[20200219]strace跟蹤設定ENABLE=BROKEN的情況(網路的情況).txt
[20200219]strace跟蹤設定ENABLE=BROKEN的情況(網路的情況).txt
--//前面的測試都是在本地服務端進行，使用lo介面。測試真實網路的情況：
1.環境：
SCOTT@book> @ ver1
PORT_STRING VERSION BANNER
------------------------------ -------------- --------------------------------------------------------------------------------
x86_64/Linux 2.4.xx 11.2.0.4.0 Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production
# echo /proc/sys/net/ipv4/tcp_keepalive* | xargs -n 1 strings -1 -f
/proc/sys/net/ipv4/tcp_keepalive_intvl: 10
/proc/sys/net/ipv4/tcp_keepalive_probes: 4
/proc/sys/net/ipv4/tcp_keepalive_time: 33
$ grep SQLNET.EXPIRE_TIME $ORACLE_HOME/network/admin/sqlnet.ora
# SQLNET.EXPIRE_TIME = 1
--//客戶端：
# echo /proc/sys/net/ipv4/tcp_keepalive* | xargs -n 1 strings -1 -f
/proc/sys/net/ipv4/tcp_keepalive_intvl: 15
/proc/sys/net/ipv4/tcp_keepalive_probes: 4
/proc/sys/net/ipv4/tcp_keepalive_time: 56
$ grep -i expire sqlnet.ora
#SQLNET.EXPIRE_TIME = 1
2.測試：
--//在客戶端執行：
$ strace -f -r -T -e trace=socket,setsockopt sqlplus scott/book@"(DESCRIPTION=(ENABLE=BROKEN)(CONNECT_DATA=(SERVICE_NAME=book)(SERVER = DEDICATED))(ADDRESS=(PROTOCOL=tcp)(HOST=192.168.100.78)(PORT=1521)))"
0.000000 socket(PF_INET6, SOCK_DGRAM, IPPROTO_IP) = 6 <0.000236>
0.001417 socket(PF_FILE, SOCK_STREAM, 0) = 6 <0.000143>
0.000447 socket(PF_FILE, SOCK_STREAM, 0) = 6 <0.000149>
SQL*Plus: Release 11.2.0.4.0 Production on Wed Feb 19 11:00:19 2020
Copyright (c) 1982, 2013, Oracle. All rights reserved.
0.021572 socket(PF_INET, SOCK_STREAM, IPPROTO_IP) = 9 <0.000179>
0.000346 setsockopt(9, SOL_SOCKET, SO_KEEPALIVE, [1], 4) = 0 <0.000137>
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
0.000863 setsockopt(9, SOL_TCP, TCP_NODELAY, [1], 4) = 0 <0.000142>
0.054498 setsockopt(9, SOL_SOCKET, SO_KEEPALIVE, [1], 4) = 0 <0.000149>
0.010003 socket(PF_NETLINK, SOCK_RAW, 0) = 10 <0.000211>
0.000690 socket(PF_FILE, SOCK_STREAM, 0) = 10 <0.000148>
0.000437 socket(PF_FILE, SOCK_STREAM, 0) = 10 <0.000149>
0.001555 socket(PF_INET, SOCK_DGRAM, IPPROTO_IP) = 10 <0.000032>
SCOTT@book> @ spid
SID SERIAL# PROCESS SERVER SPID PID P_SERIAL# C50
---------- ---------- ------------------------ --------- ------ ------- ---------- --------------------------------------------------
3 1435 14573 DEDICATED 10567 24 217 alter system kill session '3,1435' immediate;
--//在服務端執行：
# ps -ef | grep tnslsn[r]
oracle 47619 1 0 Jan15 ? 00:01:04 /u01/app/oracle/product/11.2.0.4/dbhome_1/bin/tnslsnr LISTENER -inherit
$ strace -f -r -T -e trace=socket,setsockopt -p 47619
Process 47619 attached with 3 threads - interrupt to quit
[pid 47620] 0.000000 socket(PF_INET, SOCK_STREAM, IPPROTO_TCP) = 13 <0.000037>
[pid 47619] 3.741731 setsockopt(13, SOL_TCP, TCP_NODELAY, [1], 4) = 0 <0.000020>
Process 10566 attached
Process 47619 suspended
Process 10567 attached
Process 47619 resumed
Process 10566 detached
[pid 47619] 0.002848 --- SIGCHLD (Child exited) @ 0 (0) ---
[pid 10567] 0.035160 socket(PF_INET6, SOCK_DGRAM, IPPROTO_IP) = 6 <0.000027>
[pid 10567] 0.007805 socket(PF_FILE, SOCK_STREAM, 0) = 4 <0.000025>
[pid 10567] 0.000210 socket(PF_FILE, SOCK_STREAM, 0) = 4 <0.000019>
[pid 10567] 0.006365 setsockopt(13, SOL_TCP, TCP_NODELAY, [1], 4) = 0 <0.000025>
[pid 10567] 0.001532 setsockopt(13, SOL_SOCKET, SO_KEEPALIVE, [1], 4) = 0 <0.000023>
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
[pid 47620] 1.207357 socket(PF_INET, SOCK_STREAM, IPPROTO_TCP) = 13 <0.000027>
[pid 47620] 5.000959 socket(PF_INET, SOCK_STREAM, IPPROTO_TCP) = 13 <0.000029>
[pid 47620] 5.001582 socket(PF_INET, SOCK_STREAM, IPPROTO_TCP) = 13 <0.000026>
[pid 47620] 5.001408 socket(PF_INET, SOCK_STREAM, IPPROTO_TCP) = 13 <0.000031>
[pid 47620] 5.002872 socket(PF_INET, SOCK_STREAM, IPPROTO_TCP) = 13 <0.000026>
[pid 47620] 5.000991 socket(PF_INET, SOCK_STREAM, IPPROTO_TCP) = 13 <0.000027>
[pid 47620] 5.001649 socket(PF_INET, SOCK_STREAM, IPPROTO_TCP) = 13 <0.000027>
[pid 47620] 5.001450 socket(PF_INET, SOCK_STREAM, IPPROTO_TCP) = 13 <0.000038>
^CProcess 47619 detached
Process 47620 detached
Process 47621 detached
Process 10567 detached
--//可以看出服務端與客戶端都執行了setsockopt(13, SOL_SOCKET, SO_KEEPALIVE, [1], 4).
3.看看連線情況以及各自keepalive設定是多少？
--//在客戶端執行：
$ seq 100 | xargs -iQ bash -c "netstat -npo 2>/dev/null | grep 14573|ts.awk;sleep 0.99"
--//rh 4.3版本，xargs -i引數是小寫。
[2020-02-19 11:04:06] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (0.46/0/0)
[2020-02-19 11:04:07] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (14.41/0/0)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
[2020-02-19 11:04:08] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (13.37/0/0)
[2020-02-19 11:04:09] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (12.32/0/0)
[2020-02-19 11:04:10] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (11.28/0/0)
[2020-02-19 11:04:12] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (10.23/0/0)
[2020-02-19 11:04:13] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (9.19/0/0)
[2020-02-19 11:04:14] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (8.14/0/0)
[2020-02-19 11:04:15] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (7.10/0/0)
[2020-02-19 11:04:16] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (6.05/0/0)
[2020-02-19 11:04:17] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (5.01/0/0)
[2020-02-19 11:04:18] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (3.96/0/0)
[2020-02-19 11:04:19] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (2.92/0/0)
[2020-02-19 11:04:20] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (1.87/0/0)
[2020-02-19 11:04:21] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (0.83/0/0)
[2020-02-19 11:04:22] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (40.78/0/0)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
[2020-02-19 11:04:23] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (39.73/0/0)
[2020-02-19 11:04:24] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (38.69/0/0)
[2020-02-19 11:04:25] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (37.64/0/0)
[2020-02-19 11:04:26] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (36.60/0/0)
[2020-02-19 11:04:27] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (35.55/0/0)
[2020-02-19 11:04:28] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (34.51/0/0)
[2020-02-19 11:04:29] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (33.46/0/0)
[2020-02-19 11:04:30] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (32.42/0/0)
[2020-02-19 11:04:31] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (31.37/0/0)
[2020-02-19 11:04:32] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (30.33/0/0)
[2020-02-19 11:04:33] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (29.28/0/0)
[2020-02-19 11:04:35] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (28.24/0/0)
[2020-02-19 11:04:36] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (27.19/0/0)
[2020-02-19 11:04:37] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (26.15/0/0)
[2020-02-19 11:04:38] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (25.10/0/0)
[2020-02-19 11:04:39] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (24.06/0/0)
[2020-02-19 11:04:40] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (23.01/0/0)
[2020-02-19 11:04:41] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (21.97/0/0)
[2020-02-19 11:04:42] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (20.92/0/0)
[2020-02-19 11:04:43] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (19.88/0/0)
[2020-02-19 11:04:44] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (18.84/0/0)
[2020-02-19 11:04:45] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (17.79/0/0)
[2020-02-19 11:04:46] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (16.75/0/0)
[2020-02-19 11:04:47] tcp 0 0 192.168.100.40:37596 192.168.100.78:1521 ESTABLISHED 14573/sqlplus keepalive (15.70/0/0)
--//也是分2部分 ，41+15 = 56秒.
--//在服務端執行：
$ seq 100 | xargs -IQ bash -c "netstat -npo 2>/dev/null | grep 10567|ts.awk;sleep 1"
[2020-02-19 11:06:36] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (0.41/0/0)
[2020-02-19 11:06:37] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (22.39/0/0)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
[2020-02-19 11:06:38] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (21.37/0/0)
[2020-02-19 11:06:39] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (20.35/0/0)
[2020-02-19 11:06:40] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (19.32/0/0)
[2020-02-19 11:06:41] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (18.30/0/0)
[2020-02-19 11:06:42] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (17.28/0/0)
[2020-02-19 11:06:43] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (16.26/0/0)
[2020-02-19 11:06:44] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (15.23/0/0)
[2020-02-19 11:06:45] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (14.21/0/0)
[2020-02-19 11:06:46] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (13.19/0/0)
[2020-02-19 11:06:47] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (12.17/0/0)
[2020-02-19 11:06:48] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (11.15/0/0)
[2020-02-19 11:06:49] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (10.12/0/0)
[2020-02-19 11:06:50] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (9.10/0/0)
[2020-02-19 11:06:51] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (8.08/0/0)
[2020-02-19 11:06:52] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (7.06/0/0)
[2020-02-19 11:06:53] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (6.04/0/0)
[2020-02-19 11:06:54] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (5.01/0/0)
[2020-02-19 11:06:55] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (3.99/0/0)
[2020-02-19 11:06:56] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (2.97/0/0)
[2020-02-19 11:06:57] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (1.95/0/0)
[2020-02-19 11:06:58] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (0.92/0/0)
[2020-02-19 11:06:59] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (9.92/0/0)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
[2020-02-19 11:07:00] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (8.89/0/0)
[2020-02-19 11:07:01] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (7.87/0/0)
[2020-02-19 11:07:02] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (6.85/0/0)
[2020-02-19 11:07:03] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (5.83/0/0)
[2020-02-19 11:07:04] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (4.81/0/0)
[2020-02-19 11:07:05] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (3.78/0/0)
[2020-02-19 11:07:06] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (2.76/0/0)
[2020-02-19 11:07:07] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (1.74/0/0)
[2020-02-19 11:07:08] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (0.72/0/0)
[2020-02-19 11:07:09] tcp 0 0 192.168.100.78:1521 192.168.100.40:37596 ESTABLISHED 10567/oraclebook keepalive (22.70/0/0)
--//10+23 = 33.
--//噢，終於明白為什麼分成2部分，1部分的設定來源於核心引數/proc/sys/net/ipv4/tcp_keepalive_intvl。
--//服務端/proc/sys/net/ipv4/tcp_keepalive_intvl=10,而客戶端/proc/sys/net/ipv4/tcp_keepalive_intvl=15.
--//可以看出setsockopt(N, SOL_SOCKET, SO_KEEPALIVE, [1], 4) ,SO_KEEPALIVE引數來自各自主機的核心引數tcp_keepalive_time。
--//如果設定SQLNET.EXPIRE_TIME 引數，一部分就是6，不知道為什麼？
3.另外我還測試：
--//客戶端修改SQLNET.EXPIRE_TIME = 1的情況：
$ grep -i expire sqlnet.ora
SQLNET.EXPIRE_TIME = 1
--//結果一樣，不會使用這個引數，而是使用核心引數。
4.補充使用tcpdump驗證看看:
--//從客戶端登入:
$ sqlplus scott/book@"(DESCRIPTION=(ENABLE=BROKEN)(CONNECT_DATA=(SERVICE_NAME=book)(SERVER = DEDICATED))(ADDRESS=(PROTOCOL=tcp)(HOST=192.168.100.78)(PORT=1521)))"
SCOTT@book> @ spid
SID SERIAL# PROCESS SERVER SPID PID P_SERIAL# C50
---------- ---------- ------------------------ --------- ------ ------- ---------- --------------------------------------------------
3 1589 19194 DEDICATED 13338 24 11 alter system kill session '3,1589' immediate;
--//注:不執行任何命令.
--//客戶端執行：
$ netstat -npo 2>/dev/null | grep 19194
tcp 0 0 192.168.100.40:40891 192.168.100.78:1521 ESTABLISHED 19194/sqlplus keepalive (20.73/0/0)
--//服務端執行：
# netstat -npo 2>/dev/null | grep 13338
tcp 0 0 192.168.100.78:1521 192.168.100.40:40891 ESTABLISHED 13338/oraclebook keepalive (13.46/0/0)
# tcpdump -vvnni eth0 dst port 40891
tcpdump: listening on eth0, link-type EN10MB (Ethernet), capture size 96 bytes
...
15:37:50.126104 IP (tos 0x0, ttl 64, id 18244, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.78.1521 > 192.168.100.40.40891: ., cksum 0x49ee (incorrect (-> 0xc606), 1:1(0) ack 1 win 331 <nop,nop,timestamp 713638269 2533232674>
15:37:57.424253 IP (tos 0x0, ttl 64, id 18245, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.78.1521 > 192.168.100.40.40891: ., cksum 0x49ee (incorrect (-> 0xa984), 0:0(0) ack 1 win 331 <nop,nop,timestamp 713645568 2533232674>
~~~~~
15:38:30.448234 IP (tos 0x0, ttl 64, id 18246, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.78.1521 > 192.168.100.40.40891: ., cksum 0x49ee (incorrect (-> 0xa782), 0:0(0) ack 1 win 331 <nop,nop,timestamp 713678592 2533265699>
~~~~
15:38:46.124136 IP (tos 0x0, ttl 64, id 18247, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.78.1521 > 192.168.100.40.40891: ., cksum 0x49ee (incorrect (-> 0xe943), 1:1(0) ack 1 win 331 <nop,nop,timestamp 713694267 2533298725>
15:39:03.472187 IP (tos 0x0, ttl 64, id 18248, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.78.1521 > 192.168.100.40.40891: ., cksum 0x49ee (incorrect (-> 0xa57f), 0:0(0) ack 1 win 331 <nop,nop,timestamp 713711616 2533298725>
~~~
15:39:36.496545 IP (tos 0x0, ttl 64, id 18249, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.78.1521 > 192.168.100.40.40891: ., cksum 0x49ee (incorrect (-> 0xa37d), 0:0(0) ack 1 win 331 <nop,nop,timestamp 713744640 2533331750>
~~~~~
15:39:42.122073 IP (tos 0x0, ttl 64, id 18250, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.78.1521 > 192.168.100.40.40891: ., cksum 0x49ee (incorrect (-> 0x0c82), 1:1(0) ack 1 win 331 <nop,nop,timestamp 713750265 2533364775>
15:40:09.520757 IP (tos 0x0, ttl 64, id 20500, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.40.40891 > 192.168.100.78.1521: ., cksum 0x8c3d (correct), 0:0(0) ack 1 win 145 <nop,nop,timestamp 2533397797 713750265>
~~~~~
15:40:38.138192 IP (tos 0x0, ttl 64, id 20502, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.40.40891 > 192.168.100.78.1521: ., cksum 0x1c82 (correct), 4294967295:4294967295(0) ack 1 win 145 <nop,nop,timestamp 2533426401 713750265>
15:40:42.545344 IP (tos 0x0, ttl 64, id 20504, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.40.40891 > 192.168.100.78.1521: ., cksum 0x307a (correct), 0:0(0) ack 1 win 145 <nop,nop,timestamp 2533430806 713806282>
~~~~~~
15:41:15.568301 IP (tos 0x0, ttl 64, id 20506, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.40.40891 > 192.168.100.78.1521: ., cksum 0xaf80 (correct), 0:0(0) ack 1 win 145 <nop,nop,timestamp 2533463823 713806282>
~~~~~~~
15:41:34.147020 IP (tos 0x0, ttl 64, id 20508, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.40.40891 > 192.168.100.78.1521: ., cksum 0x66ee (correct), 4294967295:4294967295(0) ack 1 win 145 <nop,nop,timestamp 2533482402 713806282>
15:41:48.592753 IP (tos 0x0, ttl 64, id 20510, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.40.40891 > 192.168.100.78.1521: ., cksum 0x53b5 (correct), 0:0(0) ack 1 win 145 <nop,nop,timestamp 2533496849 713862290>
~~~~~~~
15:42:21.616468 IP (tos 0x0, ttl 64, id 20512, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.40.40891 > 192.168.100.78.1521: ., cksum 0xd2b3 (correct), 0:0(0) ack 1 win 145 <nop,nop,timestamp 2533529874 713862290>
--//注意看下劃線的行，時間的間隔33秒。其它出現的情況是客戶端發出的探測包。比如時間 ：15:38:46.124136。對比下面的輸出就很清洗了。
--// 15:38:46.124136 ,15:39:42.122073 相差56秒。
# tcpdump -vvnni eth0 src port 40891
tcpdump: listening on eth0, link-type EN10MB (Ethernet), capture size 96 bytes
15:38:30.448364 IP (tos 0x0, ttl 64, id 20490, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.40.40891 > 192.168.100.78.1521: ., cksum 0xc4bc (correct), 2813590524:2813590524(0) ack 2927188646 win 145 <nop,nop,timestamp 2533298725 713638269>
15:38:46.123988 IP (tos 0x0, ttl 64, id 20492, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.40.40891 > 192.168.100.78.1521: ., cksum 0x8781 (correct), 4294967295:4294967295(0) ack 1 win 145 <nop,nop,timestamp 2533314401 713638269>
~~~~~~~~
15:39:03.472312 IP (tos 0x0, ttl 64, id 20494, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.40.40891 > 192.168.100.78.1521: ., cksum 0x68fc (correct), 0:0(0) ack 1 win 145 <nop,nop,timestamp 2533331750 713694267>
15:39:36.496655 IP (tos 0x0, ttl 64, id 20496, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.40.40891 > 192.168.100.78.1521: ., cksum 0xe7fa (correct), 0:0(0) ack 1 win 145 <nop,nop,timestamp 2533364775 713694267>
15:39:42.122031 IP (tos 0x0, ttl 64, id 20498, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.40.40891 > 192.168.100.78.1521: ., cksum 0xd201 (correct), 4294967295:4294967295(0) ack 1 win 145 <nop,nop,timestamp 2533370401 713694267>
~~~~~~~~
15:40:09.520757 IP (tos 0x0, ttl 64, id 20500, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.40.40891 > 192.168.100.78.1521: ., cksum 0x8c3d (correct), 0:0(0) ack 1 win 145 <nop,nop,timestamp 2533397797 713750265>
15:40:38.138192 IP (tos 0x0, ttl 64, id 20502, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.40.40891 > 192.168.100.78.1521: ., cksum 0x1c82 (correct), 4294967295:4294967295(0) ack 1 win 145 <nop,nop,timestamp 2533426401 713750265>
~~~~~~~~~
15:40:42.545344 IP (tos 0x0, ttl 64, id 20504, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.40.40891 > 192.168.100.78.1521: ., cksum 0x307a (correct), 0:0(0) ack 1 win 145 <nop,nop,timestamp 2533430806 713806282>
15:41:15.568301 IP (tos 0x0, ttl 64, id 20506, offset 0, flags [DF], proto: TCP (6), length: 52) 192.168.100.40.40891 > 192.168.100.78.1521: ., cksum 0xaf80 (correct), 0:0(0) ack 1 win 145 <nop,nop,timestamp 2533463823 713806282>
--//注意看下劃線的行，時間的間隔56秒。
5.總結：
--//11g下建議設定相關核心引數，不建議選擇增加 sqlnet.ora檔案的 SQLNET.EXPIRE_TIME引數。
# echo /proc/sys/net/ipv4/tcp_keepalive* | xargs -n 1 strings -1 -f
/proc/sys/net/ipv4/tcp_keepalive_intvl: 10
/proc/sys/net/ipv4/tcp_keepalive_probes: 4
/proc/sys/net/ipv4/tcp_keepalive_time: 600
--//12c選擇修改SQLNET.EXPIRE_TIME = 10
--//資料庫連線串建議加入(ENABLE=BROKEN)。注意客戶端相關核心引數也要修改合適,實際上就是使用核心相關引數檢測TCP KEEPALIVE套接
--//字選項來檢查連線是否仍然可用。
--//我建議如下:
# echo /proc/sys/net/ipv4/tcp_keepalive* | xargs -n 1 strings -1 -f
/proc/sys/net/ipv4/tcp_keepalive_intvl: 10
/proc/sys/net/ipv4/tcp_keepalive_probes: 4
/proc/sys/net/ipv4/tcp_keepalive_time: 600
--//不知道windows的這些引數如何修改登錄檔,有時間看看。
--//另外不必擔心這些包對網路效能的影響，現在的網路完全可以承受，還不如重點放在最佳化業務流程以及sql語句。
6.ts.awk指令碼：
$ cat /usr/local/bin/ts.awk
# /bin/bash
gawk '{ print strftime("[%Y-%m-%d %H:%M:%S]"), $0 }'
來自 “ ITPUB部落格 ” ，連結：http://blog.itpub.net/267265/viewspace-2676239/，如需轉載，請註明出處，否則將追究法律責任。
相關文章
- [20200219]strace跟蹤設定ENABLE=BROKEN的情況.txt
- iOS不同網路情況除錯iOS除錯
- 處理表鎖定的情況
- mysql索引失效的情況MySql索引
- NoClassDefFoundError的兩種情況Error
- eMarketer：印尼社會化網路發展情況
- eMarketer：亞太地區社交網路發展情況
- Mysql 可能鎖表的情況MySql
- Oracle其它情況的恢復Oracle
- DOM對映的特殊情況
- DorkBot殭屍網路近期活躍情況報告
- linux系統檢視網路連線情況Linux
- 如何處理瀏覽器的斷網情況？瀏覽器
- mysql索引不會命中的情況MySql索引
- oracle order by索引是否使用的情況Oracle索引
- 伺服器的連線情況？伺服器
- 批次查詢並設定極兔快遞延誤情況的方法
- 批次查詢並設定中通快遞延誤情況的方法
- 高階網站建設中出現的一些特殊情況網站
- 測試在cursor_sharing為force情況下執行計劃的共享情況
- 城市天氣情況APIAPI
- switch不加break情況分析
- metricbeat 監控 nginx 情況Nginx
- Linux TCP RST情況LinuxTCP
- ORACLE索引被抑制情況Oracle索引
- 查詢鎖等待情況
- 05記憶體情況記憶體
- 維護無線網路安全五種必須避免的情況(轉)
- 程式設計師修bug時的真實情況程式設計師
- 目前程式設計師的5中情況程式設計師
- 資料庫一些預設的屬性及情況設定查詢資料庫
- if 判斷為 false 的 6 種情況False
- 美國各地口罩佩戴情況的地圖地圖
- String不同定義的情況解析
- 記一次使用Oauth的情況OAuth
- Mysql 會導致索引失效的情況MySql索引
- 檢視ASM的Extent分佈情況ASM
- 關於PHPMailer漏洞情況的通報PHPAI