Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
Well eff newegg, they processed one of my fubared orders (ordered 2 bundles of qty 3 for $299.99, one received one drive in each box so 2 out of 6).
They RP (replaced) two of them (should be here tomorrow) waited a few days until I harassed them again on status of second replacement (qty 2) then promptly RF (refunded) the other two and now are saying they are out of stock and they cannot price match the $100 per drive deal now...fark. NowIi am left w/ inconsistent devices or going to hunt them down elsewhere.
I have THE WORST luck w/ newegg. They are an atrocity to deal with!!!
Anyone wanna send two my way for price match? :-(
They RP (replaced) two of them (should be here tomorrow) waited a few days until I harassed them again on status of second replacement (qty 2) then promptly RF (refunded) the other two and now are saying they are out of stock and they cannot price match the $100 per drive deal now...fark. NowIi am left w/ inconsistent devices or going to hunt them down elsewhere.
I have THE WORST luck w/ newegg. They are an atrocity to deal with!!!
Anyone wanna send two my way for price match? :-(
Last edited:
I ordered 2 with the last deal, went back later in the day and ordered 3 more and they canceled the order of 3 without any email or anything. Newegg is a pain in the butt. Probably spend the few extra $$ and go through amazon for the rest...
Looking at jet.com, I can order 3 drives, with all the discounts comes to around $320 to my door. I might just go that route.
Last edited:
Wow, that's sad to hear - I used to buy tons of stuff from the egg a few years ago, and they were great other than packing bulk drives in one layer of bubble wrap (if you bought 3 or more at once they would ship them in part of a 20-drive shipper though.) I liked that the egg was fighting the BS online shopping cart patent, and functionally I really like that you can filter categories well and actually narrow categories down to actually comparison shop and find what you are looking for on the newegg site, unlike the deliberately poorly ordered search results with bait&switch action and constant price fluctuations at scamazon. Now I'm limited to occasional shellshockers, but it sounds like this 3-drive deal went wrong in the shipping department for everyone. Last time I ordered a bunch of drives I got them from B&H since they had them in stock and shipped free and pack well, but that was over 6 months ago.
It looks like they are still $123.99 at B&H right now:
Seagate 4TB Backup Plus Portable Hard Drive STDR4000100 B&H
It looks like they are still $123.99 at B&H right now:
Seagate 4TB Backup Plus Portable Hard Drive STDR4000100 B&H
I ended up getting last two Seagate Expansion 4tb disks (STEA4000400) for $117 each off Amazon yesterday, will be here tomorrow.
DataZilla zpool so far (raidz for now, raidz2 tomorrow), should be about 13.5TB usable across 6 disks w/ two hussl4020 disks as cache devices in my SC216 chassis, it's no slouch for 5400 RPM disks w/ those slc ssd's helping out. Can saturate 1Gbps network link, can't push it much more than that so far but I'll be happy w/ that. This is my new dump/replication/CYA target pool. Probably gonna move my media off my pair of 4tb white labels from goharddrive I got a year or so ago and retire those.
Looking at my stack now and wondering (w/ the exodus from 3.5" spinners to 2.5" spinners here and ssd's of course in the mix) why I will need my Norco 4224 since it is down to just 4 disks (2 vSAN and 2 ZFS) the ZFS disks are my 2 4tb 3.5" spinner media pool, it's going bye-bye as mentioned above. Feels silly to have a 24bay chassis w/ only a small handful of disks in it...you all know what I am debating hahaha...she's CLEAN...no comment, need to think on it for a bit lol
DataZilla zpool so far (raidz for now, raidz2 tomorrow), should be about 13.5TB usable across 6 disks w/ two hussl4020 disks as cache devices in my SC216 chassis, it's no slouch for 5400 RPM disks w/ those slc ssd's helping out. Can saturate 1Gbps network link, can't push it much more than that so far but I'll be happy w/ that. This is my new dump/replication/CYA target pool. Probably gonna move my media off my pair of 4tb white labels from goharddrive I got a year or so ago and retire those.
Looking at my stack now and wondering (w/ the exodus from 3.5" spinners to 2.5" spinners here and ssd's of course in the mix) why I will need my Norco 4224 since it is down to just 4 disks (2 vSAN and 2 ZFS) the ZFS disks are my 2 4tb 3.5" spinner media pool, it's going bye-bye as mentioned above. Feels silly to have a 24bay chassis w/ only a small handful of disks in it...you all know what I am debating hahaha...she's CLEAN...no comment, need to think on it for a bit lol
Last edited:
@whitey I love the fact that you have super SAS SSDs and low end shucked hard drives 
Funny thing is I am getting pretty decent performance outta that pool w/ rough fio stats and VM boot times are like 6-7 seconds :-D
Code:
root@fio:~# fio --randrepeat=1 --ioengine=libaio --direct=1 --name=testingtons --filename=5GBtestfile --bssplit=512/10:4k/60:8k/20:64k/10 --iodepth=64 --size=5G --readwrite=randrw --rwmixread=75
testingtons: (g=0): rw=randrw, bs=512-64K/512-64K/512-64K, ioengine=libaio, iodepth=64
fio-2.2.10
Starting 1 process
testingtons: Laying out IO file(s) (1 file(s) / 5120MB)
Jobs: 1 (f=1): [m(1)] [100.0% done] [42716KB/14243KB/0KB /s] [10.6K/3597/0 iops] [eta 00m:00s]
testingtons: (groupid=0, jobs=1): err= 0: pid=20860: Tue Jun 28 15:03:22 2016
read : io=3840.5MB, bw=53413KB/s, iops=9310, runt= 73626msec
slat (usec): min=2, max=102886, avg= 6.77, stdev=124.76
clat (usec): min=173, max=1145.5K, avg=4726.87, stdev=36592.98
lat (usec): min=178, max=1145.5K, avg=4733.83, stdev=36593.16
clat percentiles (usec):
| 1.00th=[ 812], 5.00th=[ 1224], 10.00th=[ 1512], 20.00th=[ 1912],
| 30.00th=[ 2224], 40.00th=[ 2512], 50.00th=[ 2800], 60.00th=[ 3120],
| 70.00th=[ 3536], 80.00th=[ 4080], 90.00th=[ 5024], 95.00th=[ 5984],
| 99.00th=[ 9664], 99.50th=[15424], 99.90th=[839680], 99.95th=[978944],
| 99.99th=[1089536]
bw (KB /s): min= 252, max=141683, per=100.00%, avg=63664.65, stdev=33437.91
write: io=1279.6MB, bw=17796KB/s, iops=3104, runt= 73626msec
slat (usec): min=2, max=8013, avg= 8.11, stdev=26.96
clat (usec): min=445, max=1150.2K, avg=6403.53, stdev=44166.13
lat (usec): min=584, max=1150.2K, avg=6411.84, stdev=44166.12
clat percentiles (usec):
| 1.00th=[ 1496], 5.00th=[ 1976], 10.00th=[ 2288], 20.00th=[ 2704],
| 30.00th=[ 3056], 40.00th=[ 3344], 50.00th=[ 3664], 60.00th=[ 4016],
| 70.00th=[ 4448], 80.00th=[ 5024], 90.00th=[ 6112], 95.00th=[ 7520],
| 99.00th=[12608], 99.50th=[28800], 99.90th=[929792], 99.95th=[1011712],
| 99.99th=[1089536]
bw (KB /s): min= 13, max=46623, per=100.00%, avg=20988.89, stdev=11169.21
lat (usec) : 250=0.01%, 500=0.05%, 750=0.46%, 1000=1.33%
lat (msec) : 2=16.51%, 4=55.66%, 10=24.85%, 20=0.65%, 50=0.25%
lat (msec) : 100=0.01%, 250=0.01%, 500=0.02%, 750=0.05%, 1000=0.10%
lat (msec) : 2000=0.04%
cpu : usr=4.12%, sys=12.15%, ctx=215828, majf=0, minf=11
IO depths : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=0.1%, 32=0.1%, >=64=100.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.1%, >=64=0.0%
issued : total=r=685481/w=228538/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0
latency : target=0, window=0, percentile=100.00%, depth=64
Run status group 0 (all jobs):
READ: io=3840.5MB, aggrb=53413KB/s, minb=53413KB/s, maxb=53413KB/s, mint=73626msec, maxt=73626msec
WRITE: io=1279.6MB, aggrb=17796KB/s, minb=17796KB/s, maxb=17796KB/s, mint=73626msec, maxt=73626msec
Disk stats (read/write):
dm-0: ios=684147/228175, merge=0/0, ticks=3208292/1452680, in_queue=4661368, util=99.94%, aggrios=679894/227833, aggrmerge=5591/748, aggrticks=3195640/1449700, aggrin_queue=4645596, aggrutil=99.90%
sda: ios=679894/227833, merge=5591/748, ticks=3195640/1449700, in_queue=4645596, util=99.90%
root@fio:~#
A decent sale on the 8tb mybook enclosures could bring 3.5's back as competitive in the near future and make those enclosures useful again. But at 2.5x the price for the same capacity they don't make sense today. I'm curious why it's pricing is so stable while these 4tb drives seem to be all over the place(not to complain 100usd even makes them so tempting my wallet cries)
Can you do any tests without the SLC cache just to see what native performance looks like?
The big draw of these drives is the 2.5" form factor. I have this spreadsheet to work out the price/value of the hdd's HD Space
Right now 5TB drives seem to be about the best TB/$ but I can fit 16 of the 2.5 in the space of 8 3.5
Based on how SLOW of a trickle a sVMotion is running to get it back on that pool now that I have removed cache devices I am assuming these are gonna be d|ck in the d|rt SLOW w/out read/write cache devices, will report back in a bit after this piggy finishes up.
Well I wasn't expecting fast, I thought they might have expected saturating gige for sequential with an array like yours though, looking forward to the numbers.
yeah that's why i pointed at the 8tb/250 wd mybook, if it went below 200 it would be cheaper/TB than the 4tb 2.5's with an advantage of less cabling and ports required(but worse redundancy in smaller arrays due to less drives for parity) agreed on the space advantage though, I can fit these places i can't put a 3.5 ever(I actually have some laptops with 15mm bays still!)
Yea - these are not drives for speed. Slow spindle rates and 2.5" is a killer for speed.
I look at them more like cold data storage.
BTW I have 7 of them being tested right now. I need to snap a picture as you guys would enjoy this.
I look at them more like cold data storage.
BTW I have 7 of them being tested right now. I need to snap a picture as you guys would enjoy this.
OK, so yeah this is kinda pathetic, 2-2.5MB write speeds via sVMotions to either raidz or single disk (tested all of em). The cache devices are a MUST in my book w/ these even for bulk storage. I will do some file copies as high level tests in a bit to see if they can push more than 2-3MB/sec.
Least I KNOW w/ slc sas ssd devices these things will ROCK but alone I have lil' faith in them.
EDIT: Odd, CIFS copy went to a single disk @ 65-75MBps, sVMotions must be pretty damned intense and need those caching devs.
Least I KNOW w/ slc sas ssd devices these things will ROCK but alone I have lil' faith in them.
EDIT: Odd, CIFS copy went to a single disk @ 65-75MBps, sVMotions must be pretty damned intense and need those caching devs.
Last edited:
How active are the VMs you're performing Storage vMotion on? At least initially, Storage vMotion should be mainly sequential and, assuming no file fragmentation, roll along pretty fast.