Debian not using all available ram

Discussion in 'Installation/Configuration' started by KenKnight, Jan 3, 2013.

  1. KenKnight

    KenKnight New Member

    Hi all,

    I setup a Debian Perfect server a couple of years ago and we just caught recently that it is not utilizing all the available ram. This server (Dell 1750) has 2 gigs of ram which is verified by the lshw results below.



    Code:
    ======== LSWH =========
    root@mail:/etc/default# lshw -C memory
      *-firmware
           description: BIOS
           vendor: Dell Computer Corporation
           physical id: 0
           version: A12 (09/29/2006)
           size: 64KiB
           capacity: 4032KiB
           capabilities: isa pci pnp upgrade shadowing escd cdboot bootselect edd int13floppytoshiba int13floppy360 int13floppy1200 int13floppy720 int9keyboard int14serial int10video acpi usb ls120boot biosbootspecification netboot
      *-cache:0
           description: L1 cache
           physical id: 700
           size: 8KiB
           capacity: 8KiB
           capabilities: internal write-back data
      *-cache:1
           description: L2 cache
           physical id: 701
           size: 512KiB
           capacity: 1MiB
           capabilities: internal varies unified
      *-cache:0
           description: L1 cache
           physical id: 703
           size: 8KiB
           capacity: 8KiB
           capabilities: internal write-back data
      *-cache:1
           description: L2 cache
           physical id: 704
           size: 512KiB
           capacity: 1MiB
           capabilities: internal varies unified
      *-memory
           description: System Memory
           physical id: 1000
           slot: System board or motherboard
           size: 2GiB
         *-bank:0
              description: DIMM DDR Synchronous 266 MHz (3.8 ns)
              physical id: 0
              slot: DIMM_1A
              size: 1GiB
              width: 64 bits
              clock: 266MHz (3.8ns)
         *-bank:1
              description: DIMM DDR Synchronous 266 MHz (3.8 ns)
              physical id: 1
              slot: DIMM_1B
              size: 1GiB
              width: 64 bits
              clock: 266MHz (3.8ns)
         *-bank:2
              description: DIMM DDR Synchronous 266 MHz (3.8 ns) [empty]
              physical id: 2
              slot: DIMM_2A
              width: 64 bits
              clock: 266MHz (3.8ns)
         *-bank:3
              description: DIMM DDR Synchronous 266 MHz (3.8 ns) [empty]
              physical id: 3
              slot: DIMM_2B
              width: 64 bits
              clock: 266MHz (3.8ns)
    

    free -m
    Code:
    root@mail:/etc/default# free -m
                 total       used       free     shared    buffers     cached
    Mem:           248        244          4          0          5         33
    -/+ buffers/cache:        205         42
    Swap:          484        175        309
    
    We've tried putting mem=2048 in the grub file but that isn't working.

    Code:
    GRUB_DEFAULT=0
    GRUB_TIMEOUT=5
    GRUB_DISTRIBUTOR=`lsb_release -i -s 2> /dev/null || echo Debian`
    GRUB_CMDLINE_LINUX_DEFAULT="quiet"
    #GRUB_CMDLINE_LINUX=""
    GRUB_CMDLINE_LINUX="var1 var2 mem=2048m"
    

    The uname info is:

    Code:
    root@mail:/etc/default# uname -a
    Linux mail.pboss.com 2.6.32-5-686 #1 SMP Tue Mar 8 21:36:00 UTC 2011 i686 GNU/Linux
    
    Is this a bigmem issue or does anyone have any other suggestions?

    Thanks,
    Ken
     
  2. falko

    falko Super Moderator Howtoforge Staff Moderator HowtoForge Supporter ISPConfig Developer

    What's the output of
    Code:
    cat /proc/meminfo
    ?
     
  3. KenKnight

    KenKnight New Member

    Thanks falko,

    Here's the info;

    Code:
    MemTotal:         254508 kB
    MemFree:            6916 kB
    Buffers:           14484 kB
    Cached:            40984 kB
    SwapCached:        45968 kB
    Active:            98024 kB
    Inactive:         121536 kB
    Active(anon):      73864 kB
    Inactive(anon):    90260 kB
    Active(file):      24160 kB
    Inactive(file):    31276 kB
    Unevictable:           0 kB
    Mlocked:               0 kB
    HighTotal:             0 kB
    HighFree:              0 kB
    LowTotal:         254508 kB
    LowFree:            6916 kB
    SwapTotal:        496632 kB
    SwapFree:         120488 kB
    Dirty:                 4 kB
    Writeback:             0 kB
    AnonPages:        119220 kB
    Mapped:             8300 kB
    Shmem:                20 kB
    Slab:              13832 kB
    SReclaimable:       5348 kB
    SUnreclaim:         8484 kB
    KernelStack:        2480 kB
    PageTables:         4340 kB
    NFS_Unstable:          0 kB
    Bounce:                0 kB
    WritebackTmp:          0 kB
    CommitLimit:      623884 kB
    Committed_AS:    1220324 kB
    VmallocTotal:     766108 kB
    VmallocUsed:        9064 kB
    VmallocChunk:     736976 kB
    HardwareCorrupted:     0 kB
    HugePages_Total:       0
    HugePages_Free:        0
    HugePages_Rsvd:        0
    HugePages_Surp:        0
    Hugepagesize:       4096 kB
    DirectMap4k:       12124 kB
    DirectMap4M:      249856 kB
    
    
     
  4. falko

    falko Super Moderator Howtoforge Staff Moderator HowtoForge Supporter ISPConfig Developer

    The server is in fact using 248MB of RAM. Please double-check if there are really 2GB of RAM, and if so, if they are ok (maybe you need to replace it).
     
  5. KenKnight

    KenKnight New Member

    Thanks falko, we'll do. There truly is 2GB recognized by the bios. You can see in my first post for this thread, the lswh result for bank 0 and bank 1 have 1GB chips in them.

    Cheers!
    Ken
     
  6. KenKnight

    KenKnight New Member

    I personally believe it has to be a configuration or kernel issue. We took another PE1750 (identical) down to the data center with 4 gigs in it and pull the hard drives from the original server and put them in the other server. It is now reading 2GB. This is certainly better than 256meg, but doesn't explain why the O/S doesn't see all the memory... regardless of whether it can use it.

    Thanks,
    Ken
     
  7. KenKnight

    KenKnight New Member

    MYSTERY SOLVED!!!!

    Hi all,

    Been a while since I'd worked on this, but I decided to rebuild the original server that was brought back from the data center (the one reporting the 256megs) and figured out what was causing it. Someone flipped the "OS Install Mode" to "ON" in the BIOS. This limited the memory to 256megs. What a PITA!

    Anyway, toggling that off and rebooting caused the OS to see all the memory.

    Hope this helps someone else.

    Cheers!
    Ken
     

Share This Page