6 18:30:01 Tower kernel: md: disk13 read error Mar 6 18:30:01 Tower kernel: handle_stripe read error: 124409760/13, count: 1 Mar 6 18:30:01 Tower kernel: md: disk13 read error Mar 6 18:30:01 Tower kernel: handle_stripe read error: 124409768/13, count: 1 Mar 6 18:30:01 Tower kernel: md: disk13 read error Mar 6 18:30:01 Tower kernel: handle_stripe read error: 124409776/13, count: 1 Mar 6 18:30:01 Tower kernel: md: disk13 read error Mar 6 18:30:01 Tower kernel: handle_stripe read error: 124409784/13, count: 1 Mar 6 18:30:01 Tower kernel: md: disk13 read error Mar 6 18:30:01 Tower kernel: handle_stripe read error: 124409792/13, count: 1 Mar 6 18:30:01 Tower kernel: md: disk13 read error Mar 6 18:30:01 Tower kernel: handle_stripe read error: 124409800/13, count: 1 Mar 6 18:30:01 Tower kernel: md: disk13 read error Mar 6 18:30:01 Tower kernel: handle_stripe read error: 124409808/13, count: 1 Mar 6 18:30:01 Tower kernel: md: disk13 read error I have another drive ready to go in, but not sure what I should be doing at this point...
March 7, 201313 yr Author Don't like the idea of bumping my own thread, but I could definitely use some input. Last I checked, this morning disk 14 was still rebuilding, and the error count for disk 13 was still skyrocketing.
March 8, 201313 yr Author Ok, now I KNOW there is an issue.... So disk 14 finished getting rebuilt, and I put in a new drive to start replacing disk 13, ... and now this is happening
March 8, 201313 yr If you now replace the second failing disk then the first rebuild will stop and not complete, the second failing drive will also not be rebuild. The parity system can only fix 1 drive at a time, double failure means loss of 2 drives. Get a SMART report on the second failing drive. Double check issues with cabling and or power. I would suggest you try and move data off the failing drive to a known ok one... Also... It is kind of a coincedence... Did you recently change stuff in the system ?
March 8, 201313 yr Author If you now replace the second failing disk then the first rebuild will stop and not complete, the second failing drive will also not be rebuild. The parity system can only fix 1 drive at a time, double failure means loss of 2 drives. Get a SMART report on the second failing drive. Double check issues with cabling and or power. I would suggest you try and move data off the failing drive to a known ok one... Also... It is kind of a coincedence... Did you recently change stuff in the system ? I didn't really change anything, but the issue seems to be that when one drive is showing signs of failing, a different drive shows the same problem when the first drive is being rebuilt/replaced. When I replaced (disk 14), disk 13 started showing signs of failure. Now that I'm replacing disk 13, disk 5 and 12 are showing major issues. These drives have been running pedal to the metal for days now, and they're showing that they're getting quite hot, which I'm wondering if it has something to do with what's going on here. I've ordered some more powerful case fans which I'll install as soon as I get them. I'm honestly not sure how to check power related issues...
March 8, 201313 yr My drives seldom get above 40C and I have my SFF case packed. What PSU are you using?
March 8, 201313 yr Author My drives seldom get above 40C and I have my SFF case packed. What PSU are you using? Corsair 850AX... Hmm...some of these drives getting above 50C is definitely out of the ordinary and I suppose could explain these funky issues. I wish I had a huge box fan I could place right in front of the case, but for now I think I'm going to have to just wait for the drive to finish rebuilding, turn off the PC and wait until the fans come in.
March 8, 201313 yr What is your ambient temperature ( the room and outside temp), open up the case, let air go freely, open up a door if it helps.. 50 degrees is way to hot..
March 8, 201313 yr Author What is your ambient temperature ( the room and outside temp), open up the case, let air go freely, open up a door if it helps.. 50 degrees is way to hot.. Ambient temp is ~70F, the server is inside a closet which has the door open, and occasionally I open the window (outside air temp 40-50F). The case has been open for quite some time due to laziness...
March 8, 201313 yr What is your ambient temperature ( the room and outside temp), open up the case, let air go freely, open up a door if it helps.. 50 degrees is way to hot.. Ambient temp is ~70F, the server is inside a closet which has the door open, and occasionally I open the window (outside air temp 40-50F). The case has been open for quite some time due to laziness... Air flow can make a big difference. My server is inside a cabinet but I have installed fans in the cabinet with their own temperature controller. Sent from my Nexus 7 using Tapatalk 2
March 8, 201313 yr Having the sides off the case can actually INCREASE the temperature of the drives. With the case closed up the air flows through the case and across the drives - cooling them (if designed correctly). You could put a box fan where the intake for you case is and get a better cooling effect but I would try to close up the case and see if your temps drop.
March 8, 201313 yr Author Having the sides off the case can actually INCREASE the temperature of the drives. With the case closed up the air flows through the case and across the drives - cooling them (if designed correctly). You could put a box fan where the intake for you case is and get a better cooling effect but I would try to close up the case and see if your temps drop. The case here is a Norco 4224, so only the top is removable. I'm not sure if removing the top of the case helps or hurts, as the fans are very close to the hard drives... regardless, a box fan might be an upcoming purchase.
Archived
This topic is now archived and is closed to further replies.