cancel
Showing results for 
Search instead for 
Did you mean: 

TCP 10058: Can't send after socket shutdown

Anth105
Level 4
Certified
Hi Gurus

Env :  Master NBU 6.54  running on AIX 5.3
          Media NBU 6.54 running on AIX 5.3
          Client  5.1 MP3  running on Win2k sp4

Description
I am getting  a  file read error (13) on a consistent basis regardless of the  backup ( Full or incr).  The servers are  behind a DMZ  but I have set the BPCD connect-back to use the VNETD port and it worked for a long time but now all I receive are 13 errors on one specific server.   Seaching through the  logs files ( bpbkar and bpbrm), I noticed the backups seems to fail after 1 hour when the TCP socket is closed.

Client bpbkar.

17:45:54.076: [4532.2520] <4> dos_backup::OVDeInitSpecialInfo: INF - Special Folder: 3 QP00561,LocalsplOnly QP00561
17:45:54.076: [4532.2520] <4> dos_backup::OVDeInitSpecialInfo: INF - Special Folder: 3 QP00563,LocalsplOnly QP00563
17:45:54.076: [4532.2520] <4> dos_backup::OVDeInitSpecialInfo: INF - Special Folder: 3 QP00564,LocalsplOnly QP00564
17:45:54.076: [4532.2520] <4> dos_backup::OVDeInitSpecialInfo: INF - Special Folder: 3 QP00566,LocalsplOnly QP00566
17:45:54.076: [4532.2520] <4> dos_backup::OVDeInitSpecialInfo: INF - Special Folder: 3 QP00568,LocalsplOnly QP00568
17:45:54.076: [4532.2520] <4> dos_backup::OVDeInitSpecialInfo: INF - Special Folder: 3 U:\NFS NFS
17:45:54.076: [4532.2520] <4> V_LNBackup_End: INF - V_LNBackup_End <Enter> ExitStatus:1 Recycle Logs:0
17:45:54.076: [4532.2520] <2> ov_log::V_GlobalLog: INF - BEDS_Term() Enter InitFlags:0x1
17:45:54.092: [4532.2520] <2> ov_log::V_GlobalLog: INF - FS_RemoveFileSys
17:45:54.123: [4532.2520] <2> ov_log::V_GlobalLog: INF - unloading nbbedsrman.dll
17:45:54.123: [4532.2520] <2> ov_log::V_GlobalLog: INF - unloading nbbedsnt.dll
17:45:54.123: [4532.2520] <2> ov_log::V_GlobalLog: INF - unloading nbbedssql2.dll
17:45:54.123: [4532.2520] <2> ov_log::V_GlobalLog: INF - unloading nbbedsshadow.dll
17:45:54.139: [4532.2520] <16> dtcp_read: TCP - failure: recv socket (496) (TCP 10058: Can't send after socket shutdown)
17:45:54.139: [4532.2520] <16> dtcp_read: TCP - failure: recv socket (700) (TCP 10058: Can't send after socket shutdown)
17:45:54.139: [4532.2520] <4> OVShutdown: INF - Shutdown wait finished
17:45:54.139: [4532.2520] <4> OVShutdown: INF - Closing stderr connection
17:45:54.139: [4532.2520] <2> dtcp_close: TCP - success: close socket (700)
17:45:54.139: [4532.2520] <4> OVShutdown: INF - Closing stdout connection
17:45:54.139: [4532.2520] <2> dtcp_close: TCP - success: close socket (900)
17:45:54.139: [4532.2520] <4> OVShutdown: INF - Closing stdin connection
17:45:54.139: [4532.2520] <2> dtcp_close: TCP - success: close socket (496)
17:45:54.139: [4532.2520] <4> OVShutdown: INF - Finished process

30/07/2009 16:08:00 - estimated 0 kbytes needed
30/07/2009 16:08:02 - started process bpbrm (864288)
30/07/2009 16:08:10 - connecting
30/07/2009 16:08:16 - connected; connect time: 00:00:06
30/07/2009 16:08:23 - begin writing
30/07/2009 17:08:24 - Error bpbrm(pid=864288) socket read failed: errno = 119 - System call timed out.
30/07/2009 17:08:26 - Error bptm(pid=856066) media manager terminated by parent process
30/07/2009 17:08:40 - end writing; write time: 01:00:17
file read failed(13)

Any help would be much appreciated

Anth








1 ACCEPTED SOLUTION

Accepted Solutions

rj_nbu
Level 6
Employee Accredited Certified
Hi

try increasing the client read timeout on the media server. Also run netstat on the client for port 13724 . Large number of connections in TIME_WAIT state indicates a problem.


View solution in original post

2 REPLIES 2

John_Stockard
Level 5
Partner Certified
Is the firewall for the DMZ forcibly shutting down the VNETD connection between this Windows 2000 client and your AIX media server after 1 hour?  The logs in your firewall might indicate whether this is happening.

rj_nbu
Level 6
Employee Accredited Certified
Hi

try increasing the client read timeout on the media server. Also run netstat on the client for port 13724 . Large number of connections in TIME_WAIT state indicates a problem.