October 24, 2007
By Charlie Schluting

Many new workstations and servers are coming with integrated gigabit (define) network cards, but quite a few people soon discover that they can’t transfer data much faster than they did with 100 Mb/s network cards. Multiple factors can affect your ability to transfer at higher speeds, and most of them revolve around operating system settings. In this article we will discuss the necessary steps to make your new gigabit-enabled server obtain close to gigabit speeds in Linux, FreeBSD, and Windows.

Hardware considerations
First and foremost we must realize that there are hardware limitations to consider. Just because someone throws a gigabit network card in a server doesn’t mean the hardware can keep up.

For attaining maximum throughput, the most important options involve TCP window sizes.

Network cards are normally connected to the PCI (define) bus via a free PCI slot. In older workstation and non server-class motherboards the PCI slots are normally 32 bit, 33MHz. This means they can transfer at speeds of 133MB/s. Since the bus is shared between many parts of the computer, it’s realistically limited to around 80MB/s in the best case.

Gigabit network cards provide speeds of 1000Mb/s, or 125MB/s. If the PCI bus is only capable of 80MB/s this is a major limiting factor for gigabit network cards. The math works out to 640Mb/s, which is really quite a bit faster than most gigabit network card installations, but remember this is probably the best-case scenario.

If there are other hungry data-loving PCI cards in the server, you’ll likely see much less throughput. The only solution for overcoming this bottleneck is to purchase a motherboard with a 66MHz PCI slot, which can do 266MB/s. Also, the new 64 bit PCI slots are capable of 532MB/s on a 66MHz bus. These are beginning to come standard on all server-class motherboards.

Assuming we’re using decent hardware that can keep up with the data rates necessary for gigabit, there is now another obstacle — the operating system. For testing, we used two identical servers: Intel Server motherboards, Pentium 4 3.0 GHz, 1GB RAM, integrated 10/100/1000 Intel network card. One was running Gentoo Linux with a 2.6 SMP (define) kernel, and the other is FreeBSD 5.3 with an SMP kernel to take advantage of the Pentium 4′s HyperThreading capabilities. We were lucky to have a gigabit capable switch, but the same results could be accomplished by connecting both servers directly to each other.

Software Considerations
For testing speeds between two servers, we don’t want to use FTP or anything that will fetch data from disk. Memory to memory transfers are a much better test, and many tools exist to do this. For our tests, we used [ttcp](http://www.pcausa.com/Utilities/pcattcp.htm).

The first test between these two servers was not pretty. The maximum rate was around 230 Mb/s: about two times as fast as a 100Mb/s network card. This was an improvement, but far from optimal. In actuality, most people will see even worse performance out of the box. However, with a few minor setting changes, we quickly realized major speed improvements — more than a threefold improvement over the initial test.

Many people recommend setting the MTU of your network interface larger. This basically means telling the network card to send a larger Ethernet frame. While this may be useful when connecting two hosts directly together, it becomes less useful when connecting through a switch that doesn’t support larger MTUs (define). At any rate, this isn’t necessary. 900Mb/s can be attained at the normal 1500 byte MTU setting.

For attaining maximum throughput, the most important options involve TCP window sizes. The TCP window controls the flow of data, and is negotiated during the start of a TCP connection. Using too small of a size will result in slowness, since TCP can only use the smaller of the two end system’s capabilities. It is quite a bit more complex than this, but here’s the information you really need to know:

Configuring Linux and FreeBSD
For both Linux and FreeBSD we’re using the sysctl utility. For all of the following options, entering the command ‘sysctl variable=number’ should do the trick. To view the current settings use: ‘sysctl <variable name>’

• Maximum window size:
  • FreeBSD:
    kern.ipc.maxsockbuf=262144
  • Linux:
    net.core.wmem_max=8388608
• Default window size:
  • FreeBSD, sending and receiving:
    net.inet.tcp.sendspace=65536
    net.inet.tcp.recvspace=65536
  • Linux, sending and receiving:
    net.core.wmem_default = 65536
    net.core.rmem_default = 65536
• RFC 1323:
  This enables the useful window scaling options defined in rfc1323, which allows the windows to dynamically get larger than we specified above.
  • FreeBSD:
    net.inet.tcp.rfc1323=1
  • Linux:
    net.ipv4.tcp_window_scaling=1
• Buffers:
  When sending large amounts of data, we can run the operating system out of buffers. This option should be enabled before attempting to use the above settings. To increase the amount of “mbufs” available:
  • FreeBSD:
    kern.ipc.nmbclusters=32768
  • Linux:
    net.ipv4.tcp_mem= 98304 131072 196608

These quick changes will skyrocket TCP performance. Afterwards we were able to run ttcp and attain around 895 Mb/s every time – quite an impressive data rate. There are other options available for adjusting the UDP datagram sizes as well, but we’re mainly focusing on TCP here.

Windows XP/2000 Server/Server 2003
The magical location for TCP settings in the registry editor is HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters

We need to add a registry DWORD named TcpWindowSize, and enter a sufficiently large size. 131400 (make sure you click on ‘decimal’) should be enough. Tcp1323Opts should be set to 3. This enables both rfc1323 scaling and timestamps.

And, similarly to Unix, we want to increase the TCP buffer sizes:

ForwardBufferMemory 80000
NumForwardPackets 60000

One last important note for Windows XP users: If you’ve installed Service Pack 2, then there is another likely culprit for poor network performance. Explained in [knowledge base article 842264(http://support.microsoft.com/?kbid=842264), Microsoft says that disabling Internet Connection Sharing after an SP2 install should fix performance issues.

The above tweaks should enable your sufficiently fast server to attain much faster data rates over TCP. If your specific application makes significant use of UDP, then it will be worth looking into similar options relating to UDP datagram sizes. Remember, we obtained close to 900Mb/s with a very fast Pentium 4 machine, server-class motherboard, and quality Intel network card. Results may vary wildly, but adjusting the above settings are a necessary step toward realizing your server’s capabilities.

<djahandarie> we ain’t here to do e-c-e
<djahandarie> we’re here to do c-s-e on the w-e-b
<djahandarie> listen to me spit these rhymes
<djahandarie> while i program lines
<djahandarie> and commit web accessibility crimes
<djahandarie> word, son
<http402> You talk like your big on these I-Net kicks,
<http402> But your shit flows slower than a two-eighty-six.
<http402> I’m tracking down hosts and nmap scans,
<http402> While Code Igniter’s got you wringing your hands.
<http402> Cut the crap rap,
<http402> Or I’ll run ettercap,
<http402> Grab your AIM chat,
<http402> N’ send a PC bitch-slap!
<http402> peace
<djahandarie> you’re talkin bout down hosts and nmap scans
<djahandarie> while i got other plans
<djahandarie> you’re at your new job, but you can’t even do it right
<djahandarie> you just create a plight with your http rewrites
<djahandarie> i’ve been on the web since the age of three
<djahandarie> you just got on directly off the bus from mississippi
<djahandarie> respect yo’ elders, bitch
<http402> You’ve been webbin’ since three, but still ain’t grown up,
<http402> Gotta update your config and send the brain a SIGHUP.
<http402> You say you’re that old? No wonder you’re slow!
<http402> You’re knocking at the door while I run this show!
<http402> Elders my ass, you’re shit’s still in school,
<http402> Hunt and pecking at the keyboard like a spaghetti-damned fool,
<http402> Rim-riffing your hard drive like a tool,
<http402> Face it. I rule.
<djahandarie> i erase my harddrives with magnets (bitch)
<djahandarie> all you can do is troll on the fagnets
<djahandarie> and son, my brain’s wrapped in a nohup
<djahandarie> it wont be hurt by the words you throwup
<djahandarie> dont mind me while i emerge my ownage
<djahandarie> while you’re still over there apt-getting your porridge
<djahandarie> you say i’m still in school
<djahandarie> but the fact is that i know the rule
<djahandarie> cuz you need to go back to grade three
<djahandarie> and you better plea, that they take sucky graduates from c-s-e
<http402> Time to bend over and apply a patch,
<http402> Your brain’s throwing static like a CD with a scratch.
<http402> Your connection got nuked and you’ve met your match.
<http402> You run a single process like a VAX with a batch.
<http402> I’d pass the torch to a real winner
<http402> But it’d just scorch a while-loop spinner
<http402> Caught in a loop that you cant escape,
<http402> I run clock cycles around your words and flows,
<http402> Cuz your rhyme is like a PS fan: it’ blows,
<http402> Your water-cooled lyrics leak and it shows,
<http402> Take your ass back to alt.paid.for.windows.
<djahandarie> Good god, I can’t even respond to that.
<djahandarie> You win haha
* http402 takes a bow

So, I’ve been dieing to get an Ubuntu LiveUSB drive to work correctly.  I’ve been semi-successful.  It will work, but once you put it on another computer it gets flakey.

Anyway, I tried utilizing the standard Ubuntu installation from the CD, which allows you to simply install to a USB Flash drive.  Great.  I was doing this on a laptop with Windows XP already installed.  What the installer did, however, was overwrite the MBR of the Windows XP installation and installed GRUB (this is standard and OK).  However, this isn’t what I needed, because the system will not boot up if the USB drive is not connected – There’s no BOOT LOADER! ahh!

I tried using a Windows Installation Disk to fix it, however, I did not have drivers (or a floppy drive for that matter) for the SATA Hard Disk Drive (HDD) (BTW, what’s up with Windows STILL not fully supporting SATA out of the box for XP?!  Maybe it’s SP2′s fault, who knows).  So therefore, I could not use the typical fixmbr command from the Recovery Console.

To fix it I did the following.  I booted up using the Ubuntu installation on the USB drive.  Then I ran the following commands:

1. sudo apt-get install lilo
2. sudo fdisk -l
3. Find the windows installation: mine was on /dev/sda
4. sudo lilo -M /dev/sda
5. sudo reboot 0

This will place a boot loader on the windows disk and allow it to boot, without a hitch.  No, the Windows bootloader is not installed, but it works, and you will not notice the difference.

I hope this helps.  Good luck.

I have verified that this fix works, and that it is, in fact, a bug in the latest release.

EDIT: (8.11.08) Please make sure you change the quotes once you copy in the code.  WordPress will alter them, so you need to copy the code into your files and then delete each quote and retype it.  It should clear a lot of the PHP errors everyone’s getting, since it doesn’t know how to handle them, and is resulting in unpredictable behavior.

EDIT: (8.27.08) The paths to the files are as follows:

JOOMLA/components/com_docman/includes_frontend/upload.http.php
JOOMLA/administrator/components/com_docman/includes/files.php
JOOMLA/administrator/components/com_docman/classes/DOCMAN_file.class.php

Where JOOMLA is the root / home folder of the joomla installation (not to be confused with the root folder of the server).

---

Hello I’ve had the error message “error uploading. directory problem” with DOCman 1.4 RC3, debugged the code and found the solution – it is a bug in RC3:Effect:
- Error message: “error uploading. directory problem”, after step 2 of uploading a new file,Occurrence:
- with docman 1.4 RC3, (not with RC1)
- on windows-based webservers (directory paths with backslash)
- if in PHP configuration “magic_quotes_gpc” is set to “on”

Reason:
Bug in RC3:
“DOCMAN_Utils::stripslashes()” removes backslashes in ['tmp_name'], what should NOT be done!!
- in “com_docman/includes_frontend/upload.http.php”, Line 38
- in “com_docman/includes/files.php”, Line 177:

Solution:
In “com_docman/includes/files.php”, line 177
remove DOCMAN_Utils::stripslashes() function:
- incorrect: $file_upload = mosGetParam(DOCMAN_Utils::stripslashes($_FILES), ‘upload’);
- correct: $file_upload = mosGetParam($_FILES, ‘upload’);

In “com_docman/includes_frontend/upload.http.php”, line 38
remove DOCMAN_Utils::stripslashes() function:
- incorrect: $file = mosGetParam(DOCMAN_Utils::stripslashes($_FILES), ‘upload’);
- correct: $file = mosGetParam($_FILES, ‘upload’);

And instead: In “com_docman/classes/DOCMAN_file.class.php”, line 352
add DOCMAN_Utils::stripslashes() function:
- incorrect: $name = $file['name'];
- correct: $name = DOCMAN_Utils::stripslashes($file['name']);

For more information on ['tmp_name'] and backslashes
- see http://ch2.php.net/features.file-upload#42280 (11-May-2004 03:08)

That worked for me.

Good luck!
Benno

Now, there are plenty of places to find resources on this issue, because it’s quite common to have to do so.  My issue, was a little different, in that there were multiple layers of security (authentication and authorization) for me to access the files that were needed.  Here’s what the setup looked like:

Web Server -> WorkGroup -> Firewall -> Domain -> File Server -> Network Share -> SAN

As you can see, there are several issues are hand.  Not only do the accounts have to exist to allow for this communication (and an Administrator account is the only way to go), all sorts of permissions have to be valid for this to function (ie. Local Permissions within each OS on each Server, NTFS (file permissions) on each local server, Access permissions through the Firewall, Access permissions of Domain Resources, Local process level permissions, etc).  It was a nightmare, but I believe I figured it out.

There are a few things that needed to be done in this situation.  First, the permissions on the SAN needed to allow Administrators full control.  Additionally, the local user on the file server needed to have administrative privileges, to both the system and the SAN, thus allowing the network share.   Next, the firewall rules needed to allow file traffic (I can’t remember the SMB ports right now – and Windows needs netBios and something else to let file transfers go through, as well as authentication).  Finally, the Web Server local user needs admin rights.

To get this to work, I needed everything involved to run with elevated privileges, from IIS, to each thread in the application.  Since there was going to be a large amount of file movement, as well as resource and memory manipulation, it’s required.

Therefore, the admin credentials were inputted into IIS to map the network share from the Web Server to the File Server.  Next, I had to impersonate the admin user in each thread within the application, a pain, but the guide can be found via Google (if I have time, I’ll track them down again). Next, the machine.config needed to be modified to allow the ASP.NET process to run with the elevated permissions.  The web.config needs to be altered to allow impersonation with the credentials as well.

Now here’s the kicker, and something that’s not quite known but buried within Microsoft’s documentation.  For this to work (Work Group computer to authenticate to another computer) is to mirror the accounts (same username and password) on both servers (these are local accounts, not domain accounts – since the share was local on the server, the domain can be bypassed. If it were a domain resource, we would have to authenticate via NTLM to an Active Directory server, which would have been a bit more complicated).  This allows the hash sent from one system to another to be identical, and thus, you will authenticate (if you know the username and password on one workgroup computer, and it’s the same on another, chances are you’re who you say you are).

This took me a week to figure out, which was not enjoyable.  I hope this saves you some time.

So I recently got bombarded with 2 infections of this pesky beast. Some variants are easy to remove (SpyBot can simply pull them out) but the variant I came across was resilient. It loaded a DLL into the Winlogon.exe (injection) executable file (the Windows process responsible for authentication to the Operating System – Windows cannot run without it) and ran from there. So you can’t kill the process, because the OS will reboot. You can’t delete the DLL file, because the OS has it locked. Basically, it’s like a tumor in the center of your brain… there’s really no winning.

There are a few solutions out there (very few) such as Bayles’ solution and this one from a poster on TechRepublic , but unfortunately, neither was any good for me. Bayles’ solution works for variants that inject into Explorer.exe, which is loaded once the OS is loaded and you log in. However, winlogon is loaded as one of the first steps in loading the registry, so it’s loaded even before you press control-alt-delete (it’s actually the process that shows you the login screen). The other solution could have worked had I had administrative privileges on my machine, but I did not (this is my work machine). Therefore, I had to find another solution to bypass the OS from reinfecting itself, but still have access to the underlying file system.

I could have removed the hard drive, but I did not have another laptop to install the HDD into. I couldn’t boot to the Recovery Console using an XP CD, because I did not know the Administrative password. So my solution? I booted the laptop using an Ubuntu Linux CD I had in my laptop bag. Here’s what I did.

1. Boot normally in Windows XP and get the names of the infected DLLs using Bayles’ method (again, pendmove won’t work because winlogon is placed in memory (with the infected DLL) before pendmove is loaded).
2. Reboot using Ubuntu CD – hit enter at the first screen (Run or Install Ubuntu)
3. Unmount the NTFS file system (on a typical installation, it will be the entire drive) – We do this because the standard driver file is a read-only NTFS driver.
4. Open a terminal and install the ntfs-3g packages
   1. sudo apt-get install ntfs-3g
5. Create mount point
   1. sudo mkdir /mnt/test
6. Mount the hard drive
   1. sudo mount -t ntfs-3g /dev/sda1 /mnt/test
      1. /dev/sda1 – is the partition we want to mount, yours may be different
      1. /mnt/test – is the directory in which to mount the partition
7. You can then browse to the windows/system32 directory and delete the infected DLLs (in my case the path was /mnt/test/windows/system32/sbbqikklll.dll)
8. Reboot and then you must run some scanners. I suggest running SpyBot to clean up the rest of the garbage, and then maybe an anti-virus scanner (I used Symantec Corporate). Finally, run the Windows Malware Removal Tool.
   1. This step is extremely important. This trojan downloads other infections to your systems (ads, other programs, etc). Who knows if these other files contain other viruses, trojans, keyloggers, etc.

I hope this helps. Contact me if you need any special assistance.

EXT2IFSHere’s a neat little driver that allows you to mount your EXT2 partition under a wndows OS. I’ve tested this and it does indeed work.

EXT2IFS is an Installable File System Driver (IFS) for Microsoft Windows NT 4.0, Microsoft Windows 2000 (NT 5.0) and Microsoft Windows XP (NT 5.1).The driver can read the Second Extended File System (EXT2) and Third Extended
File System (EXT3)

Update:

Found this recently…

Ext2Fsd

Ext2Fsd is an open source Ext2 file system driver for Windows (NT/2K/XP). “Explorer” your linux ext2 volumes with ease under windows systems.

Update:

Another one!
Ext2 Installable File System For Windows

It provides Windows NT4.0/2000/XP with full access to Linux Ext2 volumes (read access and write access). This may be useful if you have installed both Windows and Linux as a dual boot environment on your computer.The “Ext2 Installable File System for Windows” software is freeware.

This post provides a summary of performance tuning options for a Windows Server 2003 file server. Included is information on NTFS, lanmanserver, NIC, Disk and HBA performance tuning options. None should be used in a production environment without testing. Note that the Microsoft excerpts come from a 2008 performance tuning document, after verifying that the options are supported on 2003.

General Performance Settings

Disable 8.3 name creation

HKLM\System\CurrentControlSet\Control\FileSystem\NtfsDisable8dot3NameCreation

Excerpt from Microsoft on this option:

The default is 0. This parameter determines whether NTFS generates a short name
in the 8.3 (MS DOS®) naming convention for long file names and for file names
that contain characters from the extended character set. If the value of this
entry is 0, files can have two names: the name that the user specifies and the
short name that NTFS generates. If the user-specified name conforms to the 8.3
naming convention, NTFS does not generate a short name.

Changing this value does not change the contents of a file, but it avoids the short-name attribute
creation for the file, also changing the way NTFS displays and manages the file.
For most file servers, the recommended setting is 1.

Note that when accessing files that go beyond the 260 MAX_PATH length, short filenames can be a very useful method of accessing these files.

Ignore Write Flush Commands from Clients

HKLM\System\CurrentControlSet\Services\LanmanServer\Parameters\TreatHostAsStableStorage
Excerpt from Microsoft on this option:

The default is 0. This parameter disables the processing of write flush commands
from clients. If the value of this entry is 1, the server performance and client
latency for power-protected servers can improve. Workloads similar to the
NetBench file server benchmark benefit from this behavior.

Network tuning

NIC Offloading

Newer NICs have offloading capabilities, allowing the Operating System to offload one or more tasks to the network adapter. For example, Broadcom BCM5708S NetXtreme II adapters have the following offload capability enabled:
• IPv4 Checksum offload – Calculation and validation of checksums on TX/RX of TCP and UDP packets
• IPv4 Large Send offload – Offload the segmentation of large packets to the hardware

TcpWindowSize

HKLM\System\CurrentControlSet\Services\Tcpip\Parameters\TcpWindowSize

Windows Server 2003 auto-tunes this setting, and benefits to manually choosing a window size are acknowledged in high-latency networks.

NumTcbTablePartitions

HKLM\system\CurrentControlSet\Services\Tcpip\Parameters\NumTcbTablePartitions

Increasing this number from the default of 1 can decrease contention in the TCP/IP stack. Microsoft warns of modifying this setting without significant testing, with a recommended maximum of the number of processors in the system.

MaxFreeTcbs and MaxHashTableSize

HKLM\system\CurrentControlSet\Services\Tcpip\Parameters\MaxFreeTcbs
HKLM\system\CurrentControlSet\Services\Tcpip\Parameters\MaxHashTableSize

MaxFreeTcbs can be adjusted manually to set the maximum number of TCP control blocks the system creates. This number determines the simultaneous number of connections the server can handle. If MaxFreeTcbs is changed, MaxHashTableSize should also be modified, dictating the size of the hash table that stores the control blocks.

Disk Tuning

LargeSystemCache and LanManServer file system caching

The amount of memory allocated to file system caching can help to increase the disk cache hits for a file server. In Windows Server 2003 this setting is enabled by default, although a casual glance of the system working set and the lanmanserver service working set do not show a large amount of the available physical memory for cache.

Windows 2003 also has logical block caching in addition to caching at the file system level. Further information could not be found on configuring or measuring the performance of logical block level caching.

The file cache is part of the system working set, which is protected from excessive trimming when the LargesystemCache option is usedThe lanmanserver service working set must also be protect in some way, part of services.exe (use tasklist /svc).

CacheSet from sysinternals can be used to modify the system working set size

Excerpt from Microsoft TechNet:

The Memory object performance counter System Cache Resident Bytes reports the amount of real memory currently in use by the file cache. As the number of System Cache Resident Bytes increases, we normally expect that the various measures of hit ratio will also increase. Moreover, the cache size can grow simply as a function of the size of the files that are currently in use and their pattern of access.

The Cache Resident Bytes counter reports the amount of real memory the file cache is currently occupying. The Cache Bytes counter, which sounds like it might tell you the size of the cache, actually reports the full system working
set, which includes Cache Resident Bytes and several other real memory areas. In a Windows 2000 file server (remembering Windows 2000′s heritage as the follow-on to the joint IBM/ Microsoft-developed OS2 LAN Manager), the file cache so dominates the system working set that internal documentation frequently refers to the entire system working set as the cache. This usage carries over to tools like Task Manager, which labels the system working set as the System Cache in the Performance tab, illustrated in Figure 7-2. The Windows NT version of Task Manager called this field File Cache, which is probably just as misleading. Curiously, the number of bytes in the System Cache reported by Task Manager does
not correspond exactly to the Cache Bytes counter in the System Monitor.

Disk Alignment

Using disk alignment to realign partitions can occasionally decrease the number of disk I/O operations. This occurs because the MBR is on the first 63 sectors of a disk, and the first partition starts on the 64th sector instead of the 65th sector – the beginning of the next boundary. This is vendor and disk specific, and does not apply to every disk.

Excerpt from Microsoft TechNet:

Microsoft Windows 2000 Server has an internal structure known as the master boot
record (MBR) that limits the maximum number of hidden sectors to 63. This
characteristic of the MBR causes the default starting sector for disks that
report more than 63 sectors per track to be the 64th sector. Therefore, when
programs transfer data to or from disks that have more than 63 sectors per
track, misalignment can occur at the track level, with allocations starting at a
sector other than the starting sector. This misalignment can defeat system
optimizations of I/O operations that are designed to avoid crossing track
boundaries.

Diskpar.exe is a command-line tool from the Windows 2000
Server Resource Kit that can explicitly set the starting offset in the MBR. By
doing this, the track is aligned with the physical disk partition, which results
in an improvement in disk performance. Exchange writes four kilobytes to the
database and up to 32 kilobytes for the streaming data. Therefore, make sure
that you set the starting offset to be a multiple of four kilobytes.

Write-caching option on each disk

Excerpt from Microsoft on this option:

Enabling write caching allows writes to be completed immediately after being
cached in the storage subsystem. Note that with this action a period of time
passes during which a power failure or other catastrophic event could result in
a loss of the data. However, this period is typically fairly short because write
caches in the storage subsystem are usually flushed during any period of idle
activity. Alternately, you can use time-outs at the cache level to force dirty
data out of the cache even if other active requests exist.

This option is enabled by default on all SAN-attached storage

‘Advanced Performance’ option on each disk.

Excerpt from Microsoft on this option:

The advanced performance option strips all write-through flags from disk
requests and also removes all flush-cache commands. The assumption is that if
you have power protection on your I/O path you don’t need to worry about those
two pieces of functionality; by definition, any written data is safe and
“in-order” after it is copied into power-protected storage subsystem hardware,
just as if it had been written to the physical disk media.

This option is disabled by default on all SAN-attached storage.

Emulex Lightpulse scatter/gather list elements

“HKLM\System\CurrentControlSet\Services\lp6nds35″

Add (if not already present) the following Keys under lpxnds

Parameters->Device (Parameters apply to ALL LPxxx adapters)
or
->Devicen (Parameters apply to LPxxx adapter number ‘n’ 0-99)

Under Device or Devicen add a value of “MaximumSGList” Type REG_DWORD,
with a value of from 13 to 255 (decimal). Microsoft specifies to use
this value sparingly, as these request entries come from Non-Page-Pool.

Emulex Lightpulse asynchronous requests

“HKLM\System\CurrentControlSet\Services\lp6nds35″

Add (if not already present) the following Keys under lpxnds

Parameters->Device (Parameters apply to ALL LPxxx adapters)
or
->Devicen (Parameters apply to LPxxx adapter number ‘n’ 0-99)

Under Device or Devicen add a value of “NumberOfRequests” Type REG_DWORD,
with a value of from 16 to 256 (decimal). Microsoft specifies to use
this value sparingly, as these request entries come from Non-Page-Pool
(this is especially true with the ALPHA systems).

References:

Windows Server 2008 performance tuning
http://download.microsoft.com/download/9/c/5/9c5b2167-8017-4bae-9fde-d599bac8184a/Perf-tun-srv.docx

How To Improve Windows 2003 File Server Performance
http://support.microsoft.com/kb/555041

CacheSet
http://technet.microsoft.com/en-au/sysinternals/bb897561.aspx

How to Configure the Storage Subsystem
http://technet.microsoft.com/en-us/library/bb643100.aspx

Examining and Tuning Disk Performance
http://www.microsoft.com/technet/prodtechnol/Windows2000Pro/reskit/part6/proch30.mspx?mfr=true

Aligning Disk Partitions by Using Diskpar.exe
http://technet.microsoft.com/en-us/library/bb643097.aspx

PagedPoolSize
http://technet2.microsoft.com/windowsserver/en/library/b1bda681-28b0-4339-a4ea-feb0fd9ff0c01033.mspx?mfr=true

Optimizing Your Memory Configuration
http://www.microsoft.com/technet/prodtechnol/windows2000serv/reskit/core/fnec_evl_fhcj.mspx?mfr=true

About Cache Manager in Windows Server 2003
http://support.microsoft.com/kb/837331

MaxPagedMemoryUsage for the server sevice
http://technet2.microsoft.com/WindowsServer/en/Library/357c7af1-7a0b-47a5-8af1-ff44756c498a1033.mspx

MaxPagedMemoryUsage for the server sevice
http://www.microsoft.com/technet/prodtechnol/windows2000serv/reskit/regentry/58636.mspx?mfr=true

LargeSystemCache
http://technet2.microsoft.com/WindowsServer/en/library/efa621bd-a031-4461-9e72-59197a7507b61033.mspx

Optimizing Your Memory Configuration
http://www.microsoft.com/technet/prodtechnol/windows2000serv/reskit/core/fnec_evl_fhcj.mspx?mfr=true

File Cache Performance and Tuning
http://technet.microsoft.com/en-us/library/bb742613.aspx

Cluster resource command-line operation
http://technet2.microsoft.com/windowsserver/en/library/f6b35982-b355-4b55-8d7f-33127ded5d371033.mspx

File Cache Performance and Tuning
http://technet.microsoft.com/en-us/library/bb742613.aspx

About Cache Manager in Windows Server 2003
http://support.microsoft.com/kb/837331

How to Configure the Storage Subsystem
http://technet.microsoft.com/en-us/library/bb643100.aspx

Why should you use Diskpar
http://msexchangeteam.com/archive/2005/08/10/408950.aspx

Emulex Lightpulse HBA device settings
http://contents.driverguide.com/content.php?id=106500&path=README.TXT

A document on the Microsoft Office Online site provides detailed instructions on how to export your Contacts from Outlook.

Here’s the scoop:

1. In Outlook, on the File menu, click Import and Export.
2. Click Export to a file, and then click Next.
3. Click Comma Separated Values (Windows), and then click Next.
4. In the folder list, click the Contacts folder, and then click Next.
5. Browse to the folder where you want to save the contacts as a .csv file.
6. Type a name for the exported file, and then click OK.
7. Click Next.
8. Click Finish.

That gives you a comma separate list of your Outlook Contacts. Transfer that to your new computer, then you can import them into your new version of Outlook with the following steps:

1. On the File menu, click Import and Export.
2. Click Import from another program or file, and then click Next.
3. In the list, click Comma Separated Values (Windows), and then click Next.
4. Click Browse, locate your file, and then click Next.
5. Click your Contacts folder or another folder that contains contacts, and then click Next.
6. In the list, click the file that contains your address list, and then click Map Custom Fields.

That should get you going. If you need more help here are two reference pages from Microsoft, one on Exporting Outlook Contacts and another on Importing Outlook Contacts.

Where is ASP.NET 3.5 on IIS ?

Most people on installing .NET 3.5 on the server “expect” an entry for  v3.5 under the ASP.NET version Tab in IIS manager. And not finding that entry start panicking. If you have been one of those read on. The more tech savvy folks try to run aspnet_regiis -i from the v3.5 framework folder… but even that is missing

That brings us to the question

What exactly is ASP.NET 3.5 ?

Take a look at the v3.5 framework folder you will not find most of the files needed for ASP.NET to run like aspnet_isapi.dll / aspnet_filter.dll / aspnet_regiis.exe / aspnet_regsql.exe…

The reason being .NET 3.5 is not a stand alone framework like v2.0 or v1.1 . It is just an extension of the 2.0 framework. The extension provides developers with new language features like Anonymous Types, Extension Methods, Lambda Expressions and of course LINQ …. and AJAX is now integrated … also you have a few new ASP.NET controls <asp:ListView>, <asp:LinqDataSource>, <asp:DataPager>

Will I get ASP.NET version 3.5 under the ASP.NET tab in IIS 6.0 ?

Nope. Because as far as IIS is concerned nothing has changed. The aspnet_isapi.dll from the version 2.0 folder will still handle the requests and there is no new ISAPI dll specific to version 3.5. So you don’t have to change any Script Mappings in IIS …just leave the web application pointing to v2.0 under the ASP.NET Tab. And this brings us to the next question

What exactly is the extension ?

For that if you have Visual Studio 2008 go ahead and create a Website targeting the .NET 3.5 framework. Then go to the Property Pages of the Website and select References. You will find an entry like this

And those four assembly references provide the new features of v3.5.  Here is a 10,000 ft view of the functionality implemented in these assemblies

System.Core – Provides namespaces like System.Linq, System.IO

System.Data.DataSetExtensions – LINQ to DataSet

System.Web.Extensions – AJAX

System.Xml.Linq – New XML API.. supports LINQ to XML

Do I have to configure anything manually ?

No. When you install the .NET 3.5 redistributable on the server it installs these assemblies in the GAC. And any website that refers to these assemblies can fetch them from the GAC so you do not have to configure anything manually. All you need to do is publish the content to a web application that is marked with ASP.NET v2.0

And now you should be thinking

The new language features should require a new compiler. How exactly is the new compiler used ?

The answer is your web.config file. When you create an ASP.NET 3.5 website in Visual Studio 2008 an entry is made in the web.config something like

<system.codedom>
  <compilers>
    <compiler language="c#;cs;csharp" extension=".cs" warningLevel="4"
              type="Microsoft.CSharp.CSharpCodeProvider, System, Version=2.0.0.0, Culture=neutral, PublicKeyToken=b77a5c561934e089">
      <providerOption name="CompilerVersion" value="v3.5"/>
      <providerOption name="WarnAsError" value="false"/>
    </compiler>

The new features do require a new compiler to be used and the web.config tells .NET that the v3.5 compiler has to be used.

What about IIS 7.0 ?

IIS 7.0 is no different.