Home | Contact Us | Submit Troubleshooting Info | Search | Help
Output | Software Solutions | Proofing | Color, Screening and Calibration | Scanning | Digital Photography | Consumables | POS
Home | Contact Us | Submit Troubleshooting Info | Search | Help
Output | Software Solutions | Proofing | Color, Screening and Calibration | Scanning | Digital Photography | Consumables | POS
|
|
||
Product |
Prinergy 1.0; Prinergy 1.1; Prinergy Connect 2.0 | |
Platform |
||
Operating system |
||
|
|
||
Error |
||
|
|
||
Symptoms |
The system seems like it’s slower than it should be. Running the NETPERF utility shows that network performance between the primary and secondary is much slower than expected. How to reproduce: Change the TCP/IP service’s TcpWindowSize registry setting to an inappropriate value, one that would cause excessive communication overhead. For example, when the Packet Engines gigabit Ethernet card is used, mis-type the TcpWindowSize value that you’re supposed to enter, and put in 0xFFF (4095 decimal) instead of 0xFFFF (65,535 decimal). |
|
|
|
||
Cause |
||
|
|
||
Solution |
Either remove the TcpWindowSize registry parameter completely so that NT chooses an appropriate default value (I recommend this for 100BaseT), or set it to an appropriate value (i.e. 0xFFFF for the gigabit card). This is in HKEY_LOCAL_MACHINE\System\CurrentControlSet\Services\Tcpip\Parameters Permanent fix: TBD. Somehow we should detect at runtime if inappropriate network settings are in effect. Permanent fix: |
|
|
|
||
See also |
||
|
|
||
Keywords |
networking |
|
|
|
||
Technical contact(s) |
Simon Chen |
|
Submitted by |
||
Date submitted |
||