Home server undertaking, NFS, PXE boot, 6 servers

Hey all, I have few old servers I like to play around with, and a main home server. My main one, which is a supermicro tower with 7 36 gb scsi320 drives, dual dual-core xeons. I have 5 poweredge 2600's that aren't in use due to my limitation of networking equipment, but I want to set up an NFS with pxe boot daisy chaining the poweredges to my supermicro. is this possible, and any advice?
i'm reading https://wiki.archlinux.org/index.php/Di … t_NFS_root and trying one tonight.

I think you need to have a subnetting system worked out, having a network of more networks and less hosts. So that you can "route" between each machine on either side of the network, effectively making a token ethernet.  But then it's going to get confusing if you actually try to complete the circle.  You could use iptables on each machine, one NIC would have a fixed address, serving dhcp to the PXE client machine.  So when you boot up a new PXE client, have it boot from the original server, maybe a copy of the same kernel in the /var/tftpboot directory on the main server and in the PXE client's same directory, but the root filesystem still accessed from the main server across the iptables and subnet.  Using pxelinux.cfg rules, you could have it set, to boot (possibly in fixed order) downline with the different configurations to get them to work.  I think you will be troubleshooting network bottlenecks, because no matter how you do it, it will eventually all go back to the main server.  Any errors in implementation will of course propogate across the network and cause problems.
Ditto, just get a hand-me-down switch from somewhere.
Last edited by nomorewindows (2012-06-23 16:35:33)

Similar Messages

  • Windows server 2012 R2 PXE boot is unworkable slow, I have this exact same set-up working in 2008 R2 What can I do?

    Hi there
    I'm trying to deploy a windows 7 image through Windows deployment services via PXE boot from a 2012 R2 server.
    Issue:    PXE boot is extremely slow, it takes up to more than 60 minutes for the device to download download the PXE boot
    Things I already tried to get this up and running:
    I've tried to change the TFTP block size via command prompt and via regedit
    I've changed the settings on the tab of the WDS role (go to WDS role -> properties on server -> Tab "tftp")
    Both actions resulted in PXE boot being even slower than it already was.
    To make sure this is not because of our environment I’ve set up the same configuration on a windows server 2008 R2, here PXE boot image is downloaded to the machine within 3 minutes.
    Both servers are set up through Hyper-V this is the configuration:
    2008 R2:
    Memory: 4096 MB
    1 Virtual processor
    IDE controller 2 hard drives
    Network adapter
    2012 R2:
    Memory: 4096 MB
    32 virtual processors
    SCSI controller 2 hard drives
    Network adapter
    I can’t imagine that PXE boot is not working because of the differences of the Hard drives controllers or because of the amount of the virtual processors.
    Windows server 2012 R2 seems to handle PXE boot TFTP differently in comparison to 2008 R2.
    Do you guys know what I can do/try to get this working?

    Hi Jacques Rodrigues,
    You can run Windows Deployment Services on Hyper-V virtual machines,
    that the performance will often be degraded, particularly during the Trivial File Transfer Protocol (TFTP) download phase. This phase is very resource-intensive and may
    fail if insufficient resources are available on your server running Hyper-V.
    If you are using multicast, Check if IGMP Snooping is enabled i.e. Routers that support multicasting. In particular, Internet Group Membership Protocol (IGMP) snooping should
    be enabled on all devices. This will cause your network hardware to forward multicast packets only to those devices that are requesting data. If IGMP snooping is turned off, multicast packets are treated as broadcast packets, and will be sent to every device
    in the subnet.
    The related KB:
    Windows Deployment Services Overview
    http://technet.microsoft.com/en-us/library/hh831764.aspx
    I’m glad to be of help to you!
    We
    are trying to better understand customer views on social support experience, so your participation in this
    interview project would be greatly appreciated if you have time.
    Thanks for helping make community forums a great place.

  • PXE boot issue with DHCP and SCCM server on different subnets

    I'm working with a client on the operating system deployment module of SCCM.
    Their network configuration currently has a single large subnet for client
    computers with a DHCP server on the same subnet. The SCCM subnet is
    configured on a seperate subnet with no DHCP server on the subnet. We want to
    configure client computers to be able to boot using the PXE client to deploy
    OS images to the machines but can not get PXE-boot to work correctly.
    Also, the client does not want to make changes to their network
    infrastructure routers or switches to remedy this problem. Are there settings
    on the DHCP or SCCM servers we can implement to make this work? If so, what
    needs to be installed or configured on each server. We currently already have
    WDS installed on the SCCM server and the SCCM server is configured as a PXE
    Service Point within SCCM. Both WDS and the PXE Service Point seem to be working fine.
    Any help would be appreciated.
    Thanks,
    Gary

    I am Brazilian,
    sorry for wrong english
    My DHCP is on linux,
    in my own structure VLANS
    The system center is on the network
    10.0.4.0/24
    The machines on the network 10.0.5.0/24
    The problem is that the machines that
    are not on the same network system center
    can not boot
      I tried configuring / etc/dhcp3/dhcpd.conf
    follows
    option vendor-class-identifier "PXEClient";
    option bootfile-name "\
    \ SMSBOOT \ \ x86 \
    \ wdsnbp.com";
    option tftp-server-name
    "10.0.4.101"; ---->
    IP server
    But it did not work, anyone know
    how to configure?

  • Does SCCM 2012 R2 supports PXE boot from Unix servers.

    Does SCCM 2012 R2 supports PXE boot from Unix DHCP servers.
    The client has DHCP on Unix servers. There are no Windows DHCP servers in the environment. Do we need Windows DHCP server for PXE boot.

    You can use any modern DHCP server in your environment.  DHCP is just used to help point the BIOS/Network Card to the right server, which in this case is your ConfigMgr Distribution Point with PXE Enabled.
    It is strongly recommended to use IP Helper Addresses instead of the DHCP options though.
    I hope that helps,
    Nash
    Nash Pherson, Senior Systems Consultant
    Now Micro -
    My Blog Posts
    If you found a bug or want the product to work differently,
    share your feedback.
    <-- If this post was helpful, please click the up arrow or propose as answer.

  • Problems PXE booting Gen 2 Virtual Machines after Upgrading to Windows Server 2012 R2

    My Current Setup: 
    I have two virtual machines set up with Hyper V on my Windows Server 2012 Server.
    VM1: Configured to boot from an ISO file and runs Clonezilla server. 
    VM2: Configured to PXE boot using a legacy network adapter, and with a passthrough 250Gb hard disk. 
    These two VMs are connected to an external Virtual Switch, which allows physical machines to PXE boot to the Clonezilla server. 
    For the past few months I have been using this setup for two purposes:
    1. To clone physical machines to the Clonezilla virtual server (as a backup).
    2. To restore those image to the second VM if I find that need to get access to the files on the original image.. 
    This has worked perfectly, except for the fact that the speed of the legacy network adapter on the second VM (which is required to PXE boot) is very slow. But I know this is because of the limitations in how the legacy adapter works... 
    Now my problem:
    I found out that Gen 2 VMs allow you to PXE boot without having use a legacy network adapter, allowing me to image back to the VM faster. So I upgraded the Server to Server 2012 R2 . I then created the same two virtual machines on the R2 server. I can still
    PXE boot the Gen 1 VM to the Clonezilla Server but I cannot get the Gen 2 VMs to PXE boot properly. They get an IP address from the Clonezilla Server but then just stop with the following screen. (note: I have disabled the Secure Boot). 
    Are Gen 2 VMs unable to pull down images from anything other than a WDS Server? 

    Hi J,
    >>Unfortunately WDS is not a solution for us as it is dependent on the PC being part of a domain
    If you would like to make the client not join to the domain, please check the box before “Do not join the client to a domain after an installation.”  In addition, make sure not to set the client to the domain in the unattend file and do not prestage
    the computer in AD. "
    It is quoted from following thread :
    https://social.technet.microsoft.com/Forums/windowsserver/en-US/1026c3a9-0a10-4a58-a48f-5391659a96c8/wds-set-unattend-file-for-workgroup?forum=winserversetup
    Best Regards
    Elton Ji
    We
    are trying to better understand customer views on social support experience, so your participation in this
    interview project would be greatly appreciated if you have time.
    Thanks for helping make community forums a great place.

  • HP MediaSmart Home Server EX 490 won't boot up and can't recover - running Windows XP

    Recently my Home Server launched successfully and I had it available on my new home network. I played a video on .api format via ethernet cable to my tv and I could see the server in my network directory.
    A few days ago I tried to access the server with a brand new MacBook through a wireless browse. I couldn't access as a guest so I decided to try to launch it with a direct ethernet connection from the old PC which I used to set up the server in 2010 (running Windows XP 2002 edition SP3).
    The server did not boot up properly and i have had a flashing red light since in the status light (far right) on the front of the server.
    I have attemped a recovery using the disks that came with the server and following instructions from HP support when I needed to recover it once before (while it was still under warrantly and supported by HP). The recovery process looked like it was working until it reached an "Unidentified Error" at Stage 4.
    Has anyone got any advice about how I could resintall the OS on the Server in this state while making best endeavour to not lose the data? Is the OS on the SYS drive and is that the lowest drive? Can I take the other drives out and rebuild the OS then put them back in and access the data? HP have had me chasing wild geese for hours over the phone and I'm ready to explode.

    Correction: Recovery process failure at Step 2 of 4

  • WDS Windows Server 2008 R2 network boot PXE x86 UEFI support?

    Hi,
    WDS Windows Server 2008 R2 network boot  PXE x86 UEFI support?
    -I have an SCCM 2012 SP1 site on Windows server 2008 R2.
    -I have an HP ElitePAD 900 with Atom x86 processor, and I have some problem to deploy an OSD image to the ElitePAD.
    Do I need WDS on an Server 2012 to get PXE x86 UEFI support with KB2817245?
    http://support.microsoft.com/kb/2817245/en-us
    /Per Sai
    /SaiTech

    Is this why I get this error with 2008R2?
    Windows failed to start.  A recent hardware or software change might be the cause.  To fix it... <blahblah>
    File:\Windows\System32\boot\winload.efi
    Status: 0xc0000359
    Thanks,
    Jason

  • PXE boot problem: guest VM DHCP request packets not able to reach DHCP server

    Hi Gurus,
      I'm wondering if anyone could help me with this problem. I wanted to install Linux on Oracle VMs using PXE. I set up a DHCP server and the OVM running RHEL6.4 box. The DHCP server worked fine since other PHYSICAL servers could get IPs from this DHCP server. However, DHCP requests from Oracle VMs was not able to reach the DHCP server. So I suspect this is a VM-specific issue.
    If I type in "dhcp net0" on gPXE prompt on the OVS machine(sappire), I can see the requests were being sent from the OVS server (sapphire):
    gPXE> dhcp net0
    DHCP (net0 00:21:f6:00:00:00) .............................................Connection time out (0x4c106035)
    Could not configure net0: Connection time out (0x4c106035)
    gPXE>
    [root@sapphire ~]# tcpdump -i any -n udp dst portrange 67-68
    tcpdump: WARNING: Promiscuous mode not supported on the "any" device
    tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
    listening on any, link-type LINUX_SLL (Linux cooked), capture size 96 bytes
    20:47:25.606400 IP 0.0.0.0.bootpc > 255.255.255.255.bootps: BOOTP/DHCP, Request from 00:21:f6:00:00:00, length: 387
    20:47:25.606549 IP 0.0.0.0.bootpc > 255.255.255.255.bootps: BOOTP/DHCP, Request from 00:21:f6:00:00:00, length: 387
    20:47:25.606559 IP 0.0.0.0.bootpc > 255.255.255.255.bootps: BOOTP/DHCP, Request from 00:21:f6:00:00:00, length: 387
    ^C
    12 packets captured
    14 packets received by filter
    0 packets dropped by kernel
    But if I snoop the same on the RHEL6.4 server running DHCP server and OVM, no request can be seen:
    [root@bluestone Desktop]# tcpdump -i any -n udp dst portrange 67-68
    tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
    listening on any, link-type LINUX_SLL (Linux cooked), capture size 65535 bytes
    ^C
    0 packets captured
    0 packets received by filter
    0 packets dropped by kernel
    OVS(sapphire) and OVM(bluestone) are located in the same subnet:
    [root@bluestone network-scripts]# ifconfig -a
    eth0      Link encap:Ethernet  HWaddr 00:14:22:72:7C:27 
              inet addr:192.168.2.48  Bcast:192.168.2.255  Mask:255.255.255.0
              inet6 addr: fe80::214:22ff:fe72:7c27/64 Scope:Link
              UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
              RX packets:106795 errors:0 dropped:0 overruns:0 frame:0
              TX packets:122056 errors:0 dropped:0 overruns:0 carrier:0
              collisions:0 txqueuelen:1000
              RX bytes:59173975 (56.4 MiB)  TX bytes:25362955 (24.1 MiB)
    [root@sapphire ~]# ifconfig -a
    10049df2fc Link encap:Ethernet  HWaddr 8A:C5:05:83:AF:C9 
              UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
              RX packets:80 errors:0 dropped:0 overruns:0 frame:0
              TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
              collisions:0 txqueuelen:0
              RX bytes:33200 (32.4 KiB)  TX bytes:0 (0.0 b)
    eth0      Link encap:Ethernet  HWaddr 00:1A:64:64:DA:64 
              inet addr:192.168.2.202  Bcast:192.168.2.255  Mask:255.255.255.0
              UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
              RX packets:37664 errors:0 dropped:0 overruns:0 frame:0
              TX packets:38939 errors:0 dropped:0 overruns:0 carrier:0
              collisions:0 txqueuelen:1000
              RX bytes:4537897 (4.3 MiB)  TX bytes:23127790 (22.0 MiB)
    eth0:0    Link encap:Ethernet  HWaddr 00:1A:64:64:DA:64 
              inet addr:192.168.2.212  Bcast:192.168.2.255  Mask:255.255.255.0
              UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
    [root@sapphire ~]# brctl show
    bridge name    bridge id        STP enabled    interfaces
    10049df2fc        8000.8ac50583afc9    yes        tap7.0
                                tap7.1
                                vif7.0
                                vif7.1
    I turned off iptables and SELinux on the DHCP server, the issue still remained.
    Any help will be highly appriciaited.
    Thanks in advance,
    Alex

    Hi,
    - Do you install Oracle VM Server (OVS) on an emulated environment like Oracle VM VirtualBox ? if yes so you can't do it.
    - Don't forget to configure the Virtual Machines Network and also to add this network to this Virtual Machine.
    I hope this can help you
    Best Regards

  • Server 2008 configured for PXE boot with UIU and GhostCast

    Right now, I am in the process of switching over from UIU 4, to UIU 5, and the implementation is completely different.  We used GhostCast to push images out before, but the new UIU uses WIM files to pull the drivers off the server.  I had a few
    questions about reconfiguring the new server.
    Currently, we have a server 2003 machine running ghost cast.  It pulls its images off of a server 2008 machine.  The server 2008 machine can be accessed remotely.  So my first question is, can this be configured to allow remote connections
    with the new setup?
    The things I have been reading have said that you need to enable DNS, AD, and DHCP, however if we connect this to a WAN, will it conflict with that?  Or can we configure only one of the ethernet ports to see this?  Sorry kind of new to this.
    I guess my main concern is, can we set up a separate domain for GhostCasting, while still being connected to an external domain?

    Hi,
    I’m not sure what you really want to do. For PXE boot, I think DHCP is always required just like Microsoft deployment solution WDS and MDT.
    The two products are third products, so I don’t know the detail information about how they work. Based on my experience, they should be based on broadcast or multicast.
    If so it could be a bad idea if you install it on a server facing the internet.
    I think you can also ask in the third party forums that provide the products.
    Thanks for understanding.

  • All 30 minutes my Macs (iMac/Snow Leopard, MacMini/Lion) are loosing the connection to the Windows Servers (2 Networks: Home with Home Server 2011 (Server 2008) and Office with Home Server V1 (Server 2003)).

    Hallo,
    Ich betreibe im Büro und zu Hause jeweils einen Microsoft Home Server. In beiden Netzwerken ist neben diversen Windows Rechnern auch ein Mac vorhanden.
    Die jeweilige "Konfiguration" ist wie folgt:
    - Apple MacMini (Late 2011, als HTPC) mit Lion per Wlan über AVM Fritzbox 7390 an Microsoft Windows Home Server 2011 (Windows Server 2008).
    - Apple iMac (Late 2010, Arbeitsplatzrechner) mit Snow Leopard per Lan über AVM Fritzbox 7270 an Microsoft Windows Home Server V1 (Windows Server 2003)
    Alle Betriebssysteme sind auf aktuellem Stand.
    Die Fritzboxen laufen mit der jeweils aktuellen Firmware
    Nun zu dem Problem:
    Beide Macs unterbrechen ca. alle 30 min den Kontakt zum Server. Die Verbindung wird zwar ca. 1-2 Minute später wieder aufgebaut, aber am MacMini (HTPC) macht sich dadurch bemerkbar, daß der Film oder die Musik für obige Zeitspanne stoppt und danach weiterläuft. Beim iMac (Arbeitsrechner) kann ein zuvor vom Server geladenes Dokument dann nicht mehr gespeichert oder gedruckt werden.
    Der MacMini hat dieses Phänomen direkt bei der Verbindung mit dem WHS gezeigt. Der iMac hat jedoch zuvor ca. 1,5 Jahre ohne Probleme funktioniert. Dieses Problem tritt bei dem iMac (Snow Leopard) erst seit ca. 6-9 Monate auf.
    Da unterschiedliche WHS und OS X Versionen betroffen sind und auch unterschiedliche Fritzboxen verwendet werden und scheint ein allgemeines Problem im Zusammenspiel Windows/OS X vorzuliegen.
    Bei dem iMac fällt der genaue Zeitpunkt der Verbindungsunterbrechung nicht auf.
    Bei dem MacMini läßt sich der Zeitpunkt dagegen relativ leicht bestimmen. Aus der Ereignisanzeige beim Windows Server ergibt sich leider kein eindeutiges Fehlerbild. Zu dem jeweiligen Zeitpunkt taucht aber relativ oft, aber nicht immer (!), in "Windows-Protokolle Sicherheit" der Fehler 4625 ("Fehler beim Anmelden eines Kontos") auf. Dies würde auf eine Rechteproblematik hindeuten. Das Konto des Mac kann aber alles was ihm auf dem WHS erlaubt ist.
    Beim Suchen nach einer Lösung (bisher leider erfolglos) bin ich u.a. auch über das Problem des Server Infrastructure License Services (silsvc) beim Windows Small Business Server 2011 Essentials gestolpert (auch alle 30 min!). Vielleicht ein Zusammenhang?
    Hello,
    I run the office and at home each a Microsoft Home Server. In both networks, in addition to various Windows computers, a Mac is available.
    The particular "configuration" is as follows:
    - Apple MacMini (Late 2011, as HTPC) with Lion via wifi via AVM Fritzbox 7390 to Microsoft Windows Home Server 2011 (Windows Server 2008).
    - Apple iMac (Late 2010, desktop PC) with Snow Leopard by Lan on AVM Fritzbox 7270 on Microsoft Windows Home Server V1 (Windows Server 2003)
    All operating systems on current conditions with all patches and updates.
    The Fritz boxes running with the latest firmware
    Now to the problem:
    Both Macs are loosing every 30 minutes the contact to the server. The connection is "re-"established about 1-2 minutes later. On MacMini (HTPC), the film or music is stopped for the above period. Thereafter, the film / music goes on. With the iMac (workstation) a document loaded from the server can no longer be saved or printed.
    The MacMini has shown this phenomenon directly in connection with the WHS. However, the iMac has been working about 1.5 years with no problems. This problem only occurs for about 6-9 months
    Because of different WHS and OS X versions are affected and also different Fritz boxes are used and it seems to be a general problem in the interplay Windows / OS X to be present.
    When the iMac is the exact date of disconnection does not occur.
    With the Mac Mini, the time can however be determined relatively easily. From the Event Viewer on Windows Server is unfortunately no clear error image. At the particular time but appeared relatively often, but not always (!) In Windows Security Logs the error 4625 (Failed to log on an account) on. This would indicate a rights issue.
    Searching for a solution (so far unsuccessfully, unfortunately), I stumbled over the problem of the Server Infrastructure License Services (silsvc) in Windows Small Business Server 2011 Essentials (every 30 min!). Maybe a connection?

    Hi, and thank you for your reply!
    My reason for not posting this in the server forum is that I suspect it is a clien issue. All my other clients connect happily to the server which leads me to believe the server is configured correctly.
    Only my the two MacBook Pro's are having problems.

  • PXE Boot a client to connect to a Remote Desktop Session Host/Terminal Server Farm via SCCM 2007

    Hi all,
    we've got a bunch of old desktops which are destined for disposal but i was wondering if we couldn't use them as disk-free clients to connect to our Remote Desktop farm. I did some digging and have managed to put together an iso which will boot
    the machine into WinPE and then connect to our farm. However, since it's an iso i have to use either a cd or usb for this to work. (This is pretty much what i did: http://www.iammacgyver.com/2011/02/easy-rdp-60-from-winpe-30-simple-boot.html) 
    What i want to do is to advertise the boot.wim to the client via SCCM so that it boots via PXE and then picks up the WinPE image and connect to the RD farm. I've imported the boot.wim file into SCCM under Boot Images and created an empty Task Sequence
    with the boot.wim as its Boot Image. I then advertised this Task Sequence to a collection which contains the old desktop clients. The machine PXE boots ok and loads up WinPE but rather than launch a remote desktop connection, it trundles through the normal
    WinPE stages and once it's finished 'Preparing Network Connections' it simply ends and the client reboots.
    Is what i'm trying actually possible?
    Can anyone tell me why the ISO works yet pushing the boot.wim out as a Boot Image doesn't? I know the boot.wim file (which is the same one present in the ISO) is ok since the client boots into WinPE fine when PXE booted so why doesn't the remote
    desktop session conenction launch? How is booting from the ISO different from booting from the PXE boot.wim file?
    Slightly long winded problem so many thanks in advance to anyone who can pass some info my way!

    It doesn't work because importing it into ConfigMgr injects binaries and the task sequence engine along with the process to launch the engine thus suppressing whatever you've got configured.
    Perhaps you could use a pre start command (pre-execution hook in 2007 parlance) to launch your RDP process though.
    Jason | http://blog.configmgrftw.com | @jasonsandys

  • Pxe boot sun x4170-m2 LSI raid 5 disks issues 2.10_u9 rel

    Hi Gurus,
    I am trying to image sun fire x4170-m2 series servers with Latest solaris 10.u9 patch set using Raid-5 disks setup with one hot-spare. Pxe boot works but i minboot not able to find raid-5 disks which is about 1.2 TB disks size. Where as if i used DVD to boot the servers it works properly. What i am missing. Here is my pxe entry and menu.list<mac-address> entry. I tried to insert driver LSI megraid driver with x86.miniroot, but it's already available into kernel/drv.
    -dhcpd.conf entry.
    host foobar { hardware ethernet 00:21:28:8E:3D:F2; fixed-address 10.228.150.14; next-server 140.87.100.21; filename "010021288E3DF2"; option host-name "foorbar.us.oracle.com";}
    menu.1st.<mac-address>
    cat /tftpboot/menu.lst.010021288E3DF2
    default=0
    timeout=30
    min_mem64 1024
    title Solaris_10 install_x86_64_2.10_u9
    kernel /I86PC.Solaris_10-9/multiboot kernel/unix - install - nowin -B console=i86pc,install_config=140.87.100.21:/export/jumpstart_x86_64_2.10_u9,sysid_config=140.87.100.21:/export/sysid_2.10_u9/10.228.144,install_media=152.68.248.67:/vol/local/x86/solaris/100/base/u9/mar2011,install_boot=140.87.100.21:/export/install_x86_64_2.10_u9_new/boot -v -m verbose
    module /I86PC.Solaris_10-9/x86.miniroot
    Please let me know if you have any solution for this issue.
    Thanks & Regards,
    -Sunil.

    Hi.
    Read thread:
    X4x70 (M1) with extremely slow disk I/O
    Try workaround the problem by adding the following line to /etc/system and then reboot:
    set idle_cpu_no_deep_c = 1
    Regards.

  • Starting a Home Server

    I am wanting to set up a home server (at my house) using a mac mini and an external hard drive array. I want to use the home server for a lot of different things including but not limited to:
    #1. Backing up computers to the server.
    #2. Using the server to store music and movies on for streaming and remote viewing.
    #3. Running my mail app on the server 24/7 so rules run instantly on all my devices.
    #4. To run a video surveillance software 24/7 using cameras around the house.
    #5. Remote view all my files, music and maybe videos if really needed (knowing that download speeds for music and movies might be slower).
    #6. Remotely backup my servers information to a external hard drive array at my dads house.
    #7. Allow my dad to backup his server to my external hard drive array the same way.
    I know this seams like a lot of work and money, but we are not interested in using a cloud service for this.
    I guess what I am looking for is some tips or help as i get started.
    Is there any applications that people would recommend to make the process easier.
    As far as a hard drive array, what do you suggest (8 Bay).
    Backup software?
    Will the mac mini be powerful enough to do all this?
    One last thing. I haven't boughten anything yet. I am wanting wait a few more months to see if apple releases a new mac min.
    Thanks.

    Play with OS X server on your laptop but don't plan on migrating the settings is my advice. (There are tools, but you may be better off learning how it works).  Ideally servers have static IP's, unique names & it makes more sense to me to start fresh when the server arrives.
    I'd test on a secondary installation of OSX so that the laptop can still be used after you stop testing OS X server.
    As for offsite backup. I'd also suggest you look at Arq - it backs up to Amazon S3 or Glacier.  Chronosync may be another option - it could sync between you and your dad once you have setup a secure way to access his server.
    Personally I would ignore Back To My Mac - you need a recent Mac to use it & you end up leaving your Apple ID scattered all over the place if you use it in public (ever noticed how the Apple ID gets into the iCloud settings after you login to one service). Other services like GoToMyPC, LogMeIn can allow remote access via a browser on many OS's.
    However since you are also wanting full access to the file shares you may be better suited with a VPN into the network. A VPN joins your device from the internet into the local network, so you can use whatever app or services that work on the local network. So you can use VNC/ARD iOS apps like 'Screens' to control Macs. You can also use the Screen sharing application on OS X to connect once the VPN connection is running into your network. Or VNC on Linux/ Windows machines.
    I don't think it is a good idea to setup the VPN for access from the Apple store or Library - you should really think about what level of access you want from these places - it's far safer to restrict to a few devices.
    The VPN is probably best run from your router - if you have one that is capable. The server can run it too as a service, however it's better if you are not 100% reliant on the server for the outside access - when it goes off the remote access would go down too.
    Seek out posts by MrHoffman in the server forums and there is some good advice concerning DNS on his site… you will want that working correctly if you plan to have outside access.
    http://labs.hoffmanlabs.com
    P.S. You seem to want to run Mail.app for your email rule processing - is that correct?
    That is a poor idea because the app needs to be running so that means the server must boot, auto login & launch Mail - that is a gaping hole for any server, it does require physical access, but it's not great for security should the server be stolen.
    "How did you hack it?" — "I just turned it on"

  • Users and permissions for a small home server

    Hello community,
    I have been using Linux on the desktop for many years now, but unfortunately my knowledge about servers is very limited, almost non-existent. Therefore my question is most probably equally well fitting here and into the newbie corner.
    I'm trying to set up a little home server which should be in charge of following tasks:
    - CUPS print server in the local network
    - access to shared files through NFS in the local network
    - backup (again over NFS)
    - an Owncloud server
    - maybe a mail server in the long run (NSA, paranoia, etc. )
    For now I have set up the print server, the NFS server and was working on the Owncloud installation, when Owncloud gave me some errors with users and permissions. So I was led to the idea of rethinking the users and permissions on server. So far there is only the root user who may do everything. This seems like a quite unsafe configuration. I'd like to make it safer. First, the printer, the backup and the locally shared files should be accessible from the local network only. SSH access should also be accessible locally only. The Owncloud file folder should be accessible from the internet, but of course only for the Owncloud users registered to the Owncloud server.
    What is the best way to set up users and permissions for such a set up?
    Thanks for any hints,
    PhotonX

    Hi, i think it depends who are you serving for, if you are just serving for a small office or home server or a big organization. The following quick thinking just came to me:
    I think cups set automatically a system  user of its own, and runs as it, so no trouble there. Cups also has the option to set users and it uses the system users as default, i think it depends in in how many printers/users your have in your server.Users that can manage cups are in the lp group. 
    For nfs every user should have their home, samba is also a good option if you have  windows computer in your network and it integrates better with graphical file  managers like nautilus in the clients side, but it is a hassle to configure.
    You should run the web server (owncloud ) as it own user, maybe you can manage to set something up for owncloud in the filesystem, but owncloud uses a database, and the users for owncloud are stored in there, and they are not system users.
    You can configure ssh for local use only enabling the corresponding subnets in your /etc/sshd.conf and optionally but recommended you can set a firewall and permissions. You can use iptables but i prefer ufw for simple setup.
    I think you should read the wiki:
    https://wiki.archlinux.org/index.php/users_and_groups
    and the other respective topics in the wiki.
    Also as an advice i know that arch linux is a great distribution, but you have to do more work to mantain a stable server. I would recommend debian or another more conservative distro, but of course it is your choice.
    Last edited by hydrosIII (2014-11-06 06:26:45)

  • PXE Boot and Task Sequence slowness

    I have been assured that our subnet 10.10.156.x is all 100MB.   Yet when I PXE up and run our CM2012 task sequence it takes over 1 hour
    just to resolve the task sequence.  Forget trying to actually do the OSD since that would take about 6 hours.
    Before I go back to networking I wanted to explain what I did to this forum to make sure I did things right.  
    Our CM2012 server is also our one and only MP.  It is also a DP for most of our small company.   Our CM2012 server is on subnet 10.10.141.x.  Yes, it has a static IP address.   
    Our 10.10.156.x subnet is in the same phsycal building as the 10.10.141.x subnet.  .141 is for servers and .156 is for workstations.
    We have a server that holds all of our DHCP scopes.   On this DHCP server I found the 10.10.156.x scope.  I added the following:
    066 Boot Server Host Name:  10.10.141.54 (which is our CM2012 server)
    067 Bootfile name:  \smsboot\x64\wdsnbp.com
    When I go to any workstations on the 10.10.156.x subnet I PXE boot up, it does see 10.10.141.54 and it launches the CM2012 OSD task wizard.   But as I said it is slower than glue and un-useable at this time.
    Did I setup this up right or do I need to do something else?  Or is our networking team full of it?  
    mqh7

    Ok, I did that.   I saw no errors and in fact it all looked fine.  It just took a long time.  so I talked to our top network tech.  He said the IP address means nothing.  It is all about what the wall Ports are connected too in the
    server/network room.   So this does appear to be an internal networking problem we need to solve.
    last question:  If you have 100MB ports is that considered slow for a CM2012 OSD?  our .WIM and all installed applications are close to 5GB of data.   So is 100MB good or should it all be Gigibit?
    mqh7

Maybe you are looking for

  • File path error in same screen.

    hey guys, I want to check my file path validation and display error in same screen itself. I used my GUI_UPLOAD method in START OF SELECTION. to read the file. If error it displays error in some other plain screen. I know AT selection screen will all

  • Query on SRM Shopping Cart ,plz help its very very urgent

    Hi Friends, I have a query on workflow. Users creates PO through SRM shopping cart and immedeately they receive a mail in there mail box.Below is the process flow of sending the mail to the user inbox from srm shopping cart. SRM Shopping Cart>Busines

  • What's with the "the signal and noise numbers" got to do with anything ?

    hi, i've been seeing posts about this but it's still not clear to me. can anyone help me understand this ?

  • [SOLVED] Bleachbit command line not working

    I am trying to incorporate bleachbit into a script "bleachbit -c". I have tried bleachbit from community and bleachbit-svn and bleachbit-cli from the aur. No matter what I do I get this error: info: starting BleachBit version 0.9.2 warning: cannot fi

  • IPad 16GB screen problem. Screen doesn't work as it should.

    Dear all, I would like to share with my issue. I got an IPad as a gift and started to use it... My happiness last one hour as after that on my eyes the screen started to show picture in green color and as if you watch TV with bad antenna. I would be