Jump to content
RESET Forums (homeservershow.com)

Leaderboard

  1. schoondoggy

    schoondoggy

    Moderators


    • Points

      6

    • Content Count

      8,774


  2. nrf

    nrf

    Members Plus


    • Points

      3

    • Content Count

      1,064


  3. LVLouisCyphre

    LVLouisCyphre

    Members


    • Points

      2

    • Content Count

      26


  4. fluffi444

    fluffi444

    Members


    • Points

      2

    • Content Count

      4



Popular Content

Showing content with the highest reputation since 01/25/2020 in all areas

  1. 2 points
    iLO 2.73 released https://support.hpe.com/hpsc/swd/public/detail?swItemId=MTX_ba3437a6c8d843f39ab5cace06 UPGRADE REQUIREMENTS: OPTIONAL ***ATTENTION*** Note for ESXi users: If you are booted from the Embedded SD Card, it is strongly recommended that you reboot the server immediately after updating the iLO firmware. FIRMWARE DEPENDENCY: Hewlett Packard Enterprise recommends the following or greater versions of iLO utilities for best performance: - RESTful Interface Tool (iLOREST) 2.3 - HPQLOCFG v5.2 - Lights-Out XML Scripting Sample bundle 5.10.0 - HPONCFG Windows 5.3.0 - HPONCFG Linux 5.4.0 - LOCFG v5.10.0 - HPLOMIG 5.2.0 KNOWN ISSUES: - Fibre Channel Ports are displayed with degraded status if they are configured but not attached. FIXES: The following issues are resolved in this version: - Added fix for Embedded Remote Support in an IPv6-only environment. - Added fix for Embedded Remote Support data collection for systems with multiple Smart Array Controllers. Enhancements: - Suppress SNMP traps for NIC link up/link down events that occur during POST.
  2. 1 point
    Update: I recently uninstalled the HyperV-Role from my Server and now it shows 2 cores and 2 logical processors. We can therefore state that HyperV reduces the number of processors available to the host system. And this is displayed (actually correctly) in the task manager.
  3. 1 point
    Just a small update. After working fine for a year since my last post, my Gen8 showed the issue again after a reboot last weekend. I found this thread and resolved to follow my own advice! After first trying different RAM in different slots, I found that the issue was again with slot 2 only and the sticks of RAM were fine. I pulled the Xeon and reinstalled the original Celeron, and the system booted fine with RAM in both slots. So that seemed to narrow the issue to the Xeon. There was nothing obviously wrong with the Xeon, but I thought perhaps I saw some dullness on some of the contacts, so I cleaned both sides of the chip with thermal grease remover and purifier and reinstalled. It booted up straight away with both sticks of RAM. So problem solved again, hopefully this time it is permanent.
  4. 1 point
    Thanks, I was just checking, figured after all these years maybe someone had gone ahead and modified one. It's OK, does make a nice decoration up on my desk LOL Gib
  5. 1 point
    It is a regular coin style battery. Just get the part number, likely a CR2032 or CR2025, any electronics counter should have one.
  6. 1 point
    From someone who has worked the Enterprise IT business professionally for over 35 years, I have used these for trade shows and demo gear: https://www.skbcases.com/music/products/prodlist.php?o=&c=117&s=80 https://www.pelican.com/us/en/professional/rack-mount-cases/ https://www.gatorcases.com/rack-cases-accessories/ Height, depth and shock rating vary. Craigslist, eBay and band equipment shops are good sources for used. FYI, there are lots of robust travel racks that are wood based. I have used these many times: https://ospcases.com/collections/the-protector
  7. 1 point
    Take a look at this. LackRack - Eth0Wiki https://wiki.eth0.nl/index.php/LackRack This one mite also work for you On-Stage Stands WSR7500B Rack Cabinet - Black | Sweetwater https://www.sweetwater.com/store/detail/WSR7500B--on-stage-stands-wsr7500b-rack-cabinet-black
  8. 1 point
    Guys think its a bug in linux firefox,chromium because after booting into windows in Edge it shows everything as it should. so its FIXED! kinda..
  9. 1 point
    they go in the section titled 'lan ip setup'. the alias IP is an advanced setting you should not need. these settings will still work even if you get bridge mode set up on the wan side (link above). the range for dhcp goes further down in 'ip addressing values' section did the router come with a set of defaults initially? given the newness of this topic for you I am thinking if you can get bridge mode established a factory reset of YOUR router might give good values for pretty much everything or at least a start.
  10. 1 point
    There were some odd B120i issues if you only had 2GB of memory in the server, but obviously that is not a problem for you. Have you changed the coin style battery on the system board? Not sure if it is related, but it is something I would try.
  11. 1 point
    You should be able to set the ISP modem/router into bridge mode from the Broadband menu https://www.manualslib.com/manual/615384/Zyxel-Communications-Sbg3300-N-Series.html?page=52 Regards Matt
  12. 1 point
    ok so presumably the ISP modem/router will give your router an ip like 192.168.1.10. for the lan side you have many choices, for example... LAN port 10.0.1.1 subnet mask 255.255.255.0 assign lan devices a range from 10.0.1.50 thru 10.0.1.150 note this is but one of many you can use, just an example but should work. everyone has their favorites. If you have way more devices the 10. range lets you open up an even wider space but again, this is but an example. many people prefer to avoid 192.168.x.x just to be uncommon from the 'typical' ranges.
  13. 1 point
    since you can get into the ISP modem/router, does it perhaps have a bridge mode? if so you can turn that on and just set up your router as your 'main router' with dhcp that gives the lan clients DNS assignments of your choosing. if you can't set it in bridge mode, and you really want to keep the details of your lan hidden from the ISP, you can do double - nat - let the modem/router assign your router an IP like 192.168.0.5 and then your router would set the lan up to a different subnet like 192.168.1.x, whatever doesn't conflict. given recent revelations of backdoors in ISP modems I can see why some distrust would be appropriate. in both of the configurations I mention your router's WAN port would get an address via DHCP and provide all services to the lan thru dhcp but to avoid confusion the wan segment and lan segment need to be distinct. an in-between solution where you tell the isp router not to do dhcp is fairly tricky to work through. I"d suggest one of the two above.
  14. 1 point
    Impressive, although I'm very happy with my E3-1265l v2, not the most powerful, but can run with the standard heat sink without any issues.
  15. 1 point
    Thanks, I was looking on Swiss shopping sites only, and all they'd turn up were RDIMMs. Thanks for the links, I'm going to take a look at all those. I've pulled the trigger on buying the 5029T-C with a Xeon E-2278G and will order Ram+M.2 SSD+Raid card next month and possibly a 2.5" or two so I could achieve at least having a spare drive for RAID5 or even do RAID6. This site is so full of knowledge, especially from you Schoondoggy, thanks again! It'll be a continuation of my current Microserver Gen8, but with some additional VMs. Should be fun to setup.
  16. 1 point
    It is generally hard to find high capacity ECC UDIMM; Samsung appears to make them: https://www.samsung.com/semiconductor/dram/module/M391A4G43AB1-CVF/ SuperMicro is showing a part, but I do not see Micron offering it as a part: https://store.supermicro.com/memory/ddr4/32gb-ddr4-2666-mem-dr432l-cv02-eu26.html Nemix seems to be making parts: https://www.nemixram.com/nemix-ram-32gb-ddr4-2933-pc4-23400-2rx8-ecc-unbuffered-memory.html DataRam has one as well: http://www.dataram.com/__downloads/memory/memory-specs/DVM26E2T8_32G.pdf
  17. 1 point
    Hi all, I think it has been mentioned previously that pwm for the fans is BIOS controlled, this has some interesting implications, some theory, some tested and confirmed. Theory one is that fans one and four are pwm based on ram temp sensors, later testing may confirm this. Theory number two (very strong reason to believe this) is that fans two and three are pwm based on cpu temp sensing. CPU temp goes up and fans modulate accordingly. Currently getting temps of 22-37 degrees average at idle. Through testing we confirmed that whatever the case is pwm is always happening even when the system is off but power is still being supplied i.e. power button flashing orange. With a fan connected by an external 12v supply and being given a pwm signal from the fan connector pwm occurs. With this information in mind we got to work. RyanG8 has achieved pwm from a software side using CorsairLink and a Corsair Commander device and I had it in mind to do the same thing and may do so if theory one proves false but we wanted to play around and see what we could do. At first we tried to be way to clever and build an arduino device that read the pwm from the fan controller and then managed the pwm through some code to stay within the limitations of the AIOs fans.... to hard basket lol. In the end we kept it very simple and built a module that has its own 12v supply and a pwm only from the fan connector: The two front headers are being given a pwm signal from fan twos pwm and the third header is given a signal from the theoretical ram pwm or fan four. The two pin header is the pwm supply, 12v supply is from a sata power cable. The fan connector now looks like this under the heat shrink Running without issue and no erroneous errors during post. I will post again when more testing has occurred for the "ram" pwm which will go out to two 80mm fans mounted at the rear of the chassis.
  18. 1 point
  19. 1 point
    I still have the ISO my father saved years ago. Again, it's 808MB.
  20. 1 point
    SAS is backwards compatible with SATA. HPE uses SFF-8087 cables on the majority of their backplanes. It should just require moving the SFF-8087 cable from the system board to the P420. You should update the P420 to its latest firmware before using it.
  21. 1 point
    It's most likely a bug in the Memtest86+ SMP code as it can be consistently replicated where it fails at test 7. There's a PSA on it here. TL;DR warning. Memtest86+ hasn't been updated since that PSA. I don't know if it ever will. I'm running it with SMP disabled now on one of my N54Ls and it's so far hasn't failed or locked up after over 3 hrs. Memtest86 4.3.7 does not have a SMP mode. What I've noticed with my experience with the SMP bug is that it bites me if there's > 4 GB RAM in the system (X64 mode or 64-bit addressing). I remember the old days over dealing with the IOAPIC with SMP with socket 5/7 CPUs. My first SMP machine was a home build with a Tyan Tomcat S1564D (dual socket 7) built to run FreeBSD. I also was big into the Abit BP6s with dual Celeron 533s. If the server function could handle 768 MB, it ran very well; too well in fact which is why Intel lobotomized the IOAPIC in later Celerons making it SMP impotent. What I'm going to do after I've done the initial memory burn it is put FreeBSD 12.1 on them, run a make buildworld on them and run Memtest86+ 5.01 from the FreeBSD ports collection. The old *nix school of mem testing was to do a clean kernel or world build. If you got a core dump, it's most likely a hardware problem and memory is highly suspect. If I get the same results only from the FreeBSD Memtest86+ 5.01 port, I'll chalk it up to the PSA and buggy SMP code and drop FreeNAS on them but watch them closely. It's possible the FreeBSD SMP kernel code will circumvent the bug in which case there's a (more) reliable SMP Memtest86+; the FreeBSD port.
  22. 1 point
    Update; Lenovo Thinkserver TS430 memory does indeed work. An ECC UDIMM from a TS430 should work in your G7s without issue; a pair of 8 GB UDIMMs are correctly IDed in both the stock and modded BIOSes. Memtest86+ 5.01 will ID the memory for you.
  23. 1 point
    Final update on this with a successful outcome The new card came which also had the exact same chip/controller as the other two. I found this video which shows you how to copy your Windows install disk to the mSATA drive and install from there which worked for me. The issue was that I was using a USB ISO emulator hard drive device so it was seeing another drive even through I had disconnected the rest. I burned a DVD and booted into that which confirmed it was the device I was using to install Windows. I now have the original card installed and will try and send the OWC card back.
  24. 1 point
    As I do not plan to build any more of the SDM Rev4 brackets, I have decided to share the schematic. https://1drv.ms/u/s!Ao9ObX1BCpuXsstkeE7Jc5cyRmYTFQ?e=8ouDcP
  25. 1 point
    Thanks for guidance on this! Just a note on the solution I settled with in the end for anyone else who might come across this thread looking to do the same thing - Tried a couple different cheaper eSATA cards with no success having the drives in ICYCube detected in CentOS. Looked into driver issues and again with no luck. The USB connection did work fine, but I wasn't overly happy with that as a solution. I instead got this HBA (10Gtek Host Bus Adapter - https://www.amazon.co.uk/gp/product/B01M9GRAUM) with Mini SAS connection, and swapped the ICYCube for an alternative external enclosure with compatible Mini SAS connection (SilverStone SST-TS431S-V2 - https://www.amazon.co.uk/gp/product/B0771S45X3). On setup this worked perfectly, with the drives properly detected. I'm actually also a little happier with the SilverStone unit in general so far, it feels of higher build quality and fan also seems a little quieter. I also decided to move away from using the B120i RAID controller, and hardware raid in general, and am now using ZFS / RAIDZ. Thanks again for help as I was working through this!
  26. 1 point
    There is another update -> 2.3.0.1078 for A1 https://www.station-drivers.com/index.php?option=com_kunena&view=topic&catid=7&id=125&Itemid=858&lang=en Thanks to chinobino on station-drivers.com for his nice colletion. Tested 2.3.0.1078 - Works on my Digitus DS-30104-1 with Marvell 88SE9230
  27. 1 point
    Hey Guys. Looks like there is an new official FW for the Digitus DS-30104-1 (with Marvell 88SE9230) dated 2018-12-21 on thier webseite. A bit unexpected - Found it by coincidence 🙂 PACKAGE VERSION[0xFFFFFFFF]: 2.3.0.1066 AUTOLOAD VERSION[0x00000000]: 200019 LOADER VERSION[0x0000C000]: 21001008 BIOS VERSION[0x00020000]: 1.0.0.1027 FIRMWARE VERSION[0x00030000]: 2.3.0.1076 https://www.digitus.info/en/products/computer-accessories-and-components/computer-accessories/io-cards/ds-30104-1/ http://ftp.assmann.com/pub/DS-/DS-30104-1___4016032330240/DS-30104-1_firmware_mul_DS-30104-1 Firmware_20181221.zip Installed it on my DS-30104-1 - Works good!
  28. 1 point
    @Steven Bell almost a month later, but as I just did the update I would like to share this page with all for anyone that trying to find the latest moded bios! here it is: http://www.nathanielperez.us/blog/hp-proliant-n40l-bios-modification-guide In the comments, they say that support even N36L so you can safely install it!
This leaderboard is set to Indiana - Indianapolis/GMT-05:00
×
×
  • Create New...