A list of puns related to "Serial Attached Scsi"
I'm currently wrapping up a project that focuses on hardware RAID controllers, each controller has a unique SAS address ( sort of like a MAC address for network devices ). I was wondering if anyone knows of a database of those addresses mapping them to vendors or hardware device models.
An example of such a tool for MAC addresses can be found here: http://www.coffer.com/mac_find/
I'm trying to understand what I've been reading about Serial Attached SCSI the last few hours. I haven't owned a SCSI controller for over 10 years. I used to have a Plextor CD burner and an IOmega Jaz drive attached to my Amiga, later to my PC, and back then it was SCSI-II, daisy chaining and terminating the last device. With SAS, all this seems to be gone, and now one connector on the controller equals one connectable device (except if you use an extender, whatever that is.) Wikipedia only gets me so far, the official SCSI FAQ is old and short, and various tech sites does not seem to cover SAS the same way SATA is covered.
So I rely on you, Dear Reddit, and hope some of you have insights to share about the following:
What is the difference between a port and a connector? See e.g. this product specification of a controller from Dell (PDF warning), saying it has eight ports on two connectors. I'm trying to wrap my head around this concept. The most obvious answer to me would be that four ports describes a controller limitation, where it can not address more than four devices on one physical connector (assuming you are using the mythical expander device). Except, that doesn't make sense, as in the Wikipedia article, it states that with the use of expanders, the limit is 65535 devices. (I'm now starting to assume the expander is a connector splitter)
What is the deal with the bandwidth limitation? IS it 3 Gbps or 6 Gbps? The PDF linked above says 6 Gbps when used with full duplex, which I believe means connecting two connectors on the controller to two connectors on the hard drive. Am I right? The Wikipedia article mentions 6 Gbps, but not how this is achieved. However, the top-right box on the Wikipedia page says "Full-duplex with link aggregation (4-ports wide at 24 Gbit/s)". No further explanation. Assuming "4-ports" mean four connectors, that still is 6 Gbps per connector, which means two connectors aren't used to achieve full duplex.
This page says that the 6 Gbps claim is basically bullsh*t, as with full duplex, only one of the 3 Gbps streams are used for actual data, and the other stream is used for control commands (and possibly error checking?). As I can't find these claims anywhere else, do anyone know if there is any truth to it? I ca
I found this old serial to usb cable that I could connect to the Mac SE SCSI drive port. My thinking is that I could somehow emulate a Mac's SCSI drive with my pc and mount images via from my laptop. My computer is recognizing the serial cable and the mac is trying to boot off the cable as if it was the hard drive. I feel there's some promise in this but not sure if this is even possible. I've been searching for any emulation software that could accomplish this but not having much luckl. Is this a viable means of file transfer or am I barking up the wrong tree?
I just sent a support request to Gigabyte and I'm pasting it here in case someone else ends up with this issue.
I'll keep this post up to date with their replies and will update with new benchmarks if the issue changes.
**This is not a tech support thread as the issue is beyond what we can do (either hardware or BIOS related).**
Note that I can only speculate that the issue is UAS related, since I only have one such drive.
---
I recently put together a new PC using a Gigabyte Aorus Elite x570 and a Ryzen 3700x.
The system runs fine but today I discovered a possible hardware/BIOS issue.
The issue is very serious to me, as it breaks my backup workflow.
To exclude software interference, I tested on a live linux (slax) and the issue is consistent to reproduce.
I also tested on a FRESH Windows, just in case. The issue is the same.
The issue:
Connecting a UAS (SCSI) 3.0 device to a USB 3.0 or USB 3.1 port on the motherboard makes the writes excruciatingly slow: about 70KB/s (yes, KILO BYTES).
I'll describe below my simple benchmarks.
The issue happens on a Seagate Backup Plus Slim 2TB USB 3.0 device.
It is a UAS (USB Attached SCSI) device.
Model number: STDR2000101; part number: 1K9AA7-571
Connecting the device to a mobo USB 3.0 or 3.1 triggers the issue.
It has normal performance on other computers with USB 3.0 (tested with 2 other Windows computers).
Connecting the device to a USB 2.0 works normally (but at normal USB 2.0 speeds).
Connecting the device to an external USB 3.0 (a PCI-e card) works normally at normal speeds.
ONLY the USB 3.0/3.1 fails and ONLY with the UAS device.
A SAMSUNG USB 3.0 (not UAS) has no issues and performance is normal.
** REFERENCE TEST
Tests below are using an external PCIE USB 3.0 card. Performance is normal for both disks. The SAMSUNG has similar performance when connected directly to motherboard.
SEAGATE
root@slax:/mnt# dd if=/dev/zero of=/mnt/seagate/TEST conv=fdatasync bs=384k count=1k; rm -f /mnt/seagate/TEST
1024+0 records in
1024+0 records out
402653184 bytes (403 MB, 384 MiB) copied, 8.24132 s, 48.9 MB/s
SAMSUNG
root@slax:/mnt# dd if=/dev/zero of=/mnt/samsung/TEST conv=fdatasync bs=384k count=1k; rm -f /mnt/samsung/TEST
1024+0 records in
1024+0 records out
402653184 bytes (403 MB, 384 MiB) copied, 7.78663 s, 51.7 MB/s
** using mobo USB 3.0 (larger test with 400MB data)
SEAGATE
IMPOSSIBLE TO RUN because it would take too long to writ
... keep reading on reddit β‘Hello everybody.
I just registered to Reddit, to ask my question. I'm glad to see, that this "branch" is so big and vital! My question is as mentioned in the topic. In the following i list my struggle and configuration with specs:
Specs:
Linux debian 4.19.0-6-amd64 #1 SMP Debian 4.19.67-2+deb10u2 (2019-11-11) x86_64 GNU/LinuxQEMU emulator version 3.1.0 (Debian 1:3.1+dfsg-8+deb10u3)Using virtio-win-0.1.173 drivers.SATA BD burner.Windows 10: KVM x86_64 with i440FX on OVMFWindows /: KVM x86_64 with Q35 on OVMF
Conf:
VirtIO-Blk
<disk type='block' device='cdrom or disk'>
<driver name='qemu' type='raw' cache='none' io='native' discard='ignore' detect_zeroes='off'/>
<source dev='/dev/disk/by-id/burning_device'/>
<blockio logical_block_size='2048' physical_block_size='2048'/> <!-- with and without this line -->
<target dev='vda' bus='virtio'/>
</disk>
The result is, that the Drive shows as unformatted hard disk, while the medium is a BD-RE and NOT empty at all. Using the SATA or SCSI controller doesn't produce this problem, but isn't meant for/able to burn-ing.
VirtIO-SCSI
<controller type='scsi' index='0' model='virtio-scsi'/>
<hostdev mode='subsystem' type='scsi' managed='no' sgio='unfiltered' rawio='yes'>
<source>
<adapter name='scsi_host#'/>
<address bus='0' target='0' unit='0'/>
</source>
<address type='drive' controller='0' bus='0' target='0' unit='#'/>
</hostdev>
or
<disk type='block' device='lun'>
<driver name='qemu' type='raw' cache='none' io='native' discard='ignore' detect_zeroes='off'/>
<source dev='/dev/disk/by-id/burning_device'/>
<target dev='sda' bus='scsi'/>
<address type='drive' controller='0' bus='0' target='0' unit='#'/>
</disk>
With this form of attachment, the content of the medium is displayed fine. But Windows doesn't recognize the drive as burner, as well. Neither can i format nor burn any file onto it. Although it is displayed as "BD-RE" drive. I've spent countless hours learning and trying. There are no other options available. Can anyone help? Thank you!
I could use an example of PowerShell being used to eject a USB Attached SCSI (UAS) Compatible Device (device type 3) . There must be a way to do this because the device can be ejected using 'Safely remove hardware and eject media' tool. I can eject regular USB thumb drives (device type 2) with PowerShell, no problem. My Google skills have failed me.
Hello everyone,
I am encountering a strange issue for a while now: My Hyper-V Machine has an Ubuntu VM running, which provides all others with network drives via smb,nfs and afp.
It has two USB drives and five VHDX files attached.
I noticed that when I reboot the host, my drives in Ubuntu are mixed up, for example sde1 and sdf1 are mixed up.
This creates several issues in the network including all docker containers defaulting settings.
I tried to attach the drives to different SCSI controllers and it seemed to help, but there are no more than four possible and I needed to add another VHDX.
Right now it seems that only drives attached to the same controller are mixed up. I cannot imagine this being an ubuntu error.
Does anyone know how to troubleshoot this or has ever encountered it? Does it maybe have to do with the number you set in the location dropdown?
Have tried to fix this for months now and just thought of maybe asking here.
Thanks.
Richard
I was using my new 10TB drives inside their enclosures before shucking and I was getting writes of ~130MB/s from linux (limited by the SATA drive I was reading from - that's not a benchmarked upper limit.)
After shucking I put one of my older Seagate Archive (SMR) 8TB drives into the enclosure and noticed I'm only getting writes of 30MB/s from linux. I checked how the drive was being seen in linux and noticed that it doesn't appear to support UASP in this configuration.
Can someone who hasn't shucked the drives tell me if pre-shucked the enclosures support UASP?
When I run 'lsusb -t' here's what I see:
/: Bus 02.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/10p, 5000M
|__ Port 6: Dev 2, If 0, Class=Mass Storage, Driver=usb-storage, 5000M
In contrast here's what you would see if UASP is being used:
/: Bus 02.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/6p, 5000M
|__ Port 2: Dev 5, If 0, Class=Mass Storage, Driver=uas, 5000M
|__ Port 4: Dev 3, If 0, Class=Mass Storage, Driver=usb-storage, 5000M
I'm just trying to tell if the drive is slower because it is a SMR disk or because the enclosure hobbles itself when the drive which came with it isn't inside.
(If the latter case, any recommendations on what to do with this enclosure to change this?)
The cable is a dark grey belkin, so I'm kind of guessing it's serial... I really need to get this zip drive running with it so I can actually boot
http://imgur.com/l1AOofv,sKcYnn5#0
http://imgur.com/l1AOofv,sKcYnn5#1
I got a couple of these bad boys and they are still in the packaging.
Trying to get $75.00 each with shipping.
I have only a vague whisp of a memory. Read this around 1996.
Looking for a Detective who is hunting for a serial killer. The city in which he operated was physically divided with a barrier possibly huge wall between the rich and poor areas. The killer would take women and kind of weld metal wings to there skin transforming them into angels. Think the detective had a drug problem(but dont they all)
Have a power pc 6100/66 , read the manual (here incase anyone needs for their 6100series https://download.info.apple.com/Apple_Support_Area/Manuals/desktops/0305858APM6100UM.PDF too forever to find it lol) and cannot find any manuals for the external HDD which is a microtech N50 in the picture... so my questions is do i need to terminate? I bought a Activate terminator as shown in the original SE/30 manual below which is very similar to my external drive... I cannot find anywhere to clarify if i need to terminate or not? I`ll be connecting using a DB25 to Centronics 50..
https://preview.redd.it/q1u4dey95ye81.png?width=653&format=png&auto=webp&s=b8b6ca7615f061552cedcf3afd5f58a4b3591ab9
https://preview.redd.it/8n1q7w1o5ye81.png?width=640&format=png&auto=webp&s=01b1afd12b2ef44a6043c3f3cfac4a4c93c3602b
Hi, searched a lot and so far, not found the answer.
I want to detect when a Nano is connected to the PC vial Serial so I can adjust the nRF24 radio power. When testing I need to set it to Low as it is on the bench beside me. In real use I it will be battery powered and I need it High.
I have tried writing to the Serial then printing the Bytes sent, but it shows it sends, even if it is never received.
Is there some form of ACK or some other way of doing this?
Thanks.
I have an old APS external SCSI hard drive sitting in a box. (It was used with my SE or 6100 back in the day, both of which are long gone.) Assuming the drive even works anymore, is there any way to get the data off of it other than finding an old Mac or tracking down a very expensive SCSI to USB adapter?
(I realize the answer is probably "no", but I figured I would ask anyhow.)
These are basically vintage items. The Exabyte tape drive definitely works, I've used it recently. The CD-ROM unit I haven't used, but it functions as part of the SCSI chain (I'm using an LVD connector on my PC to connect to the regular SCSI MD50 connector on the CD-ROM unit, and then that connects to the DB50 on the tape unit. These cables will be included (except for the LVD SCSI cable for my PC).
Cables and terminators included (unless you don't need them). Free but you pay for postage from Ireland.
HP Scanjet 6100C has a SCSI port.
What do I need to get it working with Win10?
Hi all,
I bought the Indigo 2 on eBay, but both drive trays are empty. It looks like this connector is proprietary (something like 80 pin?), can it be replaced with something? I have a SCSI2SD board with a DB25 connector, but I was not able to find a proper combination of adapters to connect all this together :(
There is also a SCSI-2 port at the back of the computer - can the system boot from it without having any other drives?
Tnx
I have a VM for which I wish to pass through several USB HDDs (U. What is the best way to do so? Right now I pass the USB device through. Would it better to pass the drive through to the vm as a scsi device, i.e.:
qm set <VMID> -scsiX /dev/disk/by-id/...
Are there advantages (performance or other) of one over the other?
Hello, I have a Quadra 950 that Iβm having some issues with the scsi bus. Awhile back I was having issues connecting to an MO drive that I had connected to it externally. Then it would no longer boot from the hard disk so I assumed the Hard drive finally died (itβs a double height 5 1/4 drive). I put it away for awhile until I had the time to set up a scsi2sd for it. I then installed a dummy image I made to test if it would boot from the scsi2sd correctly and then I install A/UX without issue. It worked for a bit until I removed the drive so could add another drive to the scsi2sd. When I put it back in I get the floppy with the question mark, even when I go back to the test dummy image I made.
I was wondering if anyone had any experience with the 950 that can help me troubleshoot it. I am about to take it apart to check the fuses for the scsi bus to verify the termination power to the bus, but I would like to know if there are any other steps that I can take.
This is the first time I'm using a SCSI device, so pardon my ignorance here.
I'm testing external LTO-4 SCSI tape drive : Quantum TC-L42BX LTO 4 Ultrium External Tape Drive 3.5" SCSI BRSLA-0704-AC (here) which should support 80MB/s.
Connected to LSI Logic LSI00154 LSI20320IE 1CH U320 SCSI PCI Express Card (here)
I'm using Ubuntu Server on i7-6700 but I'm getting slow write speed about 10MB/s.
$ tapestat -m 5
Linux 5.4.0-96-generic 01/24/2022 _x86_64_ (8 CPU)
Tape: r/s w/s MB_read/s MB_wrtn/s %Rd %Wr %Oa Rs/s Ot/s
st0 0 1119 0 10 0 54 54 0 0
Tape: r/s w/s MB_read/s MB_wrtn/s %Rd %Wr %Oa Rs/s Ot/s
st0 0 1222 0 11 0 59 59 0 0
Also I notice that w/s is high.
Currently, I suspect the cable but I'm not sure how to diagnose.
Any idea how to proceed from here?
Edit: adding the mt status output.
$ mt -f /dev/st0 status
SCSI 2 tape drive:
File number=0, block number=0, partition=0.
Tape block size 0 bytes. Density code 0x46 (LTO-4).
Soft error count since last status=0
General status bits on (41010000):
BOT ONLINE IM_REP_EN
Please note it says "SCSI 2 tape drive". Not sure if this is expected.
I could use an example of PowerShell being used to eject a USB Attached SCSI (UAS) Compatible Device . There must be a way to do this because the device can be ejected using 'Safely remove hardware and eject media' tool. I can eject regular USB thumb drives with PowerShell, no problem. My Google skills have failed me.
Upon further review...
USB 3.0 supports the USB Attached SCSI protocol (UASP) which uses the SCSI command set. To "Eject" the drive some type of SCSI hot plug operation is probably required. My PowerShell Fu is too weak to dive deep enough to figure this out.
Without knowing the name of the hard disk, how can I get the hard disk attached to a certain SCSI controller from one VM?
The SCSI controller will only have ONE hard disk attached so...
Please note that this site uses cookies to personalise content and adverts, to provide social media features, and to analyse web traffic. Click here for more information.