[SOLVED] Boot fails with "ata4: SRST failed (errno = -16)"

I have desktop that has four disks. They have similar three partitions sitting on RAID.
/boot and swap are on RAID1
/ is on RAID5.
This worked well until a couple of weeks ago I noticed I couldn't connect to it via SSH. I went onsite and the desktop was humming but I couldn't get picture on the screen and alt + sysrq + REISUB did nothing. I shut it down by pressing power button and started it again. It gave me following errors:
ata4: SRST failed (errno = -16)
After a while there was
reset failed, give up
I restarted the machine and went to BIOS. There was only two of four disks visible. I shut the computer down, removed the power cord and then plugged it back in. This time BIOS showed all the disks there. I tried to boot up Arch and got the following.
Booting the kernel.
running early hook [udev]
running hook [udev]
Triggering uevents
mounting '/dev/md3' on real root # md3 is the RAID 5 device containing root
mount: you must specify the filesystem type
You are now being dropped into an emergency shell.
sh: can't access tty; job control turned off
After that I tried to chroot from USB installation disk. First I check the RAID devices and see that in md1 (/boot) all seems to be ok. In md2 (swap) half of the partitions are missing. I'm not sure how to interpret the last one.
cat /proc/mdstat
Personalities : [raid1]
md1: active raid1 sdd1[3] sdb1[1] sda1[0] sdc1[2]
1048564 block super 1.0 [4/4] [UUUU]
md2: active raid1 sdb2[1] sda2[0]
2096116 block super 1.2 [4/2] [UU__]
md3: inactive sdb3[1](S) sda3[0](S) sdd3[4](S) sdc3[2](S)
608169984 block super 1.2
unused devices: <none>
I remove those devices and try to recreate them. md1 says it starts with four drives. md2 starts with 2 driver out of 4. And then the last one.
mdadm --assemble /dev/md3 /dev/sd[a-d]3
mdadm: /dev/md3 assembled from 2 drives - not enough to start the array.
Has anyone any idea where did the two partitions go? I ran SMART-tests to all four drives in long mode and all of them passed the tests.
I use GPT partitioning table, UEFI and Syslinux if it matters.
Last edited by Tha-Fox (2013-02-23 01:35:05)

I just booted to chroot environment and checked that all the partitions show up all right. Below is output of one disk and all the four disks give similar info.
gdisk -l /dev/sdd
GPT fdisk (gdisk) version 0.8.6
Partition table scan:
MBR: protective
BSD: not present
APM: not present
GPT: present
Found valid GPT with protective MBR; using GPT.
Disk /dev/sdd: 312581808 sectors, 149.1 GiB
Logical sector size: 512 bytes
Disk identifier (GUID): 8B9EBB89-366E-4C81-8916-F61CB08273DB
Partition table holds up to 128 entries
First usable sector is 34, last usable sector is 312581774
Partitions wil be aligned on 2-sector boundaries
Total free space is 2203245 sectors (1.1 GiB)
Number Start (sector) End (sector) Size Code Name
1 34 2097185 1024.0 MiB FD00 Linux RAID
2 2097186 6291489 2.0 GiB FD00 Linux RAID
3 6291490 310378529 145.0 GiB FD00 Linux RAID

Similar Messages

  • [SOLVED] Booting Arch off USB HDD fails - can't mount real root device

    I've installed Arch on the internal HDD (sda) of my laptop, and I attempted to install a second copy of Arch on partition sdb3 of an external USB HDD.  sdb3 is a primary partition that I had prepared beforehand with gparted and outfitted with the reiserfs filesystem.  I skipped section 3.2 (manually prepare hard drives) of the Arch installer.  There was no problem with section 3.3 (manually configure filesystems and mountpoints); I put all of / in sdb3.  Installing packages and configuring the system also went without a hitch.
    I wanted to boot this new install of Arch via the boot menu of GRUB that was installed in the MBR of sda with my primary install of Arch.  On the new install, I first did go through section 7 (install bootloader), thinking I would want to install GRUB in the root sector of partition sdb3 (not in the MBR of sdb where the GRUB of MintKDE resides, a boot menu that I want to keep) but that failed, so I then picked "no bootloader to be installed".  With that I exited the new install, since it appeared to have gone to completion.
    I then copied the appropriate stanza in the /boot/grub/menu.lst of this new Arch install to the /boot/grub/menu.lst of my primary Arch install on the internal HDD of my laptop.
    This stanza reads as follows:
    # (3) Arch Linux on WDPP60 blue (sdb3)
    title Arch Linux on WDPP60 blue (sdb3)
    root (hd1,2)
    kernel /boot/vmlinuz26 root=/dev/disk/by-uuid/191cc027-43ce-443b-8846-f265d60555ec ro
    initrd /boot/kernel26.img
    However, I'm not able to boot the new Arch install when I pick this entry from the GRUB menu list.  The new install starts booting, finds sda but not sdb and stops with the following errors, dropping me to a recovery shell:
    Root device /dev/disk/by-uuid... doesn't exist. Attempting to create it.
    ERROR: Unable to determine major/minor number of root device '/dev/disk/by-uuid...'
    mounting /dev/disk/by-uuid... on /new_root failed: No such file ...
    ERROR: Failed to mount the real root device.
    Running "mount -t reiserfs /dev/sdb3 /new_root" at the ramfs prompt also fails.
    I've tried a number of edits of the menu.lst stanza from the grub> command prompt, such as adding rootfstype=reiserfs, replacing root=/dev/disk/by-uuid..etc. with root=/dev/sdb3 (booting then fails with "/dev/sdb3  No such file"), using "initrd /boot/kernel26-fallback.img" etc., but booting always fails at the same stage with equivalent error messages.
    I have no trouble booting half a dozen other Linux/BSD distros that reside on two different external USB HDD's by the same method, i.e. choosing an entry from the GRUB boot menu of my primary Arch install, after I placed the appropriate stanza in its /boot/grub/menu.lst.
    How can I get this new Arch install to boot?
    Last edited by RobF (2010-05-17 23:03:27)

    Yes, this was in fact the solution.  I stumbled upon it in this thread:
    http://bbs.archlinux.org/viewtopic.php?id=65844
    When one wants to boot from a USB mass storage device, the initrd needs to include the modules for accessing USB devices, and in Arch this is accomplished by adding "usb" to the HOOKS array.  I simply reinstalled Arch to the USB HDD and in the system configuration step now edited /etc/mkinitcpio.conf by adding the "usb" item.  Adding "rootdelay=10" and/or "rootfstype=reiserfs" to the kernel line in the appropriate stanza in Arch's /boot/grub/menu.lst wasn't necessary in this case.  I left the stanza exactly as I posted it above.
    Thanks for your input.

  • Service Clients failing with "The request failed with HTTP status 401: Unauthorized"

    Hello,
    We have implemented a solution using the SSRS web service clients as produced by the WSDLs.  We have deployed the application on one server, and via the code we are calling the SSRS on another server (i.e., not using ReportViewer, etc.).  We do
    know that the user signed in has permissions since from the same staging server the Report Manager and Reports can both be accessed and viewed.  However, it's only when we run the application and attempt to execute a report on the remote SSRS server that
    we get the following error:
    The request failed with HTTP status 401: Unauthorized. at System.Web.Services.Protocols.SoapHttpClientProtocol.ReadResponse(SoapClientMessage message, WebResponse response, Stream responseStream, Boolean asyncCall) at System.Web.Services.Protocols.SoapHttpClientProtocol.Invoke(String
    methodName, Object[] parameters) at AMC.AssetTracker.Reporting.ReportExecution.ReportExecutionService.LoadReport(String Report, String HistoryID) at AMC.AssetTracker.Reporting.Report.GetReportByteArray() at AMC.AssetTracker.Reporting.Report.get_ReportStream()
    at AMC.AssetTracker.Reporting.Report.get_GetWebUIDisplay() at AMC.AssetTracker.UserControls.AssetTrackerMain.btnGenerate_Click(Object sender, EventArgs e)
    The credentials we have tried are the DefaultCredentials.  One question I have is how do you use custom credentials if neither the DefaultCredentials or the DefaultNetworkCredentials work?  Also, this service call does have to go through an ISA
    server before reaching the SSRS server.
    Any ideas?
    Thanks.

    It sounds like Kerberos authentication is needed in your situation since it's not on the same box.
    There is a one-hop limit with NTLM authentication.
    For more info please see the link below:
    http://social.msdn.microsoft.com/Forums/en-US/sqlreportingservices/thread/452e9627-cd8e-4709-bdd0-fbafcf9fd719
    Hope this helps!
    Thanks, Michael Mei

  • RAC Install fails with "OCR upgrade failed with (-1073741819) on 10.2.0

    I am trying to install RAC 10.2.0 with two Server 2003 R2 nodes connected to a fiber channel SAN. I have set up the shared storage, created the extended partitions, and setup the unformatted no drive letter volumes. Both servers see the unformatted volumes (and no others do). The preinstallation checks work fine and the installations works fine until the first configuration assistant runs. That would be the "Oracle Clusterware Configuration Assistant". It tries to run crssetup.config.bat and fails on step 3 (configuring OCR Repository) with
    ocr upgrade failed with (-1073741819)
    If I uninstall clusterware and try again it will not show the OCR and Voting disks as available disks until I delete the volumes and partitions and recreate them. I have run the precheck batch files and they show everything as good as well. When I define the OCR and Voting drives in the clusterware setup I do not select format or assign drive letter. Every time I try to do the install I get the exact same error.
    Is there anything I am missing or some trick or patch I may not have done?

    Take a look at this notes!!
    Note 341214.1 - How To clean up after a Failed Oracle Clusterware Installation on Windows
    Note 388730.1 - Oracle RAC Clusterware Installation on Windows Commonly Missed / Misunderstood Prerequisites
    Rgrs,
    Paulo.

  • Installation of HANA SP7 failed with hdbnsutil call failed.

    Hello,
    I am trying to install HANA to play around with it a little.  I tried to install using the tool hdblcmgui since the old method was deprecated and during the Installing SAP HANA database step, the install fails with the following error message.
    18:53.727 - ERR :     Installation of SAP HANA Database failed
    17:18:53.727 - INFO:       Installation failed
    17:18:53.727 - INFO:         error installing
    17:18:53.727 - INFO:           Cannot create Instance
    17:18:53.727 - INFO:             Cannot create instance
    17:18:53.727 - INFO:               Cannot create SecureStore
    17:18:53.727 - INFO:                 Starting external program /usr/sap/RSW/HDB00/exe/hdbnsutil
    17:18:53.727 - INFO:                   Command line is: /usr/sap/RSW/HDB00/exe/hdbnsutil -createSecureStore
    17:18:53.727 - INFO:                 Cannot execute program /usr/sap/RSW/HDB00/exe/hdbnsutil: /usr/sap/RSW/HDB00/exe/hdbnsutil: No such file or directory
    17:18:53.727 - INFO:                 hdbnsutil call failed
    When I look in the directory /usr/sap/RSW/HDB00/exe
    The files hdbnsutils, hdbparam and possibly others were never created.
    hana-db-001:/usr/sap/RSW/HDB00/exe # ls -l
    total 16
    dr-xr-xr-x 7 rswadm sapsys 4096 May  8 15:45 Python
    dr-xr-xr-x 2 rswadm sapsys 4096 May  8 15:45 config
    lrwxrwxrwx 1 rswadm sapsys   41 May  8 15:44 lexicon -> ../../../global/hdb/custom/config/lexicon
    -r--r--r-- 1 rswadm sapsys  549 Mar  7 09:41 manifest
    dr-xr-xr-x 7 rswadm sapsys 4096 May  8 15:45 python_support
    Does anyone have any ideas?
    Message was edited by: Tom Flanagan

    Hi Johan,
    Yes, the downloaded file was corrupted.  I checked the file against the checksum and noticed that it didn't match.  So I downloaded everything again and this time checked all files and it installed without any issue.
    Good luck

  • Client installation fails with error GetSSLCertificateContext failed with error 0x87d00281

    Hi
    Site is SCCM 2012 R2
    I am pushing SCCM client manually to a Windows Server 2012 client and it is failing with below errors
    GetSSLCertificateContext failed with error 0x87d00281
    ccmsetup 02/05/2015 12:04:11 AM
    6428 (0x191C)
    GetHttpRequestObjects failed for verb: 'GET', url: 'HTTPS://SCCMserver.GROUP.COM/CCM_Client/ccmsetup.cab'
    ccmsetup 02/05/2015 12:04:11 AM
    6428 (0x191C)
    DownloadFileByWinHTTP failed with error 0x87d00281
    ccmsetup 02/05/2015 12:04:11 AM
    6428 (0x191C)
    CcmSetup failed with error code 0x87d00281 ccmsetup
    02/05/2015 12:04:11 AM 6476 (0x194C)
    Please advice..

    0x87d00281 = "No certificate matching criteria specified"
    Have you deployed proper client auth certs to the client?
    Can you please post the entire relevant snippet of the log file? Starting with the first error code encountered is meaningless because now we can't see what happened before which is often the root cause of the issue culminating in the error message.
    Have you deployed the proper server auth certs to the MP and DP?
    Jason | http://blog.configmgrftw.com | @jasonsandys

  • Scheduling a Deski report fails with error message: "Failed to get property

    Hi,
    We have created few Desktop Intelligence report. Now when these report are schedules to be refreshed weekly (on Sunday), it fails with the following error message:
    "Failed to Get Property"
    Have tried to refresh the reports manually in Full Client too but the report fails there too.
    Have checked for the connection information. The servers are up and running.
    Have tried to re-create the connection and even tried to re-import and export the Universe.
    If we rename the report and save it and export, it works fine. Can not try this workaround as have more that 500 reports.
    Please suggest.
    regards,

    Hi Maria,
    usually the message "Failed to Get Property" is followed by an item name such as SI_FILE, or SI_VARIABLE, or SI_USE_ORIGINALDS, and so on.
    Could you please specify as much information coming from the error message as you can?
    Furthermore, is this issue happening with any Desktop Intelligence reports you schedule?
    Are the reports using the same Universe?
    Which DB are you using?
    Regards,
    Samanta F.

  • [SOLVED] Boot Resolution with Intel Haswell GPU and KMS

    Dear All,
         This thread is a follow up of this one: https://bbs.archlinux.org/viewtopic.php?id=189562. I created a new thread as I think the focus of the problem has shifted and the old title might be misleading and lead to helpful people ignoring the thread. I hope this is ok. Please correct me if I did not behave correctly.
    I think most of the "flashes" I see come from X switching from native resolution (3200x1800) which is used when starting, to 1920x1080, which I set in the configuration file.
    So my question is: how can I have the 1920x1080 resolution at boot (i.e. during early boot, using the KMS and including the i915 driver in mkinitcipo?):
    I tried:
    1) Creating the configuration file in xorg.conf.d and including it into the mkinicpio
    2) Setting the GRUB_GFX_MODE
    3) Using the EDID method (with built-in resolution 1920x1080) as described here: https://wiki.archlinux.org/index.php/ke … s_and_EDID
    4) Using uvesafb as described in the WIKI: https://wiki.archlinux.org/index.php/uvesafb
    Nothing worked. X starts during boot always with the same native resolution, the switches to 1920x1080 when GDM starts. Can you please suggest a method to set the boot resolution? It is for sure technically possible as Ubuntu does it when installed on the same hardware.. I think I just haven't found a way...
    Valerio
    Last edited by valmar (2014-11-10 15:15:07)

    Ok, answering my own question, it would appear that adding video=1920x1080 does the trick. However, I still see a lot of flashing (I think that people call it flickering)

  • Flash builder 4.5.1 ios build fails with error map failed

    I have built several test app with flash builder 4.5 and now 4.5.1.  I am working on an app for ipad, but when I run a test build it runs for several minutes as if it is going to create the ipa, but then error out and displays the following error message:
    message title: Problem occured
    'Launching [name of configuration] has encountered a problem.
    Error occurred while packaging the application:
    Map failed
    that's it, no other reference.  I have searched for any reference to this problem and haven't been able to find one.  When I first packaged a much simpler version of the app, it created an ipa just fine.
    Can someone help me.
    Thanks

    When I was try to build a debug iOS version, I got this error, too.(release version is ok)
    For the project I was use Flex SDK 4.5.1 and AIR 3 SDK.
    blew is the log, I copyed from {flash builder workspace}/.metadata/.log
    !STACK 0
    java.lang.Exception
              at com.adobe.flexbuilder.project.internal.FlexProjectCore.createErrorStatus(FlexProjectCore. java:1010)
              at com.adobe.flexbuilder.util.logging.GlobalLogImpl.log(GlobalLogImpl.java:66)
              at com.adobe.flexbuilder.util.logging.GlobalLog.log(GlobalLog.java:52)
              at com.adobe.flexide.multiplatform.ios.packaging.IPAPackager.create(IPAPackager.java:220)
              at com.adobe.flexide.multiplatform.ios.launching.IOSOnDeviceLaunchHandler.doPackage(IOSOnDev iceLaunchHandler.java:283)
              at com.adobe.flexide.multiplatform.ios.launching.IOSOnDeviceLaunchHandler.launch(IOSOnDevice LaunchHandler.java:181)
              at com.adobe.flexide.launching.multiplatform.MultiPlatformLaunchDelegate.launch(MultiPlatfor mLaunchDelegate.java:180)
              at com.adobe.flexide.launching.AbstractFlexLaunchDelegate.launch(AbstractFlexLaunchDelegate. java:238)
              at com.adobe.flexide.launching.AbstractFlexLaunchDelegate.launch(AbstractFlexLaunchDelegate. java:132)
              at org.eclipse.debug.internal.core.LaunchConfiguration.launch(LaunchConfiguration.java:853)
              at org.eclipse.debug.internal.core.LaunchConfiguration.launch(LaunchConfiguration.java:702)
              at org.eclipse.debug.internal.ui.DebugUIPlugin.buildAndLaunch(DebugUIPlugin.java:923)
              at org.eclipse.debug.internal.ui.DebugUIPlugin$8.run(DebugUIPlugin.java:1126)
              at org.eclipse.core.internal.jobs.Worker.run(Worker.java:54)
    !ENTRY com.adobe.flexbuilder.project 4 43 2011-11-22 22:10:54.816
    !MESSAGE Error occurred while packaging the application:
    Map failed
    even I try to change sdk with FLEX 4.5 and AIR 2.6, the error was still here.
    And I try to build debug project in Flash Pro. It was OK.
    So I think it was an flashbuilder bug.
    My project only 137Megabytes

  • Crypt acquire context failed with 0x8009000f after upgrade from sms 2003 to SCCM 2007 SP1

    I've upgraded clients from SMS 2003 SP3 to SCCM 2007 SP1.
    In the SCCM console some ressources are showing as not being client.
    If I look on the machine, I can see that client is installed but the client cannot communicate with the management point.
    CCMEXEC.log shows the following error message :
    Crypt acquire context failed with 0x8009000f.
    Failed to create certificate 8009000f CcmExec 21/04/2009 9:49:10 1948 (0x079C)
    CCMDoCertificateMaintenance() failed (0x8009000f).
    So, the issue is related to certificate
    If I go to C:\documents and settings\all users\Application Data\Microsoft\Crypto\RSA\MachineKeys and delete the file begining by  19c5cf... and then restart the client the problem is solved.
    Can someone explain what is happening ?
    What can I do to prevent this problem
    Kind regards

    In Server 2008 R2 SP1 (when Pushing Out SCCM 2012 clients)
    I just added another account to the security (local Administrators) for the key which starts with 19c5cf...
    and restarted the SMS AGENT HOST SERVICE.
    Thats also resolved the problem. (NETWORK was the only security account on the key file).

  • 12c agent install fails with error - The plug-in configuration for the oracle.sysman.oh monitoring plug-in may have failed

    Hi,
    I am trying to install 12c agent on windows 7 64 bit server by following m.note   .
    It is failing with Agent Configuration failed, please see below error message.
    INFO: length of temp is2
    INFO: Return value:C:\/Oracle/12.1.0.3.0_AgentCore_233/core
    INFO: ** Agent Port Check completed successfully.**
    INFO: ERROR: The Management Agent configuration failed. The plug-in configuration for the oracle.sysman.oh monitoring plug-in may have failed, or this plug-in may not be present in the Management Agent software. Ensure that the Management Agent software has the oracle.sysman.oh monitoring plug-in, if not then retry the operation. If the agent software has the oracle.sysman.oh monitoring plug-in, view the plug-in deployment log C:\Oracle\12.1.0.3.0_AgentCore_233\core\install\logs to check if the plug-in configuration for the oracle.sysman.oh monitoring plug-in failed.
    INFO:
    INFO: perform - mode finished for action: configure
    INFO:
    INFO: You can see the log file: C:\Oracle\12.1.0.3.0_AgentCore_233\core\12.1.0.3.0\cfgtoollogs\oui\configActions2014-10-03_08-48-15-AM.log
    INFO:
    INFO: C:\Oracle\12.1.0.3.0_AgentCore_233>exit /b 3
    INFO: Plugin homes:
    INFO: Plugin homes:
    INFO: C:\Oracle\12.1.0.3.0_AgentCore_233\core\12.1.0.3.0\oui\bin\runConfig.bat ORACLE_HOME=C:\Oracle\12.1.0.3.0_AgentCore_233\core\12.1.0.3.0 RESPONSE_FILE=C:\Oracle\12.1.0.3.0_AgentCore_233\core\12.1.0.3.0\agent.rsp ACTION=configure MODE=perform COMPONENT_XML={oracle.sysman.top.agent.11_1_0_1_0.xml} RERUN=true completed with status=3
    SEVERE: ERROR: Agent Configuration Failed
    Thanks,

    Looks like , J2EE is having a problem connecting to the DB.Pls. chk. the following
    -If you have configured the loop-back adapter , if installing on a local system & updated the /etc/host file with your ip address and host name.
    or if the system is part of some n/w group and has ip address assigned ?
    -Jdk 1.4 version is installed in your system
    -Enough free Disk space available in the system
    -How much RAM your system has which windows version you are using  ?
    -Is the DB is coming up properly ? its error related to this. check the logs.
    -If the ports used by sap install / run used by other services running on the system ? & user have admin. privileges on the system .
    Pls. update with your findings on above same , for the step next ...
    Thanks ,
    Uppal

  • Sccm 2012 agent not installing with error CcmSetup failed with error code 0x80041013

    We have sccm 2012 r2 enviroment , when we deploy client agent to one DC , client agent is not getting install with below error .. I can able to install agent to all the machines except one DC.
    CcmGetOSVersion failed with 0x80041013
    CcmSetup failed with error code 0x80041013
    Any idea ?

    The error code 0x80041013 translates to:
    Provider load failure
    Source: Windows Management (WMI)
    Like Briij suggested, you should try to fix the WMI repository.
    Regards,
    Nickolaj Andersen | www.scconfigmgr.com | @Nickolaja

  • Setup fails with outlook 2007

    Hi,
    Is iCloud 3 not compatable with outlook 2007? I am running windows 7 32 bit with outlook 2007. My contacts stopped syncing with my iphone. I signed out of iCloud and back in. Set up now fails with "your setup failed to start due to an unexpected error" then setup crashes. I have removed iCloud, Outlook and removed the .me pst file from the local folder. Reinstalled Outlook, icloud and then tried setup again and saddly the same error comes up.
    Is there anyway to get icloud 2?
    Regards,
    Jonathan

    Same problem, email redirector works fine. It connects to mass storage device everything works except sync. Did you find an answer to your problem?

  • [Solved] Pacman errors: "call to execv failed", system not booting

    Recent upgrade (yesterday) - problems went like this:
    1) pacman -Syu attempted "filesystem" upgrade and threw errors about some directories in /usr/local being 775 not 755.
    2) that was because I had a non-standard package in there. I changed just those relevant directories to 755.
    3) reattempted pacman -Syu but got a bunch of errors: "call to execv failed"
    4) system was immediately borked. Reboot now throws me into a shell.
    This is a 32 bit system.
    Any ideas and help would be greatly appreciated.
    Last edited by lagagnon (2013-01-30 19:10:24)

    Solved my problem with the following method:
    1) used Arch LiveCD. Mounted boot and / partitions
    2) pacman -r /path/to/root -Syyu glibc filesystem
    3) reboot gave "Unable to find root device error"
    4) ran LiveCD again following these instructions: https://wiki.archlinux.org/index.php/Pa … onger_boot

  • I have a copy of Windows 7, and I am trying to bring up Boot Camp on my MacBook with Snow Leopard. But my Snow Leopard install disk is too old for Windows 7, so attempted Boot Camp install of drivers fails.

    I have a copy of Windows 7, and I am trying to bring up Boot Camp on my MacBook with Snow Leopard. But my Snow Leopard install disk is too old for Windows 7, so attempted Boot Camp install of drivers fails.
    I have hand-installed the NVidia graphics driver, and I have keyboard, minimal trackpad, and hardwire to internet all working. So the WIndows machine is usable. But I am hurting for wi-fi, better trackpad, and whatever else is delivered by Boot Camp. I can't use any of the Boot Camp upgrades because I do not have XP or Vista.

    While we all have MacBooks in this forum not all of us use Boot Camp. There's a Boot Camp Support Community where everybody uses Boot Camp. You should also post this question there.
    https://discussions.apple.com/community/windows_software/boot_camp

Maybe you are looking for