Windows Server 2003

Add System Restore to Server 2003

Download the file http://www.windowsreference.com/files/AddSystemRestore.zip and extract it

Double click on AddSystemRestoreEntries.reg (This will add some values to the registry). Click on Yes to continue

Then insert a Windows XP CD to your CDROM drive. Right click on sr.inf and click on install

Reboot the computer

Now you can find the System Restore utility from START–>All Programs–>Accessories–> System Tools –> System Restore

Sysinternals Suite Updates

New Updates to the Sysinternals Suite

Process Monitor v2.7
This update to Process Monitor, a system monitoring utility, adds a new option to the process tree dialog that direct it to show just the timeline for displayed events, uses kernel-based thread profiling on Vista and higher for better performance, and includes a number of minor fixes and enhancements.

ProcDump v1.5
ProcDump now includes a new switch that enables the creation of a process dump upon process termination, which can help with troubleshooting unexpected process termination. It also fixes a bug where the -ma switch wouldn’t generate a full dump when combined with -r , the Windows 7-specific process reflection switch.

If you want to download these apps go to the Systinternals Live Site or go here for the whole updated package.

Posted in August that is Definitely worth a read.

Mark’s Blog: The Case of the Temporary Registry Profiles
In the latest post in Mark’s “Case of the Unexplained” series, he documents a perplexing case affecting many Microsoft and Citrix customers that Microsoft Customer Support Services solved with the use of Process Monitor’s boot logging and stack trace features.

Windows cannot access the file gpt.ini

You get the Event Log Error….

Event Type:              Error
Event Source:           Userenv
Event Category:        None
Event ID:  1058
Date:                        5/27/2009
Time:                        10:25:00 AM
User:                        NT AUTHORITY\SYSTEM
Computer:                domain controller name
Description: Windows cannot access the file gpt.ini for GPO CN={31B2F340-016D-11D2-945F-00C04FB984F9},CN=Policies,CN=System,DC=domain,DC=com. The file must be present at the location <\\domain\sysvol\champstainless.com\Policies\{31B2F340-016D-11D2-945F-00C04FB984F9}\gpt.ini>. (Access is denied. ). Group Policy processing aborted. 

Here is the quick fix… download Windows Server 200x Support Tools, run the cmd prompt inside of the new folder created in Start; Programs; Support tools… then run “dfsutil /purgemupcache” on the server which is seeing the errors.

After that the errors should be gone!

NS

Windows Server 2003 Performance Advisor and Data Collector Sets

As I am sure some people know about this installable gem, I am also sure that not everyone does. Its called “Windows Server 2003 Performance Advisor” Here is an insert from Microsoft’s website…

“Microsoft ® Windows Server ™ 2003 Performance Advisor is the latest version of Server Performance Advisor, which is a simple but robust tool that helps you diagnose the root causes of performance problems in a Microsoft Windows Server 2003 deployment. Server Performance Advisor collects performance data and generates comprehensive diagnostic reports that give you the data to easily analyze problems and develop corrective actions

Microsoft ® Windows Server ™ 2003 Performance Advisor provides several specialized reports, including a System Overview (focusing on CPU usage, Memory usage, busy files, busy TCP clients, top CPU consumers) and reports for server roles such as Active Directory, Internet Information System (IIS), DNS, Terminal Services, SQL, print spooler, and others.”

This application was deemed so useful that it was built into Windows Vista and Windows Server 2008, it can be found by going to the control panel, then clicking on “Administrative Tools”, “Reliability and Performance Monitor”, then “Data Collector Sets”. You can then create a new “Data Collector Set” using templates under “User Defined” If you would like to see how they are configured and work, check out Petri’s article “Performance Monitoring the Easy Way”.

The link to download Performance Advisor is here. Also I am providing a link to Perti’s site that explains how to use the app, click here. Is it x64 compatible you say? My answer is yes.

Let me know if you found this blog useful!

NS

Why you should backup the SQL Installation Folder

I cannot and will not take credit for the following, but since my fellow senior engineer does not have a blog I figured I’d share this with everyone out there, because when searching for a solution, this one did not present itself.

Scenario : SQL Server, slow performance, and erratic behavior.

Solution : Reboot the server.

Outcome : Server Hung, for what seems a longer than usual amount of time.

Steps taken to rectify this behavior : Hard Power off the server (don’t ever do this, I mean sometimes you have to, but wait just a little bit longer before resorting to this) Things that could have been tried before, would be to run services.msc from another workstation or server, and possible you could connect and stop some services speeding up the process.

What ended up happening? The client shut the server off with thepower button.

Server boots back up and as you’d thing, SQL Server Service failed to start…

Customer reboots again.

SQL Server fails to start once again. Customer called EPIC Information Solutions (the company I work for)

A Programmer was on the task with the senior engineer mentioned above. We came back from a well deserved lunch and the VP of Managed Services asked us to help the programmer and the senior engineer. We tried pretty much everything, permissions on files, on services, hardware profile under services was disabled(which should be enabled, but that didn’t fix it) always got SQL Server Service failed to start “Access Denied” nothing seemed to fix it. Procmon.exe couldn’t identify which files it was failing on, there was no way to check to see if the databases were okay or not. Renaming Master.mdb and Masterldf.ldf netted the same results, nothing seemed to work. We used and abused Google like you wouldn’t believe… the backups were unreliable as they were years old. Sure there were backups of the SQL Databases, but how can you restore a SQL Backup when you cannot start the SQL Server Service and restore them with SQL Management Studio…. The worst came into view, rebuilding the master.mdb. We really didn’t want to do that.

Last ditch effort from my friend the senior engineer.

Rename the “Bin” folder under the “Microsoft SQL Server” folder and restored from a backup that ran last night.

Folder was restores, back to services.msc, highlight the SQL Server Service, and click start. Done… service started. SQL Server Management Studio started up and he was able to connect to the database. Our Programmer tested the app that required SQL and sure enough, it was working fine.

Thus as the title says, for the amount of space SQL Server directory takes… Back It UP!

Cheers,

NS

IBM ServeRaid

Here’s an interesting issue that I came across. I have a few clients who are still using IBM Servers, the server referred to here is an x226 Server and a ServeRaid 7 controller.

The client said that he would come in to work in the morning to check the backups and he would have to put in a reason as to why the server shutdown. He mentioned that the backups have also not been working for the last week or so. Just to let you know this wasn’t my fault, the client is not managed services and it is up to him to monitor and let me know if they are having issues. Anyways back to the blog :). We came in to troubleshoot the backup and found that we could run a test backup. Just in case a backup to disk was also setup. Next morning it was reported that the backup failed again.

I went onsite and used IBM UpdateXpress, this CD provides an all in one firmware update for all supported components contained within the server. You can download it here. I Updated the Server’s BIOS, ServeRaid Bios/Firmware, and also the firmware of the drives themselves.

I also ran an app from IBM’s Site called Dumplog, this will “dump” the configuration and event logs from the ServeRaid controller, don’t try to decipher any of the info in the txt file, you need to send this to IBM and they will tell you what your next step it based on the info contained within that file. Download it here

Well to make a long story short… the server would crash when stressed with I/O. I figured it had to be the controller, so I ran the onboard diagnostics, and sure enough the ServeRaid test failed. I exported the test log to a text file so I could send it to an IBM Tech. Once the IBM Tech saw the dumplog files he was able to tell me that a specific drive was failing, although the drive was not reporting it to the controller properly, thus the global hot-spare wasn’t kicking in. I ended up running ServeRaid Manager and marking the bad drive  defunct, then I pulled the drive out of the server. The global hot-spare then kicked in and the rebuild started.

All seems well. It would have been nice if the drive just marked itself bad in the beginning and the issue would have been resolved much faster.

IBM Tech Support requires firmware and drivers to be up to date before they will really help you, so everything I did needed to be done. IBM is now sending a tech onsite to replace the drive and also the tape drive as It still didn’t work in the end. A backup to disk job was configured before going off site.

Till next time!

NS