casperse

Members
  • Posts

    800
  • Joined

  • Last visited

Posts posted by casperse

  1. On 8/17/2023 at 1:07 PM, JorgeB said:

    ECC can usually correct single bit errors and detect multi bit.

     Last Question 🙂
    These new Corsair Vengeance DDR5-5600 BK C40 DC - 96GB x 2 = 192G
    Are this "true" ECC? - I do like the idea to have a total of 192G of RAM

    Or is it better to get the Kingston you linked to?  (Only 4800 the above are 5600)
    I actually think getting the right RAM is the most difficult purchase in a new build today! LOL

  2. Hi All

     

    I did my last upgrade through Unraid connect, this was great! because the server was in a different location.
    And I could login and start my array (Encrypted).

     

    Today I tried this again, and for some reason I cant connect to the server anymore?

    I did read the release notes and my LAN config is standard bridge with eth0 no special changes

    Just wanted to hear if anyone else experienced the same?

    Contemplating to cut the power (remotely) for a reboot 🙂

    Cheers

     

  3. On 8/14/2023 at 12:36 PM, JorgeB said:

    I usually buy Kingston for me, in part because of availability, Hynix, Micron or Samsung should also be good options.


    But is the DDR5 "real" ECC? my old ECC RAM for my Xeon CPU is listed as:
    image.png.0d577d1a1e58ab6ec74600f6571f7707.png

    Would these new DDR5 be same level of ECC RAM as the old orig. ones?
    Sorry my reason for asking is that I read somewhere that the level of ECC validation was different?

    Shouldn't it be multi-bit?

  4. Hi All

     

    I am looking into upgrading my UNRAID server but I cant really find out if the RAM I find is "True" ECC like the one I have in my current Intel’s Xeon E-2176G build?

    So far I have found the new 96G modules:
    Corsair Vengeance DDR5-5600 BK C40 DC - 96GB x 2 = 192G
    https://www.proshop.dk/RAM/Corsair-Vengeance-DDR5-5600-BK-C40-DC-96GB/3143077

    OR
    Kingston FURY Beast - DDR5 - sæt - 64 GB: 2 x 32 GB - DIMM 288-PIN - 5200 MHz / PC5-41600 - CL38 - 1.25 V - ikke bufferet - on-die ECC
    https://www.computersalg.dk/i/8080944/kingston-fury-beast-ddr5-sæt-64-gb-2-x-32-gb-dimm-288-pin-5200-mhz-pc5-41600-cl38-1-25-v-ikke-bufferet-on-die-ecc?utm_source=pricerunner&utm_medium=911-it-elektronik-komponenter-ram&utm_campaign=kingston-technology&utm_content=0
     

    I would appreciate any recommendation on what RAM to get? (ECC minimum 128G in 4 modules)

    As always thanks for your input the constructive feedback is the best about running UNRAID 🙂

     

  5. Hi All

     

    I had to cut the power yesterday, and now again I cant stop my server keep getting the:

    Jul 28 18:53:57 PLEXZONE emhttpd: shcmd (71069): umount /mnt/cache_appdata
    Jul 28 18:53:57 PLEXZONE root: umount: /mnt/cache_appdata: target is busy.
    Jul 28 18:53:57 PLEXZONE emhttpd: shcmd (71069): exit status: 32
    Jul 28 18:53:57 PLEXZONE emhttpd: Retry unmounting disk share(s)...
    Jul 28 18:54:02 PLEXZONE emhttpd: Unmounting disks...
    Jul 28 18:54:02 PLEXZONE emhttpd: shcmd (71070): umount /mnt/cache_appdata
    Jul 28 18:54:02 PLEXZONE root: umount: /mnt/cache_appdata: target is busy.
    Jul 28 18:54:02 PLEXZONE emhttpd: shcmd (71070): exit status: 32
    Jul 28 18:54:02 PLEXZONE emhttpd: Retry unmounting disk share(s)...
    Jul 28 18:54:07 PLEXZONE emhttpd: Unmounting disks...
    Jul 28 18:54:07 PLEXZONE emhttpd: shcmd (71071): umount /mnt/cache_appdata
    Jul 28 18:54:07 PLEXZONE root: umount: /mnt/cache_appdata: target is busy.
    Jul 28 18:54:07 PLEXZONE emhttpd: shcmd (71071): exit status: 32
    Jul 28 18:54:07 PLEXZONE emhttpd: Retry unmounting disk share(s)...
    Jul 28 18:54:12 PLEXZONE emhttpd: Unmounting disks...
    Jul 28 18:54:12 PLEXZONE emhttpd: shcmd (71072): umount /mnt/cache_appdata
    Jul 28 18:54:12 PLEXZONE root: umount: /mnt/cache_appdata: target is busy.
    Jul 28 18:54:12 PLEXZONE emhttpd: shcmd (71072): exit status: 32
    Jul 28 18:54:12 PLEXZONE emhttpd: Retry unmounting disk share(s)...
    Jul 28 18:54:17 PLEXZONE emhttpd: Unmounting disks...
    Jul 28 18:54:17 PLEXZONE emhttpd: shcmd (71073): umount /mnt/cache_appdata
    Jul 28 18:54:17 PLEXZONE root: umount: /mnt/cache_appdata: target is busy.
    Jul 28 18:54:17 PLEXZONE emhttpd: shcmd (71073): exit status: 32
    Jul 28 18:54:17 PLEXZONE emhttpd: Retry unmounting disk share(s)...

     

    I did a Diagnostic file attached below, hope some one can help me?

    VM and dockers are stopped!

     

    I finally got it stopped using:
    umount -l /dev/loop2

     

    But I would like to know the cause of this 🙂

    plexzone-diagnostics-20230728-1855.zip

  6. 10 hours ago, OrneryTaurus said:

     

    That chassis does not support desktop power supplies. Best to buy at least one PSU from the manufacturer so you can utilize all the available connections it's intended to have.

     

    Server grade power supply setups are intended to be hot swappable when you have redundancy to reduce downtime.

     

    One thing to watch out for is what kind of power is needed for these PSUs. For my server, I have https://store.supermicro.com/1280w-1u-pws-1k28p-sq.html which have a rating of: 1000W: 100-140V; 1280W: 180-240V. Meaning I can only pump out 1000W of power when hooked up to a standard 120V / 15A circuit and plug. In order to get 1280W, I would need to have a 240V / 20A circuit and plug. Thankfully it supports both but that may not be the case for the 2000W PSU, it may require the 240V / 20A circuit and plug, this of course is providing it's a US plug setup.

     

     

    That's a valid point (According to the vendor it should support a normal PSU (Replacing the dual server PSU's) but not sure this is the case (They have been wrong before)

  7. On 5/26/2023 at 11:30 PM, OrneryTaurus said:

    That In-Win case is definitely half height slots as the design is exactly like a supermicro 36-bay. I would go with your original case.

    Thanks @OrneryTaurus I actually confronted the seller again that kept claiming this was full height and she told me she got wrong info from the manufactor
    So I am back to the old case and full extra height for a very big air cooler 🙂 just need to get a better deal.

    They have 2000W PSU with redundancy but I have never had anything like this? It does support a normal PSU but I can get a dual setup for the same money?
     

  8. 14 hours ago, SimonF said:

    Are you able to see what devices are mapped in the VM in HA? You may need to change the serial port

    I finally took the courage to delete and re-add the ZHA integration in HA (Hoping it would recover all my devices again!)

    And the new path is:

     

          {
            "entry_id": "1c7ac4a53b85198687c853f65f3a7fa6",
            "version": 3,
            "domain": "zha",
            "title": "/dev/ttyUSB0",
            "data": {
              "device": {
                "path": "/dev/serial/by-id/usb-QEMU_QEMU_USB_SERIAL_1-0000:00:07.0-4-if00-port0"
              },
              "radio_type": "deconz"
            },
            "options": {},
            "pref_disable_new_entities": false,
            "pref_disable_polling": false,
            "source": "user",
            "unique_id": null,
            "disabled_by": null
          }

    And it seem to be working for both USB now! 🙂
    But only one is listed as connected? the other one still says "Virsh Error"?
    Both use the port 4 as "Connect as serial" I have not been able to change them.
    Can I just ignore this error? it looks like it working ok in HA
    image.thumb.png.c61a51de496b5b4e75c6be69e528c7c4.png

  9. On 6/4/2023 at 12:42 PM, SimonF said:

    Connected out side means the device was connected outside USBM. Can you provide the output of 

    cat /usr/local/emhttp/state/usb.ini | grep error

     

    or

     

    cat /usr/local/emhttp/state/usb.ini | grep vir

     

    Thanks (Everything worked before the Unraid update)
    root@:~# cat /usr/local/emhttp/state/usb.ini | grep error
    virsherror = 1
    virsh = "error: Failed to attach device from /tmp/libvirthotplugusbbybusHassio 1-001-059.xml
    error: XML error: Duplicate USB address bus 0 port 4
    root@~# cat /usr/local/emhttp/state/usb.ini | grep vir
    virsherror = 1
    virsh = "error: Failed to attach device from /tmp/libvirthotplugusbbybusHassio 1-001-059.xml
    root@:~# 

     

    UPDATE (I am really trying to get this working not a happy house when Hassio isnt running 🙂
    I have tried detach and attach and now get this:
    image.thumb.png.8f39d66cd44bd62bda2477785a3ff45d.png

    The one that says connected dosent work in HA?
    I Dont now what the correct configuration should be in HA?
    image.thumb.png.c56d98d69125e961e03cb8d85d6923ba.png

    If anyone knows what the Blue line should be then let me know 🙂

  10. I found the HA settings, not sure what to change:

          {
            "entry_id": "2bb51ca17b7fa61b8d997ce1d3f2c8a2",
            "version": 3,
            "domain": "zha",
            "title": "ConBee II, s/n: DE2421314 - dresden elektronik ingenieurtechnik GmbH",
            "data": {
              "device": {
                "path": "/dev/serial/by-id/usb-QEMU_QEMU_USB_SERIAL_1-0000:00:07.0-4-if00-port0"
              },
              "radio_type": "deconz"
            },

    The mappings now looks like this:

    image.thumb.png.84d9007057e3df81c80256d9c601e000.png

  11. On 5/13/2023 at 8:17 AM, OrneryTaurus said:

    That's a beefy chassis. Expander backplanes should allow you to use a single cable per backplane to see the drives IIRC. You will need at least 2, maybe 3 cables to connect to it (not sure if they split the front into two backplanes or not).

     

    I was looking at something similar from the same place for more storage with full height slots. However, I personally picked up a 36-bay supermicro superchassis and upgraded from a P2000 GPU to a Tesla P4 instead ($800 all in for the two).

     

    Only issues I can think of would be related to shipping and availability of replacement parts / racking the damn thing.

    $800 Nice price! (I haven't found anything in that pricerange) I did 'nt know about the Tesla 4 card but I have other cards and I dont want to replace them all with other ones that are 1/2 height 🙂
    Also the future of cooling new CPU looks like it will requires bigger fans so the extra height would allow me to use something like a Noctua NH-D15.
    Yes the shipping and quality of the case would be a risk I agree. But so was buying my last case from the UK and so far that has been working great
    Old Norco RPC-4224 4U (UK SC-4324S) I am talking to them about another case with 36 drives that they claim have full PCie height?
    I have written to them again and they say it is full height?

     

    In-Win IW-RS436-07 - 2U Server Chassis with 36x SATA/SAS/NVMe Hot-Swap Bays - Includes 1200W CRPS Redundant PSU and Rail Kit - Server Case

     

    I think this one is better build quality and the price of this i allot higher! But I will get 2xPSU (1200W) with me in this buy! (Not sure this would

    have all the new power cables needed for new CPU's & Motherboards/GPU etc.

     

    Drawbacks: "Only 36 drives and not room for the extreme high CPU fan coolers"

     

    I just cant believe that this is really full height PCie? (The seller claims this, but the pictures doesn't look it? or does it?)
    image.png.fb3a29f55d7e6059868fc907832d37b3.png

     

    image.thumb.png.0a0908733c98e2905544998c6ab3e81b.png

     

    So either go for this or try a the Innovision case?


    Thanks for your feedback! @OrneryTaurus much appreciated!
    I just really want one new case to "rule them all" the 72 is a beast and if the quality is ok, I might dare to roll the dice on it🙂 or buy some spare parts when ordering?
    I could even have two Unraid servers running in the same case, and use the old server for backup and experimenting with Proxmox and VM's

     

     

     

     

  12. Hi All

    I always wanted a Storinator case but the cost is just to high!

    Lately I might have found a alternative, that might solve allot of my problems?

     

    Most cases that have +36 drive bays and above only have half heights on PCie slots! (You cant even have a P2000 in it!)
    Also the new CPU's are requiring allot more cooling, so having half the Hight for a fan cooler isn't really ideal.

     

    So This might be a solution? (I would appreciate any input to possible problems?)

    I am new to backplane extenders and not sure my existing HBA would be compatible with this case?

     

    My HBA is a LSI Logic SAS 9305-24i Host Bus Adapter - x8, PCIe 3.0, 8000 MB/s 

    Would this be able to utilize all these bays?

     

    I also have, from my last build:

    LSI Logic SAS 9201-8i PCI-e Controller / 9211-8i (IT-mode) / Full height bracket (2 connectors = 8 Sata)

    LSI Logic SAS 9201-16i PCI-e 2.0 x8 SAS 6Gbs HBA Card (4 connectors = 16 Sata)

     

    Again I really like the airflow and height of this case for a full height CPU air cooler!
    and the 8 x 120mm fans (According to the specs) - noise is not an issue

     

    High Dense Rack Mount 8U 72bay Top-loaded Storage Server Chassis

    image.thumb.png.6d8505191e9e03a6ca33d1adffa9a0af.png

     

     

     

     

     

  13. Hi Jorge B.

    It just finished it must have speeded up in the end....and it says that the array is ok for both drives now?

    When it was running it only stated it was rebuilding the 18TB drive and not Drive 1 (12TB) but according to Unraid my array is fine now?

    Anyway I am now looking into building another Unraid server as a Backup server (This was kind of a wakeup call)

    Again thanks for all your help so happy to be up and running again with parity drives on all drives!

  14. Hi JorgeB
     

    Speed is at highest 120MB/sec and now pretty low 8.6 MB/sec most of the time:

    image.thumb.png.d37ee83ae219c7ff71a3dde4ea24265f.png

     

    lspci -d 1000: -vv
    image.thumb.png.32c426cff4a761201a0aa8224d266c13.png

     

     

    Quote

    You can try a different PCIe slot for the HBA, if it's in a CPU slot try a PCH one, or vice versa.


    My current PCie slot and placement of HW:

    PCIe slot 1: x8 NVIDIA Quadro P2000

    PCIe slot 2: x4 NVIDIA GeForce RTX 3060

    PCIe slot 3: x8 LSI Logic SAS 9305-24i Host Bus Adapter

    PCIe slot 4: x4 M.2. NVMe ICY BOX: IB-PCI215M2-HSL adapter

     

    New placement?

    PCIe slot 1: x8 LSI Logic SAS 9305-24i Host Bus Adapter

    PCIe slot 2: x4 NVIDIA Quadro P2000

    PCIe slot 3: x8 NVIDIA GeForce RTX 3060

    PCIe slot 4: x4 M.2. NVMe ICY BOX: IB-PCI215M2-HSL adapter