HP ProLiant MicroServer Owners Club! (Attempting to sell here will result in bans)

Discussion in 'Storage & Backup' started by oli, May 10, 2011.

  1. GumbyNoTalent

    GumbyNoTalent Member

    Joined:
    Jan 8, 2003
    Messages:
    9,277
    Location:
    Briz Vegas
    Can you show out put of these commands
    Code:
    sudo hdparm -I /dev/sd[a-e]|egrep "Model|speed|Transport"
    Code:
    sudo  smartctl -a /dev/sda
    sudo  smartctl -a /dev/sdb
    sudo  smartctl -a /dev/sdc
    sudo  smartctl -a /dev/sdd
    sudo  smartctl -a /dev/sde
    if smartctl not installed, on ubuntu LTS it is part of smartmontools.
    Code:
    sudo apt-get install smartmontools
    The messages indicate a speed renegotiation, many people suggest SATA cable power issue, but you have already rulled that out how?

    Would also be useful to know;
    • what kernel / distro
    • what FS you running on the 4 drives not associated with the boot drive, which is having the issue.
    • what firmware / BIOS on MB
    • what firmware on drives
     
  2. Revenger

    Revenger Member

    Joined:
    Aug 7, 2002
    Messages:
    4,213
    Location:
    Armadale / Perth
    Running Debian 9 and I'm getting hard locks N36l
    When I was on CentOS 6 it ran pretty stable apart from software being outdated.

    When it locks not even the magic keys work.
    It'll work for a few days then hard lock randomly.
    The debug log shows this.
    Bootup after the lock shows some other stuff.

    I have no idea what to really do to get a stable server.

    It's on the modded bios if that's anything.

    Code:
    Jul 12 23:46:17 Server kernel: [    0.000000] DMI: HP ProLiant MicroServer, BIOS O41     10/01/2013
    Jul 12 23:46:17 Server kernel: [    0.000000] e820: update [mem 0x00000000-0x00000fff] usable ==> reserved
    Jul 12 23:46:17 Server kernel: [    0.000000] e820: remove [mem 0x000a0000-0x000fffff] usable
    Jul 12 23:46:17 Server kernel: [    0.000000] MTRR default type: uncachable
    Jul 12 23:46:17 Server kernel: [    0.000000] MTRR fixed ranges enabled:
    Jul 12 23:46:17 Server kernel: [    0.000000]   00000-9FFFF write-back
    Jul 12 23:46:17 Server kernel: [    0.000000]   A0000-EFFFF uncachable
    Jul 12 23:46:17 Server kernel: [    0.000000]   F0000-FFFFF write-protect
    Jul 12 23:46:17 Server kernel: [    0.000000] MTRR variable ranges enabled:
    Jul 12 23:46:17 Server kernel: [    0.000000]   0 base 000000000000 mask FFFF80000000 write-back
    Jul 12 23:46:17 Server kernel: [    0.000000]   1 base 000080000000 mask FFFFC0000000 write-back
    Jul 12 23:46:17 Server kernel: [    0.000000]   2 base 0000C0000000 mask FFFFE0000000 write-back
    Jul 12 23:46:17 Server kernel: [    0.000000]   3 disabled
    Jul 12 23:46:17 Server kernel: [    0.000000]   4 disabled
    Jul 12 23:46:17 Server kernel: [    0.000000]   5 disabled
    Jul 12 23:46:17 Server kernel: [    0.000000]   6 disabled
    Jul 12 23:46:17 Server kernel: [    0.000000]   7 disabled
    Jul 12 23:46:17 Server kernel: [    0.000000] TOM2: 0000000220000000 aka 8704M
    Jul 12 23:46:17 Server kernel: [    0.000000] e820: update [mem 0xe0000000-0xffffffff] usable ==> reserved
    Jul 12 23:46:17 Server kernel: [    0.000000] Base memory trampoline at [ffff9f0000098000] 98000 size 24576
    Jul 12 23:46:17 Server kernel: [    0.000000] BRK [0x16eb34000, 0x16eb34fff] PGTABLE
    Jul 12 23:46:17 Server kernel: [    0.000000] BRK [0x16eb35000, 0x16eb35fff] PGTABLE
    Jul 12 23:46:17 Server kernel: [    0.000000] BRK [0x16eb36000, 0x16eb36fff] PGTABLE
    Jul 12 23:46:17 Server kernel: [    0.000000] BRK [0x16eb37000, 0x16eb37fff] PGTABLE
    Jul 12 23:46:17 Server kernel: [    0.000000] BRK [0x16eb38000, 0x16eb38fff] PGTABLE
    Jul 12 23:46:17 Server kernel: [    0.000000] BRK [0x16eb39000, 0x16eb39fff] PGTABLE
    Jul 12 23:46:17 Server kernel: [    0.000000] BRK [0x16eb3a000, 0x16eb3afff] PGTABLE
    Jul 12 23:46:17 Server kernel: [    0.000000] ACPI: Local APIC address 0xfee00000
    Jul 12 23:46:17 Server kernel: [    0.000000] On node 0 totalpages: 2096941
    Jul 12 23:46:17 Server kernel: [    0.000000]   DMA zone: 64 pages used for memmap
    Jul 12 23:46:17 Server kernel: [    0.000000]   DMA zone: 21 pages reserved
    Jul 12 23:46:17 Server kernel: [    0.000000]   DMA zone: 3997 pages, LIFO batch:0
    Jul 12 23:46:17 Server kernel: [    0.000000]   DMA32 zone: 14271 pages used for memmap
    Jul 12 23:46:17 Server kernel: [    0.000000]   DMA32 zone: 913296 pages, LIFO batch:31
    Jul 12 23:46:17 Server kernel: [    0.000000]   Normal zone: 18432 pages used for memmap
    Jul 12 23:46:17 Server kernel: [    0.000000]   Normal zone: 1179648 pages, LIFO batch:31
    Jul 12 23:46:17 Server kernel: [    0.000000] ACPI: Local APIC address 0xfee00000
    Jul 12 23:46:17 Server kernel: [    0.000000] ACPI: IRQ0 used by override.
    Jul 12 23:46:17 Server kernel: [    0.000000] ACPI: IRQ9 used by override.
    Jul 12 23:46:17 Server kernel: [    0.000000] pcpu-alloc: s105304 r8192 d29864 u524288 alloc=1*2097152
    Jul 12 23:46:17 Server kernel: [    0.000000] pcpu-alloc: [0] 0 1 2 3 
     
    Last edited: Jul 13, 2019
  3. GumbyNoTalent

    GumbyNoTalent Member

    Joined:
    Jan 8, 2003
    Messages:
    9,277
    Location:
    Briz Vegas
    Code:
    sudo find / -name core -type f -atime -30
    ^ find all "core" files created in the last 30 days

    Once you have a core dump we can use gdp to see what the issue is.
     
  4. Revenger

    Revenger Member

    Joined:
    Aug 7, 2002
    Messages:
    4,213
    Location:
    Armadale / Perth
    this is what happened running that,

    Code:
    Last login: Wed Jun 26 12:46:56 2019 from 192.168.1.148
    root@Server:~# find / -name core -type f -atime -30
    root@Server:~#
     
  5. GumbyNoTalent

    GumbyNoTalent Member

    Joined:
    Jan 8, 2003
    Messages:
    9,277
    Location:
    Briz Vegas
    Good no core dumps, which is now pointing to hardware.
     
  6. Revenger

    Revenger Member

    Joined:
    Aug 7, 2002
    Messages:
    4,213
    Location:
    Armadale / Perth
    Quite strange as with Centos I had no issues like this at all.
    Dunno if its because I updated the modded bios.
    Was thinking Debian may be incompatiable and I should go back to Centos , but would like to get this running.
    My N36L has standard 8GB ram no ECC if that makes any diffrence.
    And all I have in it thats diffrent is a Remorte Access Card.
     
  7. sjp770

    sjp770 Member

    Joined:
    Apr 23, 2009
    Messages:
    1,722
    Location:
    Sticks, NSW
    Anyone got non ECC ram working in a Gen10? Saw a post that confirmed it should boot and work fine, grabbed 2x mta16atf2g64az-2g3e1 and it now reboots every now and then.
     
  8. Mace

    Mace Member

    Joined:
    Aug 16, 2001
    Messages:
    493
    Location:
    Canberra
    In my experience, if it loads your OS, the RAM is fine. The wrong type of RAM either won't even POST or won't load the OS.
    If so, then the reboots are caused by something else. Try removing anything else you have added, eg all the HDD bays are full might be overloading the power supply. Or check logs for an overheating CPU.
    If you want to confirm it isn't the RAM put in some ECC RAM. 2GB is enough to load any OS. If its stable, then it's the Non ECC RAM.
    I've got 16GB in mine, but you could also try reducing the RAM by half. I've found some systems have a max RAM limit.
     
  9. sjp770

    sjp770 Member

    Joined:
    Apr 23, 2009
    Messages:
    1,722
    Location:
    Sticks, NSW
    Agreed - it turns out it was most likely lack of drivers in the esxi install. Either for the system in general or the HP raid card that is certified for the gen10. In the meantime my friend ended up going with hyper-v and it seems stable on that ram with the raid controller.
     
    Mace likes this.
  10. Revenger

    Revenger Member

    Joined:
    Aug 7, 2002
    Messages:
    4,213
    Location:
    Armadale / Perth
    Just for info I'm keeping a eye on things for the next possible lockup.
    If I can get a pattern down I'll let you know.
    There was a kernel update recently I installed so let's see.

    Netdata shows warnings about ipv4 timeouts etc but I think it's just being overly sensitive.
     
  11. sjp770

    sjp770 Member

    Joined:
    Apr 23, 2009
    Messages:
    1,722
    Location:
    Sticks, NSW
    2x mta16atf2g64az-2g3e1 ram = 32Gb works fine in the Gen10 btw. Working for sometime now with windows 2016 install and hypervisored vms.
     
    Mace likes this.
  12. Mace

    Mace Member

    Joined:
    Aug 16, 2001
    Messages:
    493
    Location:
    Canberra
    Tell us about your Gen10 experience? I have a Gen8 and it's great to be able to upgrade the CPU. You're the first person posting about the Gen10 so far I think.
     
  13. sjp770

    sjp770 Member

    Joined:
    Apr 23, 2009
    Messages:
    1,722
    Location:
    Sticks, NSW
    Gen8 was nicer, I upgraded the CPU in that fine, and it was a lot cheaper. If we ever do the CPU I'll report back.
     
  14. Mace

    Mace Member

    Joined:
    Aug 16, 2001
    Messages:
    493
    Location:
    Canberra
    It was on the market when I upgraded from Gen7, but I read that the CPU was not upgradeable for the 10, so I bought the 8. It would be very interesting to find out now I could have upgraded the 10.
    Either way, even stock these units are great. So compact and so versatile for modding.
     
  15. TheWedgie

    TheWedgie Insert Custom Title Here

    Joined:
    Jun 16, 2002
    Messages:
    3,027
    Is anyone running a 2.5" SSD in their N36L's drive cage?
    Thinking about picking up a cheap SSD as a boot drive but obviously will need an adapter - obviously it will have to line up with the cage plugs, hoping someone has already found one that fits!
     
  16. Mace

    Mace Member

    Joined:
    Aug 16, 2001
    Messages:
    493
    Location:
    Canberra
  17. TheWedgie

    TheWedgie Insert Custom Title Here

    Joined:
    Jun 16, 2002
    Messages:
    3,027
  18. al'Thor

    al'Thor Member

    Joined:
    Jul 11, 2003
    Messages:
    353
    Location:
    Brisbane
    Thats SAS not SATA .

    I have a Samsung 860 evo 2TB in an Orico 1125SS adapter
    https://www.ebay.com.au/sch/i.html?....0+6Gbps+&_sacat=0&_sop=15&rt=nc&LH_PrefLoc=2
    In my n40L - works great but not tried as a boot drive

    for mine, I've been wondering PCIe x4 adapter to M.2 nvme pcie ssd , would "Blow the Sata into the weeds" if you not using the pcie x16 slot
    it won't make much difference being on PCIe 2.0
    https://www.techpowerup.com/review/pci-express-4-0-nvme-ssd-test-amd-x570-ryzen-3000/

    Anyone tried that
     
  19. Mace

    Mace Member

    Joined:
    Aug 16, 2001
    Messages:
    493
    Location:
    Canberra
    Well spotted!
    I'll have to doublecheck I haven't bought the wrong thing as that link was from the last lot I bought to have spare.
    It looks identical to the pair I have been using for ages.

    Edit.
    It seems the interface is the same. I'll test it tonight to be sure.
    https://www.programering.com/a/MjN4EDMwATU.html
    [​IMG]
     
    Last edited: Aug 15, 2019
    TheWedgie likes this.
  20. TheWedgie

    TheWedgie Insert Custom Title Here

    Joined:
    Jun 16, 2002
    Messages:
    3,027
    Good thing I'm slack and hadn't ordered them yet. Did notice the SAS in the item title but didn't look any closer.
     
    Mace likes this.

Share This Page

Advertisement: