The T620 plus is a nice little box but until you guys stop jacking up the prices I'm going to wait.
$70 per unit looks just fine to me, and these comes with the AT27Mini Fiber NIC, so the carrier for (potentially) mounting a SATA SSD is there - like I've said before, those machines will work on any HP branded laptop/docking station power bricks as long as the wattage looks correct, so that can simply be a matter of scrounging an IT admin's closet for spares. The only thing I would make sure of would be to ask specifically for a Jamestown Rev. A board with the populated mSATA headers so you can conceivably slap an adapter to work with the SATA SSD.The T620 plus is a nice little box but until you guys stop jacking up the prices I'm going to wait.
Here you go.Does anyone have a link to the recycling breakdown guide for the t730? Mine arrived today and a motherboard screw is stuck somewhere inside underneath the mainboard. I heard it rattling around while unpacking the unit but it's stuck fast now.
Does the T730 have the Jamestown Rev. A or B board? I am looking for a replacement for my WatchGuard XTM515 appliance running pfSense.$70 per unit looks just fine to me, and these comes with the AT27Mini Fiber NIC, so the carrier for (potentially) mounting a SATA SSD is there - like I've said before, those machines will work on any HP branded laptop/docking station power bricks as long as the wattage looks correct, so that can simply be a matter of scrounging an IT admin's closet for spares. The only thing I would make sure of would be to ask specifically for a Jamestown Rev. A board with the populated mSATA headers so you can conceivably slap an adapter to work with the SATA SSD.
Also, isn't this the thread for the t730 replacement?
I'll take some readings after I'm finished testing ram. I've got a t730 sitting on my desk with a CX2 VPI dual 40Gb Infiniband card in right now, was planning on running 40Gb IB between two (dual port CX3s may be too hot.)My guess is more than 20W, especially if the PCIe x8 slot is populated with a 10GbE or even 40GbE card. I threw a SolarFlare Flareon in mine and the power draw was so bad that it locked the machine up in Win10 IoT (although it could also be heat build-up. Jesus, that Flareon PTP card was a beast). Not sure if the situation will be significantly better in Linux.
If it's chasing watts for a firewall (say, with a quadport card), the t620 Plus with the Jaguar cores will be more efficient - that being said, how much of that idle watt consumption can be attributed to the Radeon R7 GPU, and is it possible to throttle the CPUs/GPUs down, either in Linux or on BIOS? I might have to do another downtime window and look at what tricks can be done on the t730 side to reduce power consumption.
Of course, if this little monster can do SR-IOV, it might change the equation entirely, since if it's possible for Proxmox to pass PCI virtual functions into multiple pfsense VMs talking to multiple VLANs, you can then run multiple firewall instances at close-to-wire speeds on the same physical device...
Hmm, mine after booting ubuntu server is ~16.5-17.5W. A lot of that must be your CX-2 card (and maybe the PCIe link?).Oof. Idling at the proxmox install screen right now with a CX-2 dual port VPI card installed: 33W.
your numbers are close to mine. i got 15.8W at idle after booting up pfsense with nothing else installed (no PCI-E card).Hmm, mine after booting ubuntu server is ~16.5-17.5W. A lot of that must be your CX-2 card (and maybe the PCIe link?).
durr@tinybox:~⟫ sudo smartctl -a /dev/sda
smartctl 6.5 2016-01-24 r4214 [x86_64-linux-4.4.0-31-generic] (local build)
Copyright (C) 2002-16, Bruce Allen, Christian Franke, www.smartmontools.org
=== START OF INFORMATION SECTION ===
Device Model: PSSBN016GA27MC0
Serial Number: 90530773070807202513
LU WWN Device Id: 5 000000 000000000
Firmware Version: SBFH00B8
User Capacity: 16,013,942,784 bytes [16.0 GB]
Sector Size: 512 bytes logical/physical
Rotation Rate: Solid State Device
Form Factor: M.2
Device is: Not in smartctl database [for details use: -P showall]
ATA Version is: Unknown(0x0ff8) (minor revision not indicated)
SATA Version is: SATA 3.2, 6.0 Gb/s (current: 6.0 Gb/s)
Local Time is: Mon Aug 20 21:57:46 2018 PDT
SMART support is: Available - device has SMART capability.
SMART support is: Enabled
durr@tinybox:~⟫ sudo hdparm -Tt /dev/sda
/dev/sda:
Timing cached reads: 3484 MB in 2.00 seconds = 1742.05 MB/sec
Timing buffered disk reads: 1268 MB in 3.00 seconds = 422.34 MB/sec
durr@tinybox:~⟫ bonnie++
Writing a byte at a time...
done
Writing intelligently...done
Rewriting...done
Reading a byte at a time...done
Reading intelligently...done
start 'em...done...done...done...done...done...
Create files in sequential order...done.
Stat files in sequential order...done.
Delete files in sequential order...done.
Create files in random order...done.
Stat files in random order...done.
Delete files in random order...done.
Version 1.97 ------Sequential Output------ --Sequential Input- --Random-
Concurrency 1 -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
tinybox 7G 647 99 246420 37 164373 24 2691 99 597928 38 +++++ +++
Latency 17138us 694ms 761ms 5228us 20737us 2444us
Version 1.97 ------Sequential Create------ --------Random Create--------
tinybox -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP
16 +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++
Latency 1719us 734us 537us 393us 66us 285us
1.97,1.97,tinybox,1,1534836828,7G,,647,99,246420,37,164373,24,2691,99,597928,38,+++++,+++,16,,,,,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++,17138us,694ms,761ms,5228us,20737us,2444us,1719us,734us,537us,393us,66us,285us
There's no such thing as the Jamestown Rev A or B on the t730 - Jamestown is the codename for the t620 motherboard. This is a t730 we are talking about here. Completely different machine with completely different performance characteristics.Does the T730 have the Jamestown Rev. A or B board? I am looking for a replacement for my WatchGuard XTM515 appliance running pfSense.
Yeah, Phison is one of those Taiwanese ODM (original design manufacturers) that almost no one outside the industry has ever heard of, like Wistron or Quanta. They usually make their marks on fabbing thumb drives for resellers like Sony.Misc stuff:
Power on hours is 1, so it sounds like this was indeed brand new.Code:durr@tinybox:~⟫ sudo smartctl -a /dev/sda smartctl 6.5 2016-01-24 r4214 [x86_64-linux-4.4.0-31-generic] (local build) Copyright (C) 2002-16, Bruce Allen, Christian Franke, www.smartmontools.org === START OF INFORMATION SECTION === Device Model: PSSBN016GA27MC0 Serial Number: 90530773070807202513 LU WWN Device Id: 5 000000 000000000 Firmware Version: SBFH00B8 User Capacity: 16,013,942,784 bytes [16.0 GB] Sector Size: 512 bytes logical/physical Rotation Rate: Solid State Device Form Factor: M.2 Device is: Not in smartctl database [for details use: -P showall] ATA Version is: Unknown(0x0ff8) (minor revision not indicated) SATA Version is: SATA 3.2, 6.0 Gb/s (current: 6.0 Gb/s) Local Time is: Mon Aug 20 21:57:46 2018 PDT SMART support is: Available - device has SMART capability. SMART support is: Enabled
I can find literally NO information on the disk (PSSBN016GA27MC0). Just looking at it, it's got a Phison PS3111-S11-13.
------Code:durr@tinybox:~⟫ sudo hdparm -Tt /dev/sda /dev/sda: Timing cached reads: 3484 MB in 2.00 seconds = 1742.05 MB/sec Timing buffered disk reads: 1268 MB in 3.00 seconds = 422.34 MB/sec durr@tinybox:~⟫ bonnie++ Writing a byte at a time... done Writing intelligently...done Rewriting...done Reading a byte at a time...done Reading intelligently...done start 'em...done...done...done...done...done... Create files in sequential order...done. Stat files in sequential order...done. Delete files in sequential order...done. Create files in random order...done. Stat files in random order...done. Delete files in random order...done. Version 1.97 ------Sequential Output------ --Sequential Input- --Random- Concurrency 1 -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks-- Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP tinybox 7G 647 99 246420 37 164373 24 2691 99 597928 38 +++++ +++ Latency 17138us 694ms 761ms 5228us 20737us 2444us Version 1.97 ------Sequential Create------ --------Random Create-------- tinybox -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete-- files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP 16 +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++ Latency 1719us 734us 537us 393us 66us 285us 1.97,1.97,tinybox,1,1534836828,7G,,647,99,246420,37,164373,24,2691,99,597928,38,+++++,+++,16,,,,,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++,17138us,694ms,761ms,5228us,20737us,2444us,1719us,734us,537us,393us,66us,285us
Oh wow, if you let the display output go to sleep, or just disconnect it, the idle power draw drops to ~11.7 watts.
Edit: Hey, the SSD is actually Phison branded:
View attachment 9110
It's a "GA5HSN27P92-S110J0", but I don't get anything for that either, with or without the S110J0 suffix.
It has 2 flash memory ICs:
View attachment 9111
TA16G5LARA
K1706
E72T026A00
I'm not sure who makes them either.
Hm. Which version of the BIOS did you upgrade to, and did it cause any regressions on the AMD DMAR/PCI passthrough functionality?So power settled down later to around ~26W idle which is closer to what I expected (~16W idle + ~10W PCIe card.)
Sweet, the bios sees the m.2 drive as a uefi target after the update. The bios fan speed seems to stick once booted now too which is a good thing -- the Mellanox nic was getting *hot*.
Edit 2: If anyone receives a machine with a wiped SSD you'll be missing the EFI partition and won't be able to update directly from within the BIOS.
I had to download HP's "ThinUpdate" here HP t730 Thin Client - Driver Downloads | HP® Customer Support and run it under windows to download the ThinPro 7 image and write it to a USB key to install. You also have to unpack the BIOS update on a windows machine and copy that over to usb as well. I'm going to pull an image off of a usb stick that has the bios files pre-included in case anyone wants to skip all of the windows steps and jump right into booting and flashing.
01.11 Rev AHm. Which version of the BIOS did you upgrade to, and did it cause any regressions on the AMD DMAR/PCI passthrough functionality?
enp1s0f0:
Boot image Option ROM and UEFI
Link speed Negotiated automatically
Link-up delay time 5 seconds
Banner delay time 2 seconds
Boot skip delay time 5 seconds
Boot type PXE
PF MSI-X interrupt limit 32
SR-IOV Enabled
Virtual Functions on each PF 127
VF MSI-X interrupt limit 1
enp1s0f1:
Boot image Option ROM and UEFI
Link speed Negotiated automatically
Link-up delay time 5 seconds
Banner delay time 2 seconds
Boot skip delay time 5 seconds
Boot type PXE
PF MSI-X interrupt limit 32
SR-IOV Enabled
Virtual Functions on each PF 127
VF MSI-X interrupt limit 1
[Tue Aug 21 23:40:19 2018] pci 0000:01:00.0: [1924:0803] type 00 class 0x020000
[Tue Aug 21 23:40:19 2018] pci 0000:01:00.0: reg 0x10: [io 0xe100-0xe1ff]
[Tue Aug 21 23:40:19 2018] pci 0000:01:00.0: reg 0x18: [mem 0xfd000000-0xfdffffff 64bit]
[Tue Aug 21 23:40:19 2018] pci 0000:01:00.0: reg 0x20: [mem 0xfe090000-0xfe09ffff 64bit]
[Tue Aug 21 23:40:19 2018] pci 0000:01:00.0: reg 0x30: [mem 0xfe040000-0xfe07ffff pref]
[Tue Aug 21 23:40:19 2018] pci 0000:01:00.0: PME# supported from D0 D3hot
[Tue Aug 21 23:40:19 2018] pci 0000:01:00.0: reg 0x184: [mem 0x00000000-0x00001fff 64bit]
[Tue Aug 21 23:40:19 2018] pci 0000:01:00.0: VF(n) BAR0 space: [mem 0x00000000-0x000fdfff 64bit] (contains BAR0 for 127 VFs)
[Tue Aug 21 23:40:19 2018] pci 0000:01:00.0: reg 0x18c: [mem 0x00000000-0x0000ffff 64bit]
[Tue Aug 21 23:40:19 2018] pci 0000:01:00.0: VF(n) BAR2 space: [mem 0x00000000-0x007effff 64bit] (contains BAR2 for 127 VFs)
[Tue Aug 21 23:40:19 2018] pci 0000:01:00.1: [1924:0803] type 00 class 0x020000
[Tue Aug 21 23:40:19 2018] pci 0000:01:00.1: reg 0x10: [io 0xe000-0xe0ff]
[Tue Aug 21 23:40:19 2018] pci 0000:01:00.1: reg 0x18: [mem 0xfc000000-0xfcffffff 64bit]
[Tue Aug 21 23:40:19 2018] pci 0000:01:00.1: reg 0x20: [mem 0xfe080000-0xfe08ffff 64bit]
[Tue Aug 21 23:40:19 2018] pci 0000:01:00.1: reg 0x30: [mem 0xfe000000-0xfe03ffff pref]
[Tue Aug 21 23:40:19 2018] pci 0000:01:00.1: PME# supported from D0 D3hot
[Tue Aug 21 23:40:19 2018] pci 0000:01:00.1: reg 0x184: [mem 0x00000000-0x00001fff 64bit]
[Tue Aug 21 23:40:19 2018] pci 0000:01:00.1: VF(n) BAR0 space: [mem 0x00000000-0x000fdfff 64bit] (contains BAR0 for 127 VFs)
[Tue Aug 21 23:40:19 2018] pci 0000:01:00.1: reg 0x18c: [mem 0x00000000-0x0000ffff 64bit]
[Tue Aug 21 23:40:19 2018] pci 0000:01:00.1: VF(n) BAR2 space: [mem 0x00000000-0x007effff 64bit] (contains BAR2 for 127 VFs)
[Tue Aug 21 23:40:19 2018] pci 0000:00:02.1: PCI bridge to [bus 01]
[Tue Aug 21 23:40:19 2018] pci 0000:00:02.1: bridge window [io 0xe000-0xefff]
[Tue Aug 21 23:40:19 2018] pci 0000:00:02.1: bridge window [mem 0xfc000000-0xfe0fffff]