Archive for the ‘Troubleshooting’ Category

15 Minute Beeps

I was asked to look into a beeping in a server room which no one could find. 

It was understandably difficult as there were three server racks with modems, routers, switches, servers, PCs, UPSs, CCTV, and audio equipment.  To add to that they were single beeps, quite wide spread apart, and within a noisy, concrete room.  I determine after a little while they were single beeps and exactly 15 minutes apart.  Typically first suspicions are the UPS units, but 15 minutes is not typical of APC UPS devices.  To make a long story short, it turns out in a far corner where 2 fiber internet connections entered there were small UPSs units supplied by the service provider.  One of the UPS units showed a failed battery.  Being behind the open steel door, no one had seen the light. 

This seems like a trivial post and an obvious solution but I am posting as there are many forums on the Internet with people not being able to locate a device creating a beep every 15 minutes.  This is common with FIOS, Verizon, and Bell equipment.  One story goes as far as to explain a fellow turning everything off in the server room, one device at a time, every 15 minutes, till everything was shut down and unplugged, but could still hear the beeps.

Fibre1

DFSR Event ID: 2147485861

It seems Server Essentials and possibly others will often show in the daily report a DFSR Event ID: 2147485861 error. There is also a corresponding Warning Event logged in the DFS Replication log (sub-folder of Applications and Services Logs) of the event viewer with Source DFSR and Event ID 2013, followed by an Event ID 2212. This is most often caused by a dirty shutdown.

image

There are numerous articles on the Internet explaining how to resolve this, but I was told they were not clear to some readers. So, hopefully to clarify:

Firstly open Regedit and locate the key below, and verify it is set to 1. If not change it to 1.   (Note: you should always backup the registry before making any changes)

HKLM\System\CurrentControlSet\Services\DFSR\Parameters\StopReplicationOnAutoRecovery

Then in the error message within the report locate the command as highlighted in the image above, but cut and past it from your server report as it will have the correct volume GUID. Make sure it is all one line. You may want to use Notepad to reassemble if broken in the report. You need to use an elevated command prompt to run this.

If you run the command with the wrong GUID it will display; “No Instance(s) Available”.

If successful your command window should look similar to this:

image

This should resolve the problem, but it may return at some point in the future.  The Windows Server 2012 Essentials Build document (wiki article) suggests to prevent this in the future, after running the command successfully, change the aforementioned registry key from 1 to 0.  Doing so enables DFS Replication automatic recovery.

http://social.technet.microsoft.com/wiki/contents/articles/13620.windows-server-2012-essentials-build-document.aspx#DFSR_Error_in_Health_Reports

Cisco/Linksys Bad Gateway

Recently after installing a new Linksys (Cisco) EA6300 router we received a “502 – Bad Gateway” error message when trying to access the web management page.

image

(Note: the IP in the image is correct for this site, but the default of course is 192.168.1.1)

Installation of the router and configuring went as expected, and internet access by client machines was fine, but after completion and reboot the Bad Gateway message appeared when trying to log back into the router. This seems to be a common issue with numerous models not just the EA6300, but also the EA6500, EA4500, EA2700, E4200, E1200, and more.  There are numerous posts about this issue on many blogs and message boards, all of which suggest starting to resolve by pressing the reset button.  If you have a detailed configuration with custom IP configurations, port forwards, DDNS and more, that would be quite a nuisance and time consuming.

We discovered simply disconnecting the WAN / Internet connection to the router and refreshing the web management page allowed access and log in.  Once logged on the WAN connection could be reconnected.  Logging out and back in reproduced the problem.  The solution was to gain access as described and then using the built-in utility update the router’s firmware.  Once updated there were no further problems.

.NET 3.5 & .NET 2.0 on Win 10

There are numerous 3rd party applications that require .NET 3.5 and/or .NET 2.0 such as QuickBooks, Profile, and more.  Normally you simply go to: Control Panel, Programs and Features, Turn Windows Features On and Off, select .NET Framework 3.5 (includes .NET 2.0 and 3.0) and install.

image

However with Windows 10 it will want to “Download files from Windows Update” and then fail, primarily when joined to a domain that has WSUS (Windows Server Update Services) enabled, resulting in an Error code: 0x800F081F.

image

The problem also exists on Windows 8 and 8.1, with numerous suggestions to resolve including removing specific updates.  These updates do not exist on Win 10, but if they relate to your problem see: http://www.askvg.com/fix-0x800f0906-and-0x800f081f-error-messages-while-installing-net-framework-3-5-in-windows-8/

To give credit where credit is due, Microsoft’s solutions to the Windows 10 problem are outlined in the following article: https://support.microsoft.com/en-us/kb/2734782

From that I was able to resolve using the following steps. :

  • Attach a Windows 10 install ISO, either by inserting the install CD, USB, or a path to an ISO file on the network.  The latter can be achieved by using the USB/ISO creation tool which downloads the files from Microsoft and creates the ISO from Microsoft: https://www.microsoft.com/en-us/software-download/windows10
  • If you are using an ISO file you can mount it so that it can be accessed using a drive letter within Windows 8/10 by highlighting the ISO file and choose manage then mount, from the menu bar in Windows Explorer.
  • Edit local group policy to look for your ISO when Microsoft Update cannot be accessed.  Open group Policy by entering  gpedit.msc  in the search box or from an elevated command line.
  • Locate the following policy: Computer Configuration | Administrative Templates | System
  • In the right hand window, scroll down past the folders and locate the specific policy: “Specify settings for optional component installation and component repair”.
  • image
  • Double click on the policy to open it, click the radio button “Enable”, and in the box “Alternate source path” enter the path to the necessary files.  They are located in the \sources\sxs folder of the ISO.  In my case this would be E:\sources\sxs
  • Note: there is an option in the policy to “contact Windows Update directly” but this did not work for me or others.
  • Force group policy to update by rebooting or from an elevated command prompt enter   gpupdate  /force
  • Now you can return to: Control Panel, Programs and Features, Turn Windows Features On and Off, select “.NET Framework 3.5 (includes .NET 2.0 and 3.0)” and install.  It should locate the files and install without a problem.
  • I recommend a reboot after doing so.
NOTE:  WINDOWS 10

–  I recently had difficulties applying this method to a Windows 10 machine.  I believe it may be related to it being an upgrade from 8.1, but I had to download the Windows 10 trial .ISO file from TechNet, mount it, and run the following command from an elevated command prompt.  (Substitute your drive letter for ‘D:’, and path if necessary)

dism /online /add-package /packagepath:E:\sources\sxs\microsoft-windows-netfx3-ondemand-package.cab

The Mailbox has been Quarantined

I recently came a cross issue with an Exchange server running out of room and after the cleanup, one user only, would receive the message; “Cannot open your default e-mail folders.  Microsoft Exchange is not available.  Either there are network problems or the Exchange server is down for maintenance.”

image

The same user could also not log into Outlook Web Access.  After verifying DNS was working correctly, and then reviewing the server’s Event Viewer Application  logs at the time of the failure, a MSExchangeIS error with Sevent ID 10018 was noted.  “The mailbox for user 10c98e9d-1bcb-441c-a5bc-1a3fa19336f8: /o=First Organization/ou=Exchange Administrative Group (FYDIBOHF23SPDLT)/cn=Recipients/cn=John Doe has been quarantined. Access to this mailbox will be restricted to administrative logons for the next 6 hours.”

image

The Microsoft articles listed below reference this issue and how to resolve.  Assuming there is no corruption with the mailbox you can verify there is a quarantined mailbox by viewing and simply deleting the registry key under:  HKLMSYSTEMCurrentControlSetServicesMSExchangeIS<Server Name>Private-{db guid}QuarantinedMailboxes

I recommend, as always, backing up the registry before doing so.  Then to apply the change restart the Microsoft Exchange Information Store service.

https://support.microsoft.com/en-us/kb/2603736

https://technet.microsoft.com/library/bb331958.aspx#SH

Microsoft MVP Virtual Conference

Sign up for the first ever Microsoft online Virtual conference.  Two days of free sessions. 
http://mvp.microsoft.com/en-us/virtualconference.aspx 

image

  • · World-class free online conference that features technical content presented by Americas’ region MVPs that is open to the public
  • · More technical content (Level 200, 300, 400), less marketing
  • · 5 tracks: IT Pro English, Dev English, Consumer English, Mixed Spanish, Mixed Portuguese
  • · Event will be broadcast via Lync using L+ which enhances the conferencing capabilities of Lync
  • · Two full days of sessions with simultaneous webcasts running across all 5 tracks
  • · Thursday May 14th and Friday, May 15th
  • · Start at 8am PT and running until 6pm PT (Pacific)
  • · Day 1: 45 sessions + Keynote, Day 2: 50 sessions
  • · Keynote on Day 1 to be delivered by Steve ‘Guggs’ Guggenheimer, Corporate VP of DX
  • · On Demand content available via Channel9
  • · This event is not just for MVPs, it’s for everyone!
  • #MVPvConf   #MVPBuzz

Conference Promo Video:  https://www.youtube.com/watch?v=k9mnBYT3xro

image

Virtual Machine Management Service failure

I recently came across a Hyper-V host which after a power outage and a dirty shut  down, due the absence of a UPS, could not connect to the Virtual Machine Management Service.  When opening the Hyper-v console the VMs were not listed, and selecting “Connect to Server” resulted in the following error message; “An error occurred while attempting to connect to server ‘ServerName’.  Check that the Virtual Machine Service is running and that you are authorized to connect to the server.”

image

The Hyper-V services displayed running in the Services management console, and permissions had not changed, however when checking for dependencies for the Hyper-V Virtual Management Service it show no dependencies.  Comparing this to another Hyper-V host reveals the service depends on the Remote Procedure Call (RPC) and Windows Management Instrumentation service. Both of these were running as well.

When services and system information is not present in various consoles, and the WMI service is running, it can often be a result of missing or corrupt WMI files.  You can confirm this by running msinfo32.exe (System Information).  If so, you will receive a message “Can’t collect Information.  Cannot access the Windows Management Instrumentation software.  Windows Management files may be moved or missing.”

image

If this is the case there is an easy fix:

  • Always have a full backup of your system before making any changes!
  • Note:  Microsoft recommends troubleshooting, and restoring the WMI files rather than deleting them as I describe here.  Deleting the files can affect 3rd party software which may require reinstalling.  For more information see the links at the end of this article.  In this case the WMI checks showed no inconsistencies and a reset failed, so rebuilding was the best option.
  • Stop the Windows Management Instrumentation service, which will warn you that it will also stop the Hyper-V Virtual Machine Management and other services, verifying the link between the two.
  • Rename the WMI information file folder to something such as .old  C:\Windows\System32\wbem\Repository.old
  • image
  • Restart the host.   Upon reboot it will recreate the Repository folder and its contents.  You may find the first logon hangs for a few minutes and also it may be slow for a while as it rebuilds the files.
  • image
Additional information regarding troubleshooting WMI:

http://blogs.technet.com/b/askperf/archive/2009/04/13/wmi-rebuilding-the-wmi-repository.aspx

http://technet.microsoft.com/en-us/library/ff404265.aspx

Office 365 authentication did not succeed

I have had a few questions regarding a message “Office 365 authentication did not succeed” suddenly appearing both in the daily reports and the Alert Viewer of Server Essentials.  The alert viewer suggests changing the admin account (or refresh it) in the Office 365 tab of the Essentials Dashboard, however doing so fails with a message stating you are using the wrong account or password.

image

In most cases if you log into the Office 365 site using the domain’s admin e-mail account you will find the password has expired and you are asked to update it.  Do so and return to the Dashboard entering the new password which should now allow it to validate and eliminate the error.

image

Disable WSUS on Managed Computers

The past 8 or more years most of us have managed PC updates using WSUS (Windows Server Update Service) and Group policy.  However, the structure of the modern office has changed to a large percentage of mobile employees who never ‘touch down’ at headquarters.   If these devices do not connect to the domain they do not have updates applied.

A client who has not returned to the office in 18 months, and likely will not for the life of their laptop, recently asked how they could update their machine manually.  Currently they were not able to do so as Windows Update showed “settings are managed by your system administrator”, in other words, by WSUS

image

It is quite simple to disable WSUS management in the registry, however remember if the device is reconnected to the domain, the WSUS policies will be reapplied.  Therefore you may want to move the device to an OU not linked to the WSUS policy or remove the device in the policy under security filtering.

Disclaimer:  Be aware making incorrect registry changes can have disastrous effects to the health of the device.  Be sure to backup the registry before editing.  To do so see the following Microsoft article; “How to back up and restore the registry in Windows”  http://support.microsoft.com/kb/322756 

  • Open the registry editor, by entering Regedit in the Start / Run box, and browse to:  HKLM\Software\Policies\Microsoft\Windows\
  • Locate the WindowsUpdate  Key and delete it
  • Reboot the PC (may take 2 reboots)
  • Now you can manually update and configure Windows updates to automatically check for and install updates directly from the Microsoft Update site

image

You may want to consider using a newer service such as Windows Intune to manage your computers, especially mobile devices.  http://www.microsoft.com/en-us/server-cloud/products/windows-intune/

Java – page cannot be displayed

It seems the Java download site has been down for 24 hours.  Many internet posts by folk receiving “This page can’t be displayed” after clicking the download button.  The “See all Java downloads” page results in the same message.

A site test reveals it is down as well:  http://downforeveryoneorjustme.com/javadl.sun.com

I did find the Oracle site worked without any problems: http://www.oracle.com/technetwork/java/javase/downloads/jre7-downloads-1880261.html

Tag Cloud