| From | Sent On | Attachments |
|---|---|---|
| Asankha C. Perera | Oct 29, 2012 6:20 am | |
| Christopher Schultz | Oct 29, 2012 9:12 am | |
| Asankha C. Perera | Oct 29, 2012 9:28 am | |
| Christopher Schultz | Oct 29, 2012 2:21 pm | |
| Asankha C. Perera | Oct 29, 2012 8:56 pm | |
| Christopher Schultz | Oct 31, 2012 8:56 am | |
| Asankha C. Perera | Nov 1, 2012 12:38 am | |
| Esmond Pitt | Nov 1, 2012 3:27 pm | |
| Asankha C. Perera | Nov 2, 2012 12:12 pm | |
| Mark Thomas | Nov 2, 2012 12:51 pm | |
| Christopher Schultz | Nov 2, 2012 1:30 pm | |
| Asankha C. Perera | Nov 5, 2012 5:36 am | |
| Christopher Schultz | Nov 5, 2012 3:25 pm | |
| Asankha C. Perera | Nov 5, 2012 11:56 pm | |
| Esmond Pitt | Nov 6, 2012 5:17 pm | |
| Asankha C. Perera | Nov 6, 2012 7:24 pm | |
| Asankha C. Perera | Nov 6, 2012 8:01 pm | |
| Mark Thomas | Nov 6, 2012 10:22 pm | |
| Mark Thomas | Nov 6, 2012 10:25 pm | |
| Asankha C. Perera | Nov 6, 2012 11:12 pm | |
| Mark Thomas | Nov 6, 2012 11:37 pm | |
| Asankha C. Perera | Nov 7, 2012 7:18 am | |
| Esmond Pitt | Nov 7, 2012 3:27 pm | |
| Asankha C. Perera | Nov 7, 2012 6:51 pm | |
| Esmond Pitt | Nov 7, 2012 7:02 pm | |
| Asankha C. Perera | Nov 7, 2012 7:35 pm | |
| Mark Thomas | Nov 8, 2012 1:49 am | |
| Asankha C. Perera | Nov 8, 2012 7:02 am | |
| Pid | Nov 8, 2012 12:46 pm | |
| Asankha C. Perera | Nov 8, 2012 9:11 pm | |
| Meik Suchlich | Nov 8, 2012 10:40 pm | |
| Konstantin Kolinko | Nov 9, 2012 12:03 am | |
| Meik Suchlich | Nov 9, 2012 12:39 am | |
| Pid | Nov 9, 2012 12:48 am | |
| Caldarale, Charles R | Nov 9, 2012 6:11 am | |
| Tim Watts | Nov 9, 2012 7:10 am | |
| André Warnier | Nov 9, 2012 7:16 am | |
| Konstantin Kolinko | Nov 9, 2012 10:13 am | |
| Christopher Schultz | Nov 9, 2012 11:41 am | |
| Asankha C. Perera | Nov 9, 2012 8:52 pm | |
| Pid | Nov 11, 2012 6:14 am | |
| Esmond Pitt | Nov 11, 2012 7:18 pm | |
| Terence M. Bandoian | Nov 12, 2012 9:16 am | |
| Christopher Schultz | Nov 12, 2012 11:31 am | |
| Asankha C. Perera | Nov 12, 2012 2:30 pm | |
| Asankha C. Perera | Nov 12, 2012 2:36 pm | |
| Mark Thomas | Nov 12, 2012 4:10 pm | |
| Mark Thomas | Nov 12, 2012 5:02 pm | |
| selvakumar netaji | Nov 12, 2012 6:32 pm | |
| Caldarale, Charles R | Nov 12, 2012 7:07 pm | |
| Esmond Pitt | Nov 13, 2012 3:16 pm | |
| chris derham | Nov 14, 2012 8:01 am | |
| Christopher Schultz | Nov 14, 2012 1:05 pm | |
| Asankha C. Perera | Nov 14, 2012 9:47 pm |
| Subject: | Handling requests when under load - ACCEPT and RST vs non-ACCEPT | |
|---|---|---|
| From: | Asankha C. Perera (asan...@apache.org) | |
| Date: | Oct 29, 2012 6:20:03 am | |
| List: | org.apache.tomcat.users | |
Hi All
During some performance testing I've seen that Tomcat resets accepted TCP connections when under load. I had seen this previously too [1], but was not able to analyze the scenario in detail earlier.
As per this dump from Wireshark [2], it seemed like Tomcat ACKed the client request, accepted part of the request, and then suddenly decided to close the connection and hence RSTed it. What I would expect Tomcat to have done instead is to refuse a connection when under load, and not accept and RST. The problem occurs for a client that would not know if a RSTed connection could be safely retried. If the connection was not accepted, a fail-over is straight forward.
Hope to hear some details from the developer community, to understand this behavior better
regards asankha
[1] http://markmail.org/message/v7cpj6oqumtn5gtp [2] http://troll.ws/image/6b38f283
-- Asankha C. Perera AdroitLogic, http://adroitlogic.org





