• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Project: Rackmount Overkill

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
I'm not sure if it attempts to charge the battery when there isn't one hooked up; that is a good question though. I know the battery has electronics inside to help charge it and get information on the battery, so it may attempt to "talk" to the battery before it starts charging it.
 
It follows the same premise as other rechargeables, it looks for a feedback current IIRC (long time since I talked to an EE about batteries).

Basically, it doesnt see ANY signal with no battery, with a dead but good battery it does. With a totally dead no good battery it gets a zero signal the same as a no battery indicator.

IIRC, YMMV, IANAEE
 
I'll throw the battery on my other card. I really hope that card isn't toasted.
 
Exactly what I needed tonight. On the other platter (see what I did there?), I know that the hotspare works as intended. I attempted to move back to the old drive, but it canceled half way through with the "failed" message as you see here. I'm letting it rebuild now.

failed_hitachi.png
 
So, I'm not sure what happened with the drive. I yanked it on Monday and checked the SMART data with a few programs. Each told me the drive was in perfect shape and there is no reported issues. I fired up HDTune and started a full platter scan. It finished 5 hours later with absolutely zero errors. I was a bit confused, I just put it under a fair amount of stress. So, still trying to figure out what is wring, I fired up IOMeter, added a 4kb read (100%random) test to the list and fired it up. I let it run for a minute and started the HDTune test again. This ran for over 8 hours of pure thrashing and reported no errors. I pulled the server out of the rack, reseated the cables and dropped the drive back in. I restarted the server, got into the RAID card's BIOS, added the "failed" Hitachi as a global hot spare and forced my normal hot spare offline. This forced it to rebuilt on the "failed" drive. I restarted the server while it was doing the rebuild and got the OS back up. The rebuild ran for ~6 hours and finished successfully. I'm really hoping it was just a fluke or a cable.
 
I wonder if the equivalent to WD's TLER, CCTL is configured correctly. I had issues with my WD 640 blacks until I enabled TLER preventing the drive from spending too much time on error recovery causing it to drop from my array.
 
It was my understanding they did not have this feature, but I would sure like to know. :shrug:
 
I wonder if the equivalent to WD's TLER, CCTL is configured correctly. I had issues with my WD 640 blacks until I enabled TLER preventing the drive from spending too much time on error recovery causing it to drop from my array.


Western Digital drives no longer support this feature, so I won't be buying anymore WD storage drives. Main drives like VR's are a maybe, but big storage drives definitely not.

And the issue that thideras described is spot on, I think it is a tler-like issue.

This is why I sent several caviar blacks back to newegg. I remember thideras thinking I was stupid/crazy... no offense thid but I'm sort of glad you're seeing what I saw.
 
Western Digital drives no longer support this feature, so I won't be buying anymore WD storage drives. Main drives like VR's are a maybe, but big storage drives definitely not.
That really bites. I miss being able to use a ton of "cheap" drives as storage. Just wish there was an alternative. I'd be using 5400/4300 RPM drives if it was substantially cheaper since I don't need the speed. 660mb/sec read is a little excessive for a network server when the NIC caps out at 85mb/sec. :p

2tb drives for $100 with no "features" would be absolutely amazing. Want. To. Buy.

This is why I sent several caviar blacks back to newegg. I remember thideras thinking I was stupid/crazy... no offense thid but I'm sort of glad you're seeing what I saw.
I'll have to do some research, but it was my understanding that these drives did not have that feature; the whole reason I bought them! Good thing we aren't busy today, I can check some reviews.
 
I wonder if the equivalent to WD's TLER, CCTL is configured correctly. I had issues with my WD 640 blacks until I enabled TLER preventing the drive from spending too much time on error recovery causing it to drop from my array.


That's why you should not use consumer disk in raid, the sector repair clock is set to high and it will spend up to 20 seconds trying to repair and error. Where as a WD RE or Raid Edition Sata disk will spend maybe 2-4 seconds and then mark itself failed. I see people trying to run a low end consumer disk, like the Green drives in raid on a decent hardware controller and constantly complain how the 'drive has failed' and blame the manufacture for there own ignorance. All drives will have bad sectors that need fixed at some point, and the chance of bad sectors will obviously increase with the larger capacity disk modern day.

Ever wonder why 146GB SAS is the sweet spot for arrays still?

:bday:

Good luck Corey :)
 
This thread's been going quite a while. I found it while searching the forum for ideas about how to set up my own rackmount system. I got an Equipto rack last year and have just gotten some servers to stick in it. I'm thinking Beowulf.

I like the look of your rack and your computer room.
 
Thank you, I put a lot of thought, time and effort into the systems :)

I was thinking about picking up some 1u servers and running a cluster. I've been very slowly finding uses for the systems and only have two servers offline at the moment. What are you going to use yours for and what seems to be the most common programs to run on it?
 
I got a hold of 5 - Proliant DL380's (dual Xeon 2.8) just last week. I've had the rack for a year and have been looking for something cheap to stick in it, I refuse to pay more than $35/each for any server so it's taken a while to find some. Am planning on using it for SETI or folding, something like that - just nerd stuff really. As far as set up, well, I'm just starting my research so I'll get back to you. It'll definitely be a cluster however, which may not be optimal for what I've got, but I want to build one so there it is.
 
Very nice, create a thread here so I can keep an eye on it; I'm very interested in your results.
 
Great, so I was having issues with my Western Digital MyBook connecting to find my server is completely turned off when I got home. I was working on it earlier and it just stopped responding; PuTTY didn't even report that it lost the connection, it just stopped working.

Investigating the issue and swapping some hard drives around (meant to do this a long time ago).
 
System threw a kernel panic as soon as it went to mount other file systems. Disabled the mount for my RAID array and it fired right up. Currently running a fsck; already hit 50 or so inodes that were corrupted. Hopefully shouldn't lose anything...
 
You have way more problems than me, dude. I have no idea how you run into so many issues!I setup my server and it just sits there and gives me any file I want, not to mention it's a great foot stool!
 
Weeeeelp. Found the issue.

Started the fsck.......BAM, breaker flips off and on. Server turns off. I think my UPS bit the dust and was tripping the breaker temporarily. Removed it from the equation and am trying again.
 
Weeeeelp. Found the issue.

Started the fsck.......BAM, breaker flips off and on. Server turns off. I think my UPS bit the dust and was tripping the breaker temporarily. Removed it from the equation and am trying again.

That'd explain it! But are your machines set to turn back on?
 
Back