Search String: Display: Description: Sort:

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*many\s+outgoing\s+tcp\s+sockets\s+are\s+slower\s+than\s+a\s+few\s*$/: 22 ]

Total 22 documents matching your query.

1. many outgoing tcp sockets are slower than a few (score: 1)
Author: xxx>
Date: Mon, 21 Feb 2005 10:01:21 +0100
Incoming our outgoing data mostly? I'm a bit confused, it is a download service so you are probably *sending* data? So? You could be a bit more helpful. Sample over 5 seconds, you won't overflow that
/archives/netdev/2005-02/msg00721.html (11,222 bytes)

2. Re: many outgoing tcp sockets are slower than a few (score: 1)
Author: xxxxxxx>
Date: Mon, 21 Feb 2005 11:36:14 +0100
bert hubert wrote: On Mon, Feb 21, 2005 at 01:35:33AM +0100, Christian Schmid wrote: New connections get made without any problems. Just existing connections slow down painfully. Incoming our outgoin
/archives/netdev/2005-02/msg00722.html (14,544 bytes)

3. Re: many outgoing tcp sockets are slower than a few (score: 1)
Author: 藤英明 <yoshfuji@xxxxxxxxxxxxxx>
Date: Mon, 21 Feb 2005 13:02:23 +0100
O_NONBLOCK send() is really nonblocking, but O_NONBLOCK sendfile() really isn't, as it still does the disk read (if any) synchronously. How are you making sure that you're sending "pure-cache data"?
/archives/netdev/2005-02/msg00725.html (10,141 bytes)

4. Re: many outgoing tcp sockets are slower than a few (score: 1)
Author: is Romieu <romieu@xxxxxxxxxxxxx>
Date: Mon, 21 Feb 2005 13:25:41 +0100
It is? So it returns EAGAIN if the data to be sent is not in the page cache? -- http://www.PowerDNS.com Open source, database driven DNS Software http://netherlabs.nl Open and Closed source services
/archives/netdev/2005-02/msg00726.html (10,616 bytes)

5. Re: many outgoing tcp sockets are slower than a few (score: 1)
Author: xxxxx>
Date: Mon, 21 Feb 2005 13:36:53 +0100
No, it just blocks. It only returns EAGAIN if the data is in the page cache but the socket TX queue doesn't have enough space. (At least, when I last tried it.) --L
/archives/netdev/2005-02/msg00727.html (10,133 bytes)

6. Re: many outgoing tcp sockets are slower than a few (score: 1)
Author: <ak@xxxxxx>
Date: Mon, 21 Feb 2005 13:59:19 +0000
Christian Schmid wrote: bert hubert wrote: On Mon, Feb 21, 2005 at 01:35:33AM +0100, Christian Schmid wrote: Outgoing data. I am using sendfile() to send the file on a non-blocking socket but the cal
/archives/netdev/2005-02/msg00730.html (11,156 bytes)

7. Re: many outgoing tcp sockets are slower than a few (score: 1)
Author: ndy.Dunlap" <rddunlap@xxxxxxxx>
Date: Mon, 21 Feb 2005 18:17:16 +0100
Lennert Buytenhek wrote: On Mon, Feb 21, 2005 at 11:36:14AM +0100, Christian Schmid wrote: Outgoing data. I am using sendfile() to send the file on a non-blocking socket but the call blocks for 100 m
/archives/netdev/2005-02/msg00749.html (11,108 bytes)

8. Re: many outgoing tcp sockets are slower than a few (score: 1)
Author: p" <rddunlap@xxxxxxxx>
Date: Mon, 21 Feb 2005 18:24:58 +0100
Do you poll/select for socket readiness before doing write() with EAGAIN? (shouldn't matter all that much, but could narrow down debugging) 100ms is a bit of an odd number - how big is your write (in
/archives/netdev/2005-02/msg00750.html (11,246 bytes)

9. Re: many outgoing tcp sockets are slower than a few (score: 1)
Author: mas Graf <tgraf@xxxxxxx>
Date: Mon, 21 Feb 2005 18:29:43 +0100
'syswrite'. Is your application written in C or perl? --L
/archives/netdev/2005-02/msg00751.html (10,881 bytes)

10. Re: many outgoing tcp sockets are slower than a few (score: 1)
Author: xxxx>
Date: Mon, 21 Feb 2005 20:11:30 +0100
Lennert Buytenhek wrote: On Mon, Feb 21, 2005 at 06:17:16PM +0100, Christian Schmid wrote: Outgoing data. I am using sendfile() to send the file on a non-blocking socket but the call blocks for 100 m
/archives/netdev/2005-02/msg00753.html (11,646 bytes)

11. Re: many outgoing tcp sockets are slower than a few (score: 1)
Author: id <webmaster@xxxxxxxxxxxxxx>
Date: Mon, 21 Feb 2005 20:10:36 +0100
bert hubert wrote: On Mon, Feb 21, 2005 at 06:17:16PM +0100, Christian Schmid wrote: Because thats the first I excluded. I changed the program by replacing sendfile with a caching-routine plus syswri
/archives/netdev/2005-02/msg00754.html (12,691 bytes)

12. many outgoing tcp sockets are slower than a few (score: 1)
Author: bert hubert <ahu@xxxxxxx>
Date: Mon, 21 Feb 2005 10:01:21 +0100
Incoming our outgoing data mostly? I'm a bit confused, it is a download service so you are probably *sending* data? So? You could be a bit more helpful. Sample over 5 seconds, you won't overflow that
/archives/netdev/2005-02/msg01741.html (11,348 bytes)

13. Re: many outgoing tcp sockets are slower than a few (score: 1)
Author: Christian Schmid <webmaster@xxxxxxxxxxxxxx>
Date: Mon, 21 Feb 2005 11:36:14 +0100
New connections get made without any problems. Just existing connections slow down painfully. Incoming our outgoing data mostly? Outgoing data. I am using sendfile() to send the file on a non-blockin
/archives/netdev/2005-02/msg01742.html (14,825 bytes)

14. Re: many outgoing tcp sockets are slower than a few (score: 1)
Author: Lennert Buytenhek <buytenh@xxxxxxxxxxxxxx>
Date: Mon, 21 Feb 2005 13:02:23 +0100
O_NONBLOCK send() is really nonblocking, but O_NONBLOCK sendfile() really isn't, as it still does the disk read (if any) synchronously. How are you making sure that you're sending "pure-cache data"?
/archives/netdev/2005-02/msg01745.html (10,290 bytes)

15. Re: many outgoing tcp sockets are slower than a few (score: 1)
Author: bert hubert <ahu@xxxxxxx>
Date: Mon, 21 Feb 2005 13:25:41 +0100
It is? So it returns EAGAIN if the data to be sent is not in the page cache? -- http://www.PowerDNS.com Open source, database driven DNS Software http://netherlabs.nl Open and Closed source services
/archives/netdev/2005-02/msg01746.html (10,831 bytes)

16. Re: many outgoing tcp sockets are slower than a few (score: 1)
Author: Lennert Buytenhek <buytenh@xxxxxxxxxxxxxx>
Date: Mon, 21 Feb 2005 13:36:53 +0100
No, it just blocks. It only returns EAGAIN if the data is in the page cache but the socket TX queue doesn't have enough space. (At least, when I last tried it.) --L
/archives/netdev/2005-02/msg01747.html (10,348 bytes)

17. Re: many outgoing tcp sockets are slower than a few (score: 1)
Author: Baruch Even <baruch@xxxxxxxxx>
Date: Mon, 21 Feb 2005 13:59:19 +0000
On Mon, Feb 21, 2005 at 01:35:33AM +0100, Christian Schmid wrote: Outgoing data. I am using sendfile() to send the file on a non-blocking socket but the call blocks for 100 ms per socket if I get ov
/archives/netdev/2005-02/msg01750.html (11,613 bytes)

18. Re: many outgoing tcp sockets are slower than a few (score: 1)
Author: Christian Schmid <webmaster@xxxxxxxxxxxxxx>
Date: Mon, 21 Feb 2005 18:17:16 +0100
Outgoing data. I am using sendfile() to send the file on a non-blocking socket but the call blocks for 100 ms per socket if I get over 3000 sockets. Thats causing the massive slowdown in sum. I first
/archives/netdev/2005-02/msg01769.html (11,345 bytes)

19. Re: many outgoing tcp sockets are slower than a few (score: 1)
Author: bert hubert <ahu@xxxxxxx>
Date: Mon, 21 Feb 2005 18:24:58 +0100
Do you poll/select for socket readiness before doing write() with EAGAIN? (shouldn't matter all that much, but could narrow down debugging) 100ms is a bit of an odd number - how big is your write (in
/archives/netdev/2005-02/msg01770.html (11,479 bytes)

20. Re: many outgoing tcp sockets are slower than a few (score: 1)
Author: Lennert Buytenhek <buytenh@xxxxxxxxxxxxxx>
Date: Mon, 21 Feb 2005 18:29:43 +0100
'syswrite'. Is your application written in C or perl? --L
/archives/netdev/2005-02/msg01771.html (11,084 bytes)


This search system is powered by Namazu