Recommended NAS?

[KJake]

Originally Posted by mikesm

At least thats working out ok. What happened with the 1tb drives? Were the seals tampered with? Whoever did that swapping around is a real tool. Look at this page of the fatwallet thread: http://www.fatwallet.com/forums/mess...2358&start=140

I'm lacking motivation to leave the warm house since I can't really hook them up any way. I have to imagine that going to stores on the day after xmas isn't fun. maybe tomorrow, i need to visit the post office anyway.

[KJake]

Originally Posted by KJake

I'm lacking motivation to leave the warm house since I can't really hook them up any way. I have to imagine that going to stores on the day after xmas isn't fun. maybe tomorrow, i need to visit the post office anyway.

Got the two replaced with two more and made the Geek Squad (oh noes!) test them before leaving the store. Now I have 3 working ones. I'll have to get at least one more PMP, figure out why my motherboard hates multiple Sil chips and whether there is any type of warranty for my other PMP if it is really dead.

Then I can migrate out the 500gb drives, and talk to seagate since NewEgg only had a 30 day warranty on them (5 year manu however).

[mikesm]

Originally Posted by KJake

Got the two replaced with two more and made the Geek Squad (oh noes!) test them before leaving the store. Now I have 3 working ones. I'll have to get at least one more PMP, figure out why my motherboard hates multiple Sil chips and whether there is any type of warranty for my other PMP if it is really dead.

Then I can migrate out the 500gb drives, and talk to seagate since NewEgg only had a 30 day warranty on them (5 year manu however).

Good. Did the geek squad guys hear about the disk swapping fraud?

On the PMP, can you try swapping it with the working PMP, or do you just have one? The addonics people are very nice to work with - if it's dead, I'm sure they will swap it for you. Also make sure you have the floppy power connector connected, etc... Do any of the indicator lights light up on the existing PMP?

[KJake]

Originally Posted by mikesm

Good. Did the geek squad guys hear about the disk swapping fraud?

Nope.

Originally Posted by mikesm

On the PMP, can you try swapping it with the working PMP, or do you just have one? The addonics people are very nice to work with - if it's dead, I'm sure they will swap it for you. Also make sure you have the floppy power connector connected, etc... Do any of the indicator lights light up on the existing PMP?

I do have two right now, but things have been recording since and I just haven't had the time to play yet. I will try swapping, etc to determine if it is dead and then talk to addonics. Mine don't have any indicator lights - at least I don't think they do...

[Richard.Wilkes@]

Hello!

Finally made it through all 12 or 13 pages of this particular thread. Lots of good stuff... however, the one thing I was hoping to get was a clear, concise set of recommendations for the components one would buy for a brand-new Linux system (probably Ubuntu, although I'm not attached to any particular distribution) to do file serving for my Sage setup. For reference, my current SageTV server records up to 4 SD streams and 2 HD streams at once. I'll eventually move to all HD recording and the use of 2-4 HD extenders for viewing.

I plan on using RAID-5. I'd like to end up with at least 2 TB of usable storage, but more would be better (HD sucks down disk space!). Ideally, someone (Mike seems to be particularly knowledgeable) would post a list of specific components here of what they think would be a great system, hopefully based on experience.

Thanks in advance for any suggestions you folks can provide.

Rich

[mikesm]

Originally Posted by Richard.Wilkes@

Hello!

Finally made it through all 12 or 13 pages of this particular thread. Lots of good stuff... however, the one thing I was hoping to get was a clear, concise set of recommendations for the components one would buy for a brand-new Linux system (probably Ubuntu, although I'm not attached to any particular distribution) to do file serving for my Sage setup. For reference, my current SageTV server records up to 4 SD streams and 2 HD streams at once. I'll eventually move to all HD recording and the use of 2-4 HD extenders for viewing.

I plan on using RAID-5. I'd like to end up with at least 2 TB of usable storage, but more would be better (HD sucks down disk space!). Ideally, someone (Mike seems to be particularly knowledgeable) would post a list of specific components here of what they think would be a great system, hopefully based on experience.

Thanks in advance for any suggestions you folks can provide.

Rich

Rich, my current setup is with an older socket 939 X2 3800 CPU and a NF4 based motherboard that does not have AHCI capable SATA ports, and given how expensive DDR memory is, I wouldn't recommend building that.


First off, note the hardware notes on the SATA PMP driver page: http://home-tj.org/wiki/index.php/Libata-tj-stablep://

Given this, I would probably recommend a good p35 based intel motherboard that has an ich9r on it, like the abit ip35 (not the -e version or the -v version), or the MSI P35 Combo-F http://www.clubit.com/product_detail...emno=CA4854032(which is on sale now for$69 ar at clubit, which is also running a combo sale with cpus). The ip35 has 6 ich9r ports, two of them have issues as per the libata-tj page for pmp use, but they work fine for opitical drives and such I think. The MSI has 5 sata ports, and an additional 1x pci-e slot compared to the ip35 (which has 2). Both use the realtek 8111b pcie ethernet controller which shouldn't bottleneck due to PCI issues.

Both of these boards can take pci-e based video cards, but I'd probably just use a cheap pci based vga card from a surplus store like this one: http://www.outletpc.com/c4338.html. This allows you to insert a x1 pcie-e disk controller in the x16 pci-e slot, since you'll probably not be caring about graphics performance in this system. But either have enough x1 PCI-E slots to be able to support plenty of disk if you exceed the onboard AHCI ports when using pmp's.

If you wait a bit, you may be able to get g35 based motherboards with integrated graphics, but if you are like me and have some old pcie vid cards around, these boards aren't that much more expensive, have more expansion and can overclock well too if you want to go there.

As far as cpu, an e2180 should be plenty, but I'd probably get a e4500 to have some headroom. If you plan on doing commercial detect as well, go with the 4500 or a faster cpu. It's not that expensive.

For ram 2x1gb is great, but you could go to 4GB in 2x2gb for not that much more, and have even more buffering.

In terms of the 5in3 racks, the supermicro CSE-M35-1B is very robust, but it's a little larger than the addonics or ams 5in3's, so I tend to reccommend the latter, and I think they look better too. The AMS is available at fry's for $95, which is a very good price.

For pmp's, you really want to go with a 3736 based pmp like this one: http://www.addonics.com/products/hos.../ad5sapm-e.asp.
The "rackmount" version is easy to attach to the case with double sticky tape. You'll an esata to sata cable to connect it to a motherboard sata port, the 5in3's all come with sata cables that go from the pmp to each drive port.

SI3132 based pci-e controllers also work very well in my experience. I personally got this cheap siig one: http://www.siig.com/ViewProduct.aspx?pn=SC-SAE012-S2, and along with a 2nd one, I can drive 4 pmp's with their sata ports. But pretty much any si3132 based pci-e controller will work fine.

That's about it. I have mentioned my love of the original coolermaster stacker case http://www.coolermaster.com/products...ate=168&id=166, but the newer cm 810 (not the 830) seems just as good.

Use these components and you'll end up with a superfast NAS. But a lot of other variations are possible too, as pretty much any p35 based mb will work. Those with an actual intel GbE controller like the 82566DC may be a bit faster on the network, but I don't think it's a material difference.

Does that help?

thx
mike

[Richard.Wilkes@]

Thank you, Mike!

[KJake]

My thanks also. I may need to replace my motherboard. I should have known better, but it is my own fault for not doing research first (in my defence I repurposed the current one). I knew back when nForce2 board were out that they were crap, so I figured nForce4 based ones would be OK now...not so. At this point, another $100 to keep this running happily is a drop in the bucket, lol. So, I'll probably go for the ICH9 based boards. The PMPs should work connected to those ports. My other thought is to let this system limp along until SATAIII comes out and upgrade then.

Oh, and a revelation about fully removing the 500GB drives to try and return them to Seagate... XFS can not be shrunk, only grown. So, I'm pretty sure that I'm going to do an advanced replacement (after I first convince them that they need to replace them) and then swap each drive one-by-one and wait for the array to rebuild after each swap. That _should_ work.

[mikesm]

Originally Posted by KJake

My thanks also. I may need to replace my motherboard. I should have known better, but it is my own fault for not doing research first (in my defence I repurposed the current one). I knew back when nForce2 board were out that they were crap, so I figured nForce4 based ones would be OK now...not so. At this point, another $100 to keep this running happily is a drop in the bucket, lol. So, I'll probably go for the ICH9 based boards. The PMPs should work connected to those ports. My other thought is to let this system limp along until SATAIII comes out and upgrade then.

Oh, and a revelation about fully removing the 500GB drives to try and return them to Seagate... XFS can not be shrunk, only grown. So, I'm pretty sure that I'm going to do an advanced replacement (after I first convince them that they need to replace them) and then swap each drive one-by-one and wait for the array to rebuild after each swap. That _should_ work.

Kjake, the Nf4 will work just fine. Mine does. You just have to use a 3132 PCIE card, and it will perform just fine. I just answered Rick's question aboyt what I would build today, but in reality, a new system is hardly needed for fast performance.

Eventually, I'll probably rebuild mine, but right now it just chugs along... Nice and fast... :-)

What happened to those 1 TB disks?

Also, if you are willing to take the array offline, you can just clone the old disks to the new disks, insert the new disks in the rack and then bring the array back online. No need to do all the rebuilds if you are willing to take a little downtime.



thx
Mike

[KJake]

Originally Posted by mikesm

Kjake, the Nf4 will work just fine. Mine does. You just have to use a 3132 PCIE card, and it will perform just fine. I just answered Rick's question aboyt what I would build today, but in reality, a new system is hardly needed for fast performance.

Yup, I've got two of those Syba cards, but the system doesn't POST right when there is something connected to the second card. I assume this has something to do with the motherboard being a POS nForce4 board. Maybe I need to disable something else? I already disabled the 3132 port that is on it, but I'm using the NVRAID ports for my OS disk (RAID disabled).

Originally Posted by mikesm

What happened to those 1 TB disks?

I haven't had the chance to test that PMP out yet. Without it, I can't possibly install any more drives. Probably can test tomorrow or Monday.

Originally Posted by mikesm

Also, if you are willing to take the array offline, you can just clone the old disks to the new disks, insert the new disks in the rack and then bring the array back online. No need to do all the rebuilds if you are willing to take a little downtime.

This is true, hadn't thought of that.

[KJake]

Originally Posted by mikesm

This will add 3 TB of needed space to my NAS server, and allow me to retire my 5x320GB original raid array and let me use these for system disks on other systems.

By the way, how did you end up retiring the 5x320GB set? As I mentioned in a previous post, you can't shrink XFS...so how exactly did you do this?

[KJake]

Originally Posted by KJake

Probably can test tomorrow or Monday.

Yay, it is still working - no idea what happened, but it is working for now. This happened last time too, so if it freaks out again...

Since I was poking around, I decided to mount the 5-in-3, and wire everything up. I moved the 5x500gb disks back to the PMP and then plugged all the ports on the 5-in-3 to any left over ports (NV and 3132). I put the 3x1TB in there (2 spots to grow, plus two more spots open inside the case).

If there is a proper way to migrate with EVMS/XFS, then I'd like to plan on swapping a drive out for a 1TB drive. Each new drive can be added to the 3x1TB MD array and I would be able to grow the array by 1TB (RAW, then minus what I replaced). This seems like a more practical arrangement, but something I would only do if I'm running out of space.

488384001 sda1
488384001 sdg1
488384001 sdh1
488384001 sdi1
488384001 sdj1
488384001 sdk1
732572001 sdb1
732572001 sdc1
732572001 sdd1
732572001 sde1
976760001 sdl1
976760001 sdm1
976760001 sdn1

[KJake]

It freaked out already. One of the drives randomly disappeared and then was re-added to the array as a spare/failed. I removed it and added it and it figured things out and is rebuilding (I didn't have to fiddle with anything hardware-wise). Something tells me that all is not well with this PMP.

I'm seeing this in the logs now too:

Jan  7 01:03:28 orion kernel: [ 4971.428639] ata1.00: limiting SATA link speed to 1.5 Gbps
Jan  7 01:03:29 orion kernel: [ 4972.212487] ata1.15: exception Emask 0x100 SAct 0x0 SErr 0x200000 action 0x6 frozen
Jan  7 01:03:29 orion kernel: [ 4972.212526] ata1.15: irq_stat 0x00020002 <PMP DMA CS errata>
Jan  7 01:03:29 orion kernel: [ 4972.212557] ata1.00: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen
Jan  7 01:03:29 orion kernel: [ 4972.212592] ata1.00: cmd 25/00:08:4f:34:22/00:03:00:00:00/e0 tag 3 cdb 0x0 data 397312 in
Jan  7 01:03:29 orion kernel: [ 4972.212593]          res 99/37:00:00:00:00/00:00:40:28:99/00 Emask 0x2 (HSM violation)
Jan  7 01:03:29 orion kernel: [ 4972.212628] ata1.01: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen
Jan  7 01:03:29 orion kernel: [ 4972.212663] ata1.01: cmd 25/00:00:47:33:22/00:04:00:00:00/e0 tag 2 cdb 0x0 data 524288 in
Jan  7 01:03:29 orion kernel: [ 4972.212664]          res 99/37:00:00:00:00/00:00:40:28:99/00 Emask 0x2 (HSM violation)
Jan  7 01:03:29 orion kernel: [ 4972.212699] ata1.02: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen
Jan  7 01:03:29 orion kernel: [ 4972.212733] ata1.02: irq_stat 0x00020002 <ran out of SGEs while writing>
Jan  7 01:03:29 orion kernel: [ 4972.212765] ata1.02: cmd 35/00:e8:57:2f:22/00:03:00:00:00/e0 tag 4 cdb 0x0 data 512000 out
Jan  7 01:03:29 orion kernel: [ 4972.212766]          res 99/37:00:00:00:00/00:00:40:28:99/00 Emask 0x2 (HSM violation)
Jan  7 01:03:29 orion kernel: [ 4972.212801] ata1.03: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen
Jan  7 01:03:29 orion kernel: [ 4972.212836] ata1.03: cmd 25/00:18:3f:34:22/00:03:00:00:00/e0 tag 0 cdb 0x0 data 405504 in
Jan  7 01:03:29 orion kernel: [ 4972.212837]          res 99/37:00:00:00:00/00:00:40:28:99/00 Emask 0x2 (HSM violation)
Jan  7 01:03:29 orion kernel: [ 4972.212872] ata1.04: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen
Jan  7 01:03:29 orion kernel: [ 4972.212907] ata1.04: cmd 25/00:00:47:33:22/00:04:00:00:00/e0 tag 1 cdb 0x0 data 524288 in
Jan  7 01:03:29 orion kernel: [ 4972.212908]          res 99/37:00:00:00:00/00:00:40:28:99/00 Emask 0x2 (HSM violation)
Jan  7 01:03:29 orion kernel: [ 4972.212942] ata1.05: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen
Jan  7 01:03:29 orion kernel: [ 4972.212979] ata1.15: hard resetting link
Jan  7 01:03:29 orion kernel: [ 4972.212981] ata1: controller in dubious state, performing PORT_RST
Jan  7 01:03:31 orion kernel: [ 4974.444066] ata1.15: SATA link up 3.0 Gbps (SStatus 123 SControl 0)
Jan  7 01:03:31 orion kernel: [ 4974.604026] ata1.00: hard resetting link
Jan  7 01:03:32 orion kernel: [ 4975.327886] ata1.00: SATA link up 1.5 Gbps (SStatus 113 SControl 310)
Jan  7 01:03:32 orion kernel: [ 4975.343882] ata1.01: hard resetting link
Jan  7 01:03:33 orion kernel: [ 4976.067741] ata1.01: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
Jan  7 01:03:33 orion kernel: [ 4976.083737] ata1.02: hard resetting link
Jan  7 01:03:33 orion kernel: [ 4976.791599] ata1.02: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
Jan  7 01:03:33 orion kernel: [ 4976.807597] ata1.03: hard resetting link
Jan  7 01:03:34 orion kernel: [ 4977.531455] ata1.03: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
Jan  7 01:03:34 orion kernel: [ 4977.547451] ata1.04: hard resetting link
Jan  7 01:03:35 orion kernel: [ 4978.271310] ata1.04: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
Jan  7 01:03:35 orion kernel: [ 4978.271314] ata1.05: hard resetting link
Jan  7 01:03:35 orion kernel: [ 4978.835200] ata1.05: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
Jan  7 01:03:36 orion kernel: [ 4978.973688] ata1.00: configured for UDMA/100
Jan  7 01:03:36 orion kernel: [ 4979.099086] ata1.01: configured for UDMA/33
Jan  7 01:03:36 orion kernel: [ 4979.224352] ata1.02: configured for UDMA/33
Jan  7 01:03:36 orion kernel: [ 4979.340861] ata1.03: configured for UDMA/66
Jan  7 01:03:36 orion kernel: [ 4979.465328] ata1.04: configured for UDMA/66
Jan  7 01:03:36 orion kernel: [ 4979.483086] ata1: EH complete

[mikesm]

Originally Posted by KJake

By the way, how did you end up retiring the 5x320GB set? As I mentioned in a previous post, you can't shrink XFS...so how exactly did you do this?

This is what LVM is for. If you add a PV that is the same size or greater as the PV you want to remove, you can then do a pvmove and LVM will move off the stuff in the old array to the new one. This operates under the filesystem, so you don't have to resize XFS downwards.

Now, if you want to remove an array because the XFS filesystem is say 50% full, and you want to take 30% of it away, that you cannot do.

thx
mike

[KJake]

Originally Posted by mikesm

This is what LVM is for. If you add a PV that is the same size or greater as the PV you want to remove, you can then do a pvmove and LVM will move off the stuff in the old array to the new one. This operates under the filesystem, so you don't have to resize XFS downwards.

Now, if you want to remove an array because the XFS filesystem is say 50% full, and you want to take 30% of it away, that you cannot do.

thx
mike

OK, I sort of had a feeling that something like that could be done. Unfortunately, my 6x500GB MD size is 2.3TB and I'm not going to get a full 2TB from the new Hitachi disks, so that won't work, right?

[KJake]

Originally Posted by KJake

It freaked out already. One of the drives randomly disappeared and then was re-added to the array as a spare/failed. I removed it and added it and it figured things out and is rebuilding (I didn't have to fiddle with anything hardware-wise). Something tells me that all is not well with this PMP.

I'm seeing this in the logs now too:

Jan  7 01:03:28 orion kernel: [ 4971.428639] ata1.00: limiting SATA link speed to 1.5 Gbps
Jan  7 01:03:29 orion kernel: [ 4972.212487] ata1.15: exception Emask 0x100 SAct 0x0 SErr 0x200000 action 0x6 frozen
Jan  7 01:03:29 orion kernel: [ 4972.212526] ata1.15: irq_stat 0x00020002 <PMP DMA CS errata>
Jan  7 01:03:29 orion kernel: [ 4972.212557] ata1.00: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen
Jan  7 01:03:29 orion kernel: [ 4972.212592] ata1.00: cmd 25/00:08:4f:34:22/00:03:00:00:00/e0 tag 3 cdb 0x0 data 397312 in
Jan  7 01:03:29 orion kernel: [ 4972.212593]          res 99/37:00:00:00:00/00:00:40:28:99/00 Emask 0x2 (HSM violation)
Jan  7 01:03:29 orion kernel: [ 4972.212628] ata1.01: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen
Jan  7 01:03:29 orion kernel: [ 4972.212663] ata1.01: cmd 25/00:00:47:33:22/00:04:00:00:00/e0 tag 2 cdb 0x0 data 524288 in
Jan  7 01:03:29 orion kernel: [ 4972.212664]          res 99/37:00:00:00:00/00:00:40:28:99/00 Emask 0x2 (HSM violation)
Jan  7 01:03:29 orion kernel: [ 4972.212699] ata1.02: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen
Jan  7 01:03:29 orion kernel: [ 4972.212733] ata1.02: irq_stat 0x00020002 <ran out of SGEs while writing>
Jan  7 01:03:29 orion kernel: [ 4972.212765] ata1.02: cmd 35/00:e8:57:2f:22/00:03:00:00:00/e0 tag 4 cdb 0x0 data 512000 out
Jan  7 01:03:29 orion kernel: [ 4972.212766]          res 99/37:00:00:00:00/00:00:40:28:99/00 Emask 0x2 (HSM violation)
Jan  7 01:03:29 orion kernel: [ 4972.212801] ata1.03: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen
Jan  7 01:03:29 orion kernel: [ 4972.212836] ata1.03: cmd 25/00:18:3f:34:22/00:03:00:00:00/e0 tag 0 cdb 0x0 data 405504 in
Jan  7 01:03:29 orion kernel: [ 4972.212837]          res 99/37:00:00:00:00/00:00:40:28:99/00 Emask 0x2 (HSM violation)
Jan  7 01:03:29 orion kernel: [ 4972.212872] ata1.04: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen
Jan  7 01:03:29 orion kernel: [ 4972.212907] ata1.04: cmd 25/00:00:47:33:22/00:04:00:00:00/e0 tag 1 cdb 0x0 data 524288 in
Jan  7 01:03:29 orion kernel: [ 4972.212908]          res 99/37:00:00:00:00/00:00:40:28:99/00 Emask 0x2 (HSM violation)
Jan  7 01:03:29 orion kernel: [ 4972.212942] ata1.05: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen
Jan  7 01:03:29 orion kernel: [ 4972.212979] ata1.15: hard resetting link
Jan  7 01:03:29 orion kernel: [ 4972.212981] ata1: controller in dubious state, performing PORT_RST
Jan  7 01:03:31 orion kernel: [ 4974.444066] ata1.15: SATA link up 3.0 Gbps (SStatus 123 SControl 0)
Jan  7 01:03:31 orion kernel: [ 4974.604026] ata1.00: hard resetting link
Jan  7 01:03:32 orion kernel: [ 4975.327886] ata1.00: SATA link up 1.5 Gbps (SStatus 113 SControl 310)
Jan  7 01:03:32 orion kernel: [ 4975.343882] ata1.01: hard resetting link
Jan  7 01:03:33 orion kernel: [ 4976.067741] ata1.01: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
Jan  7 01:03:33 orion kernel: [ 4976.083737] ata1.02: hard resetting link
Jan  7 01:03:33 orion kernel: [ 4976.791599] ata1.02: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
Jan  7 01:03:33 orion kernel: [ 4976.807597] ata1.03: hard resetting link
Jan  7 01:03:34 orion kernel: [ 4977.531455] ata1.03: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
Jan  7 01:03:34 orion kernel: [ 4977.547451] ata1.04: hard resetting link
Jan  7 01:03:35 orion kernel: [ 4978.271310] ata1.04: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
Jan  7 01:03:35 orion kernel: [ 4978.271314] ata1.05: hard resetting link
Jan  7 01:03:35 orion kernel: [ 4978.835200] ata1.05: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
Jan  7 01:03:36 orion kernel: [ 4978.973688] ata1.00: configured for UDMA/100
Jan  7 01:03:36 orion kernel: [ 4979.099086] ata1.01: configured for UDMA/33
Jan  7 01:03:36 orion kernel: [ 4979.224352] ata1.02: configured for UDMA/33
Jan  7 01:03:36 orion kernel: [ 4979.340861] ata1.03: configured for UDMA/66
Jan  7 01:03:36 orion kernel: [ 4979.465328] ata1.04: configured for UDMA/66
Jan  7 01:03:36 orion kernel: [ 4979.483086] ata1: EH complete

It completely crashed again. I'm calling it - it's faulty.

[mikesm]

Originally Posted by KJake

OK, I sort of had a feeling that something like that could be done. Unfortunately, my 6x500GB MD size is 2.3TB and I'm not going to get a full 2TB from the new Hitachi disks, so that won't work, right?

Obviously, this requires another 1 TB drive purchase!

[mikesm]

Originally Posted by KJake

It completely crashed again. I'm calling it - it's faulty.

You should check the SATA cabling as well as the power cabling to the 5in3 rack. Not supplying enough power to the rack can yield bad behavior.

I would be surprised if the PMP failed - they are pretty simple devices. Make sure the cabling is solid before getting a RMA.

thx
mike

[KJake]

Originally Posted by mikesm

Obviously, this requires another 1 TB drive purchase!

haha, I was afraid of that - those are $359 now or something and would ruin my savings. Looks like there are Western Digital drives out there for about $250, not terrible, though I've vowed to never buy them again since WD have been the worst drives I've ever bought.

Seen any other deals out there?

[KJake]

Originally Posted by mikesm

You should check the SATA cabling as well as the power cabling to the 5in3 rack. Not supplying enough power to the rack can yield bad behavior.

I would be surprised if the PMP failed - they are pretty simple devices. Make sure the cabling is solid before getting a RMA.

thx
mike

The cables the drives are using is fine, as they're all using the same cables and work fine for over a week not connected to the PMP. Connected them to the PMP and boom.

I should check the cable from the PMP to the controller, I don't think that I did that. I think that's about the only possibility left here.

The drives (500GB) that are connected to the PMP aren't in the 5-in-3, but I used 2 SATA power plugs to power it, direct from the PSU, not split or anything (since the 5-in-3 is going to "split" the power anyways).