Search String: Display: Description: Sort:

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*TCP\-Protection\s+is\s+really\s+a\s+pain\.\.\.\s*$/: 24 ]

Total 24 documents matching your query.

1. TCP-Protection is really a pain... (score: 1)
Author: Christian Schmid <webmaster@xxxxxxxxxxxxxx>
Date: Thu, 03 Feb 2005 05:07:30 +0100
Your new dynamically adjusted socket-buffer in 2.6.10 is really a pain for big servers. PLEASE tell me a way how to disable it. Thank you. Best regards, Christian Schmid
/archives/netdev/2005-02/msg00073.html (8,065 bytes)

2. Re: TCP-Protection is really a pain... (score: 1)
Author: Stephen Hemminger <shemminger@xxxxxxxx>
Date: Wed, 2 Feb 2005 20:54:37 -0800
sysctl -w net.ipv4.tcp_wmem="4096 8192 16384" Your single stream will be slower, but the memory footprint will be smaller.
/archives/netdev/2005-02/msg00074.html (8,716 bytes)

3. Re: TCP-Protection is really a pain... (score: 1)
Author: Christian Schmid <webmaster@xxxxxxxxxxxxxx>
Date: Thu, 03 Feb 2005 06:21:46 +0100
Actually, thats my problem. Single streams are too slow! Before I had buffers up to 500 KB. This was very nice to CPU because I only needed to "push" more data once in 5 seconds. I am doing this ever
/archives/netdev/2005-02/msg00075.html (10,207 bytes)

4. Re: TCP-Protection is really a pain... (score: 1)
Author: Jeff Garzik <jgarzik@xxxxxxxxx>
Date: Thu, 03 Feb 2005 01:50:45 -0500
Christian Schmid wrote: btw: Another problem I am experiencing is that downloads suddenly break in speed from 360 kb/sec to 8-12 kb/sec. 5 seconds later they stall completely. But the interesting par
/archives/netdev/2005-02/msg00076.html (9,731 bytes)

5. Re: TCP-Protection is really a pain... (score: 1)
Author: Christian Schmid <webmaster@xxxxxxxxxxxxxx>
Date: Thu, 03 Feb 2005 18:04:24 +0100
Several users complain about downloads stalling at 99% as well... I was unable to solve this issue. Jeff Garzik wrote: Christian Schmid wrote: btw: Another problem I am experiencing is that downloads
/archives/netdev/2005-02/msg00084.html (10,171 bytes)

6. Re: TCP-Protection is really a pain... (score: 1)
Author: Stephen Hemminger <shemminger@xxxxxxxx>
Date: Thu, 3 Feb 2005 11:33:35 -0800
Are you using a board that support TCP Segmentation Offload. The problem may well be that before we were not doing congestion control properly with TSO. A pre-2.6.8 host with TSO was violating all so
/archives/netdev/2005-02/msg00090.html (10,132 bytes)

7. Re: TCP-Protection is really a pain... (score: 1)
Author: Christian Schmid <webmaster@xxxxxxxxxxxxxx>
Date: Thu, 03 Feb 2005 20:58:48 +0100
Hm how can I check that? and what do you mean with "board"? Mainboard? NIC? its an onboard-nic on this mainboard: http://www.supermicro.com/products/motherboard/Xeon/E7501/X5DP8-G2.cfm The worst thin
/archives/netdev/2005-02/msg00092.html (11,123 bytes)

8. Re: TCP-Protection is really a pain... (score: 1)
Author: Stephen Hemminger <shemminger@xxxxxxxx>
Date: Thu, 3 Feb 2005 12:15:03 -0800
E1000 support TSO. To check if it is enabled do: ethtool -k eth0 To turn it off use. ethtool -K eth0 tso off You get the idea.. -- Stephen Hemminger <shemminger@xxxxxxxx>
/archives/netdev/2005-02/msg00095.html (9,273 bytes)

9. Re: TCP-Protection is really a pain... (score: 1)
Author: Christian Schmid <webmaster@xxxxxxxxxxxxxx>
Date: Thu, 03 Feb 2005 21:19:19 +0100
Offload parameters for eth0: tx-checksumming: on scatter-gather: on tcp segmentation offload: on What exactly is tcp segmentation offload? Where can I read more about it? Should I disable it or is th
/archives/netdev/2005-02/msg00096.html (9,899 bytes)

10. Re: TCP-Protection is really a pain... (score: 1)
Author: Rick Jones <rick.jones2@xxxxxx>
Date: Thu, 03 Feb 2005 12:47:14 -0800
Christian Schmid wrote: Offload parameters for eth0: rx-checksumming: on tx-checksumming: on scatter-gather: on tcp segmentation offload: on What exactly is tcp segmentation offload? TCP Segmentation
/archives/netdev/2005-02/msg00099.html (12,205 bytes)

11. Re: TCP-Protection is really a pain... (score: 1)
Author: Jeff Garzik <jgarzik@xxxxxxxxx>
Date: Thu, 03 Feb 2005 17:13:26 -0500
Christian Schmid wrote: What exactly is tcp segmentation offload? Where can I read more about it? Should I disable it or is this not a good idea? It's an optimization designed to reduce CPU usage for
/archives/netdev/2005-02/msg00102.html (9,869 bytes)

12. Re: TCP-Protection is really a pain... (score: 1)
Author: Christian Schmid <webmaster@xxxxxxxxxxxxxx>
Date: Thu, 03 Feb 2005 23:29:10 +0100
I turned it off and without it, I was unable to reproduce any stalls anymore. Jeff Garzik wrote: Christian Schmid wrote: What exactly is tcp segmentation offload? Where can I read more about it? Shou
/archives/netdev/2005-02/msg00104.html (10,276 bytes)

13. TCP-Protection is really a pain... (score: 1)
Author: Christian Schmid <webmaster@xxxxxxxxxxxxxx>
Date: Thu, 03 Feb 2005 05:07:30 +0100
Hi. Your new dynamically adjusted socket-buffer in 2.6.10 is really a pain for big servers. PLEASE tell me a way how to disable it. Thank you. Best regards, Christian Schmid
/archives/netdev/2005-02/msg01093.html (8,027 bytes)

14. Re: TCP-Protection is really a pain... (score: 1)
Author: Stephen Hemminger <shemminger@xxxxxxxx>
Date: Wed, 2 Feb 2005 20:54:37 -0800
sysctl -w net.ipv4.tcp_wmem="4096 8192 16384" Your single stream will be slower, but the memory footprint will be smaller.
/archives/netdev/2005-02/msg01094.html (8,764 bytes)

15. Re: TCP-Protection is really a pain... (score: 1)
Author: Christian Schmid <webmaster@xxxxxxxxxxxxxx>
Date: Thu, 03 Feb 2005 06:21:46 +0100
Actually, thats my problem. Single streams are too slow! Before I had buffers up to 500 KB. This was very nice to CPU because I only needed to "push" more data once in 5 seconds. I am doing this ever
/archives/netdev/2005-02/msg01095.html (10,344 bytes)

16. Re: TCP-Protection is really a pain... (score: 1)
Author: Jeff Garzik <jgarzik@xxxxxxxxx>
Date: Thu, 03 Feb 2005 01:50:45 -0500
Christian Schmid wrote: btw: Another problem I am experiencing is that downloads suddenly break in speed from 360 kb/sec to 8-12 kb/sec. 5 seconds later they stall completely. But the interesting par
/archives/netdev/2005-02/msg01096.html (9,831 bytes)

17. Re: TCP-Protection is really a pain... (score: 1)
Author: Christian Schmid <webmaster@xxxxxxxxxxxxxx>
Date: Thu, 03 Feb 2005 18:04:24 +0100
Several users complain about downloads stalling at 99% as well... I was unable to solve this issue. btw: Another problem I am experiencing is that downloads suddenly break in speed from 360 kb/sec to
/archives/netdev/2005-02/msg01104.html (10,289 bytes)

18. Re: TCP-Protection is really a pain... (score: 1)
Author: Stephen Hemminger <shemminger@xxxxxxxx>
Date: Thu, 3 Feb 2005 11:33:35 -0800
Are you using a board that support TCP Segmentation Offload. The problem may well be that before we were not doing congestion control properly with TSO. A pre-2.6.8 host with TSO was violating all so
/archives/netdev/2005-02/msg01110.html (10,235 bytes)

19. Re: TCP-Protection is really a pain... (score: 1)
Author: Christian Schmid <webmaster@xxxxxxxxxxxxxx>
Date: Thu, 03 Feb 2005 20:58:48 +0100
Hm how can I check that? and what do you mean with "board"? Mainboard? NIC? its an onboard-nic on this mainboard: http://www.supermicro.com/products/motherboard/Xeon/E7501/X5DP8-G2.cfm The worst thin
/archives/netdev/2005-02/msg01112.html (11,369 bytes)

20. Re: TCP-Protection is really a pain... (score: 1)
Author: Stephen Hemminger <shemminger@xxxxxxxx>
Date: Thu, 3 Feb 2005 12:15:03 -0800
E1000 support TSO. To check if it is enabled do: ethtool -k eth0 To turn it off use. ethtool -K eth0 tso off You get the idea.. -- Stephen Hemminger <shemminger@xxxxxxxx>
/archives/netdev/2005-02/msg01115.html (9,431 bytes)


This search system is powered by Namazu