我们在ASP.NET应用程序中使用ERPConnect,但有时当用户尝试通过我们的应用程序连接到SAP时,会收到一些错误消息。
CPIC-CALL:'CMRCV:rc = 19位置主机上的SAP-Gateway SERVERNAME / sapgw10 ERROR会话53230591未找到时间2011年11月16日星期三13:49:58发布701组件SAP-Gateway版本2 RC 728 MODULE gwxxrd。 c LINE 6278 COUNTER 1277701
通过ERPConnect打开与SAP的连接时总会发生这种情况。
R3Connection connection = Connection;
connection.Open(); // the error occurs here
这很随机。有些日子我们没有收到任何错误消息,其他日子我们每天收到20条消息。 ERPConnect使用System32文件夹中定义的librfc32.dll,版本为7200.1.39.6461(这是最新版本)。
我的本地开发计算机(Windows 7 x86)上未发生此问题。它始终出现在我们的生产服务器(Windows Server 2008 R2)上。
知道什么可能导致这个问题吗?
编辑(添加了错误日志提取):
* * LOCATION SAP-Gateway on host HERCULES / sapgw10 * ERROR partner 'hercules:sapgw01' not reached * * TIME Thu Nov 24 09:10:20 2011 * RELEASE 701 * COMPONENT NI (network interface) * VERSION 38 * RC -10 * MODULE nixxi.cpp * LINE 2835 * DETAIL NiPConnect2 * SYSTEM CALL connect * ERRNO 10061 * ERRNO TEXT WSAECONNREFUSED: Connection refused * COUNTER 1385122 * *****************************************************************************
GwTraceHdlInfo:
HANDLE = 137
TIME = Thu Nov 24 09:10:19 2011
SOCKET = 2108
STAT = NI_CONN_WAIT
TYPE = STREAM IPv4
OUT = 0 messages 0 bytes
IN = 0 messages 0 bytes
LOCAL = 0.0.0.0:1865
REMOTE = -
OPTIONS = BUFFERED
ni hdl = 137
type = CLIENT
net_stat = CONNECT_TO_REM_GW_PENDING
hostaddr = 100.100.103.15
accept_hostaddr = 100.100.103.15
opcode = NORMAL_CLIENT
conn opcode = REMOTE_GATEWAY
conn vers = 2
index = 10
data = 0000000000000000
ext_info = 000000000EF1D870
offset = 0
rest_len = 0
snc_forced = 0
remote gateway infos:
hostname =
hostaddr = 100.100.103.15
service = sapgw01
tpname = sapdp01
GwSelClear: clear RWC for hdl 137
NiICloseHandle: called for hdl 137 while waiting for connection
NiICloseHandle: shutdown and close hdl 137 / sock 2108
GwFreeHdlInfoMemory: free 000000000EF1D870 (ext info)
GwIFreeMemForLU_TP: freed memory for sys 10
GwListRemove: remove elem 10 from sys_r3_list
GwListRemove: 2 elems in sys_r3_list
GwListRemove: elem 10 not in sys_reg_list
GwListRemove: remove elem 10 from sys_inuse_list
GwListRemove: 104 elems in sys_inuse_list
GwUpdateClient: act_sys: 104
GwIDelR3Conn: send error message to client
GwSendRc3: send (appc_rc=CM_PRODUCT_SPECIFIC_ERROR / sap_rc=GW_CONNECT_FAILED) to client HERCULES / sapgw10 (0)
GwRqDpSendTo: (HERCULES / sapgw10, gw_id=0, appc_ca_blk=-1, len=335, comm_index=-1)
LOCK APPC ca_blk 3
DpGetCpicCommIdx: found index 3 via appc_hdr
found comm entry 3 (tid/uid/mode/conv_id/a_r: 158/13268/0/ /0)
GwIsAWaitingSWP: wp 1 waiting
GwSapWpWakeUp: send answer to sap wp 1
GwSapWpWakeUp: make wp 1 owner of appc_ca_blk 3
-OUT- tid -1 uid -1 mode 255
-OUT- appc_ca_blk 3 len 335 rq_id 43133
-OUT- wait_f_answ 0 cmd 0
NiIWrite: hdl 109 sent data (wrt=28,pac=1,MESG_IO)
GwIDelR3Conn: decrement conv_no of client 0: 5
GwFreeMemory: free 000000000EEF3480 (len=3700)
GwFreeMemory: act_overflow_size = 20352 (- 3700)
GwIDelR3Conn: idx/conv = ((-1/0) | (10/0)), delete conv 48 from conv-table
GwClearConn: conv_no/tcp_conv_no/sna_conv_no: 6/6/0
GwClearConn: free r3 conv info
GwClearConn: free buffer info
GwFiSearchConvId: deleted 27077110 local, conn=48
GwListInsert: insert elem 48 into conn_free_list (at begin)
GwListInsert: 494 elems in conn_free_list
GwListRemove: remove elem 48 from conn_search_list
GwListRemove: 0 elems in conn_search_list
GwListRemove: elem 48 not in conn_write_list
GwListRemove: remove elem 48 from conn_inuse_list
GwListRemove: 6 elems in conn_inuse_list
GwReadReqIndex: (process_id: -1)
GwReadReqIndex: no filled request-index-block
GwCheckClient: check clients
LOCK APPC ca_blk 4
GwCheckClient: raise timeout for conv 27059938 (intr/uid/mode/comm_idx=9/13261/0/90)
GwSendRcToDp: send appc_rc/sap_rc 10001/221 to dp
make DISP owner of appc_ca_blk 4
DpSemRq: key: 3, units: 1, timeout: -1
DpSemRel: key: 3, units: 1
DpRqPutIntoQueue: put request into queue (reqtype 0, prio LOW, rq_id 43134)
GwDispWakeUp: send wakeup with ni handle 2
NiIWrite: hdl 2 sent data (wrt=1,pac=1,MESG_IO)
-OUT- sender_id APPC_SERVER tid 165 wp_ca_blk -1 wp_id -1
-OUT- action SEND_TO_WP uid 13261 appc_ca_blk 4 type NOWP
-OUT- new_stat NO_CHANGE mode 0 len 80 rq_id 43134
GwListRemove: remove elem 90 from comm_wait_list
GwListRemove: 0 elems in comm_wait_list
GwListRemove: remove elem 90 from comm_wait_list
GwListRemove: 0 elems in comm_wait_list
编辑2:添加代码示例
我创建了一个小代码示例,如果注意到最后一个连接保持打开状态。如果我再次启动该示例,则示例末尾仍保持打开两个连接。 所以出了点问题......
protected void SAP( object sender, EventArgs e ) {
ThreadStart threadStart1 = new ThreadStart( CreateSAP );
Thread thread = new Thread( threadStart1 );
Thread thread2 = new Thread( threadStart1 );
thread.Start();
Thread.Sleep( 2000 );
thread2.Start();
while ( thread2.IsAlive ) {
}
Response.Write( "done" );
}
private void CreateSAP() {
R3Connection conn = ERPConnection.Connection;
conn.MultithreadingEnvironment = true;
for ( int i = 0; i < 12; i++ ) {
try {
conn.Open();
Thread.Sleep( 1000 );
} finally {
conn.Close();
}
}
}
答案 0 :(得分:2)
不幸的是,这是我在我的代码中做的事情,这不是很......很好...... ERPConnect有一个R3Connection对象,我在整个应用程序中使用了一个单例。
如果用户正在使用连接,而另一个用户也请求了连接,则R3Connection对象会创建一个新连接(您可以使用SAP中的SM04事务来验证这一点)。 如果两个用户都关闭了连接,则两个连接中的一个保持打开状态(我认为这是ERPConnect中的一个错误,但我们无法联系他们,因为我们的支持暂时已过期)。 在生产环境中,连接的限制(在SAP中定义)很快就会达到。
我现在为每个用户创建一个R3Connection对象,因此在我的代码中调用Close()方法(或使用语句)时,我确信SAP系统中的连接已正确关闭。
<强>代码强>
public static R3Connection Connection {
get {
ERPConnect.LIC.SetLic( "MyLicenseKey" );
R3Connection connection = new R3Connection(
[set connection settings]
);
connection.MultithreadingEnvironment = true;
return connection;
}
}
答案 1 :(得分:0)
我正在寻找一个完全不同的问题的答案,但看到我偶然发现你的情况,这看起来非常像网络配置 - 确保你的SAP服务器“赫拉克勒斯”有端口3210,3310和3610(我认为那些是正确的 - 可能是3201/3301/3601,很难通过SMGW日志告诉)在Windows防火墙中打开,没有任何其他防火墙阻止相关流量,并且“赫拉克勒斯”上的默认网关是设置为可以成功将流量路由到100.100.103.15的设备(我假设,从您的日志中,这是运行.NET应用程序的服务器的IP地址),以及100.100.103.15上的默认网关设置为可以成功将流量路由到“大力神”的设备。此外,如果您使用完全限定的域名(例如hercules.sap.local)来建立与SAP服务器的连接,那么您将获得更好的运气,因为那时您不再依赖史前的变态,而是找不到SAP服务器。