Jump to content
Strawberry Orange Banana Lime Leaf Slate Sky Blueberry Grape Watermelon Chocolate Marble
Strawberry Orange Banana Lime Leaf Slate Sky Blueberry Grape Watermelon Chocolate Marble

MSFN is made available via donations, subscriptions and advertising revenue. The use of ad-blocking software hurts the site. Please disable ad-blocking software or set an exception for MSFN. Alternatively, register and become a site sponsor/subscriber and ads will be disabled automatically. 


Sign in to follow this  
Tripredacus

[2003] Need to determine where bottleneck resides

Recommended Posts

I've got some questions about which Counters I should be using to determine the problem with our file server. This server is not listed in my Servers thread, but here are the specs:

Server 2003 Standard x64 SP2

Intel Server board (likely 3210SHLC or other entry level Intel Server board)

4GB DDR2 533

2x Xeon quad-core E5320 1.86GHz

healthy RAID5 array on 3Ware 9550 8 port

A main concern is that the OS volume is also in the RAID5 as well as the data volume. They are as follows:

C: is a 50GB partition and has 6.34GB free

D: is a 1.7TB partition and has 287GB free

While under load (load meaning an active GhostCast Multicast session is in progress) doing anything on the server is very slow. Examples are: navigating the Start Menu, opening Windows, traversing folders, connecting to 3DM2 (using IE6) and even the GhostCase itself is slow. Network throughput out during the Cast sits around 1%, and average perhaps 100Mbps. The network is not the bottleneck of the system, as our WDS Server has no problem achieving respectable speeds, and of course the physical system seems to suffer. After load is released from the server, it returns to normal.

I ran the following Counters and all results appear nominal:

Disk

Current Disk Queue Length: < 1 (I did not get to capture this value before the load was released)

Avg. Read Queue Length: 0.229

Avg Write Queue Length: 0.116

% Disk Time: 20.104

Network

Output Queue Length: 0

CPU

CPU Time: 21.870

What other Counters should I try to help determine the problem?

Share this post


Link to post
Share on other sites

check the network card for any NIC offloading features, if the NIC is offloading to the CPU task and it is slowing the PC you may have to turn them off

Share this post


Link to post
Share on other sites

I tried to both apply the hotfix and disabling the options manually. Both the same result. This occurs during the two scenarios:

1. Running the hotfix, before the screen with the Finish button comes up

2. Deselect/disable the offloading options in Device Properties, click OK and before Properties screen goes away

DRIVER_IRQL_NOT_LESS_OR_EQUAL

STOP: 0xD1 (0x141, 0x2, 0x0, 0xFFFFFADF8EF179E2)

e1e5132e.sys <--- this is the Intel Pro 1000 driver

Address FFFFFADF8EF179E2 base at FFFFFADF8EF0D000 Datestamp 45f08ad6

I stopped the machine from doing a memory dump because there would be no feasible way for me to get the DMP file off the drive. Both times I restarted and booted back into Windows using Last Known Good Config. So it seems Server 2003 doesn't like turning off those features.

I did notice in the Intel NIC properties, concerning TCP/IP offloading, that it should not be enabled if servicing DOS clients, which we use the DOS client of Symantec Ghost.

Any other ideas?

Share this post


Link to post
Share on other sites

If disabling offloading causes a bugcheck, you've definitely got driver problems. Those are classic TCP chimney behaviors, honestly.

Share this post


Link to post
Share on other sites

I updated the driver for the NIC to the latest, which is still over a year old, and rebooted. Now I do not even have an Advanced tab, although the device still has the same name. How can I determine that these features are not being used?

I won't be able to see if the issue is resolved as we won't be using Ghost likely until early next week. Hopefully this driver update was enough.

Share this post


Link to post
Share on other sites

If you follow the instructions in the KB article linked and do it manually, it won't matter what the driver was set to. Disabling the interfaces at the OS level will mean the driver will start up, query support, and won't enable those functionalities once loaded if the OS says "no".

Share this post


Link to post
Share on other sites

The driver update seems to have worked. We will likely end up doing some more testing tomorrow, as I only ghosted a netbook and its image is pretty small. We shall see how ghosting a server goes, but I think this issue is solved.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...