• Please be aware that ADSM.ORG will be going through forum software conversion and migration over the next several weeks. The cut-over date and time will be posted as we near the target date. Please visit the following link for more details ADSM News & Announcements
  • Community Tip: Please Give Thanks to Those Sharing Their Knowledge.

    If you receive helpful answer on this forum, please show thanks to the poster by clicking "LIKE" link for the answer that you found helpful.

  • Community Tip: Forum Rules (PLEASE CLICK HERE TO READ BEFORE POSTING)

    Click the link above to access ADSM.ORG Acceptable Use Policy and forum rules which should be observed when using this website. Violators may be banned from this website. This notice will disappear after you have made at least 3 posts.

Failure Code 42 and 14 - Job Failure

archyis

ADSM.ORG Member
Joined
Jan 3, 2008
Messages
13
Reaction score
0
Points
0
Server is on Linux and the client is on windows. The client can ping the server and communication ports are set correctly (default). The job fails at about 8% everytime.

I enabled more detailed logging and the most notable parts included below...
"dtcp_read: TCP - failure: recv socket (372) (TCP 10058: Can't send after socket shutdown)
10:03:38.816 AM: [1000.1564] <16> dtcp_read: TCP - failure: recv socket (372) (TCP 10058: Can't send after socket shutdown)
10:03:39.831 AM: [1000.1564] <16> dtcp_read: TCP - failure: recv socket (372) (TCP 10058: Can't send after socket shutdown)
10:03:40.847 AM: [1000.1564] <16> dtcp_read: TCP - failure: recv socket (372) (TCP 10058: Can't send after socket shutdown)"

Has anyone ran into this before? I am not able to find anything helpful on the symantec site regarding this.
 

heada

ADSM.ORG Moderator
Joined
Sep 23, 2002
Messages
2,549
Reaction score
168
Points
0
Location
Indiana
Looks like one side has shutdown the TCP connection without notifying the other side.. You could sniff the network and look for the FIN packet to find out which side.

Is this backup taking a long time? If so, is there a timeout value you can increase?

-Aaron
 

archyis

ADSM.ORG Member
Joined
Jan 3, 2008
Messages
13
Reaction score
0
Points
0
This backup fails quite quickly (usually after about 5 minutes) and ends up getting to about 5-8% completed before failing. All other servers are backing up correctly except for this one.

Unfortunatly I work for a very large company and I am involved on the Windows Team, Server belongs to the Unix Team and then the network team would have to be the ones to run the sniffer. :p Talk about red tape!

Anyway, I found this article here which is pretty informative and is the same thing as I am getting - [FONT='Calibri','sans-serif']https://forums.symantec.com/syment/board/message?board.id=21&thread.id=5862 [/FONT]

[FONT='Calibri','sans-serif']So I think what I am going to try is uninstalling the Netbackup agent completely and then reinstalling it and updating it to MP3 to see if that fixes it.[/FONT]
 

sgabriel62

ADSM.ORG Senior Member
Joined
Apr 7, 2005
Messages
1,356
Reaction score
5
Points
0
Location
Michigan
Website
www.sgsolutionsinc.com
Verify your NDD variables at the OS level, If you are failing after 5 minutes, you've run into a contention issue of I would expect validation of files to expire.
Run a check against your Comm and Idle timeout variables, and number of retries as well.
Give me the exact error and I'll go to the support site and dig into the knowledge base.
If the meantime, check your media server, and the media being requested. Perhaps you have run into a bad sector on the tape itself or dust on the tape drive head. Run some queries on the mount requests and see if there is a common fault.
Good luck - let me know
 

archyis

ADSM.ORG Member
Joined
Jan 3, 2008
Messages
13
Reaction score
0
Points
0
This is the error messages -
02/06/2008 15:48:35 - Error bpdm (pid=24384) system call failed - Connection reset by peer (at child.c.1174)
02/06/2008 15:48:35 - Critical bpbrm (pid=24149) from client %servername%: FTL - tar file write error (10054)
02/06/2008 15:48:35 - Error bpdm (pid=24384) unable to perform read from client socket, connection may have been broken
02/06/2008 15:48:36 - Error bpdm (pid=24347) media manager terminated by parent process
02/06/2008 15:48:38 - Error bpbrm (pid=24149) could not send server status message
02/06/2008 15:48:40 - Warning bpbrm (pid=24149) Could not set linger value on socket. Errno = 22: Invalid argument
02/06/2008 15:48:44 - end writing; write time: 0:01:25

Verify your NDD variables at the OS level, If you are failing after 5 minutes, you've run into a contention issue of I would expect validation of files to expire.
Run a check against your Comm and Idle timeout variables, and number of retries as well.
Give me the exact error and I'll go to the support site and dig into the knowledge base.
If the meantime, check your media server, and the media being requested. Perhaps you have run into a bad sector on the tape itself or dust on the tape drive head. Run some queries on the mount requests and see if there is a common fault.
Good luck - let me know
 

archyis

ADSM.ORG Member
Joined
Jan 3, 2008
Messages
13
Reaction score
0
Points
0
Ok, so I ended up getting this backup to work without changing the NIC.

The server in question was a Biscom Faxcom Server...before I had thought it was a regular 2003 Windows Server with a fax software installed on it. My company was alread having problems with our disk staging running out of space with other servers which were reporting a 129 disk full error, while this particular one was reporting the error that this thread is about.

So for kicks, I changed the job to backup straight to tape and ran it manually for the one that always failed and WTF it worked.

So I am concluding that this servers is running into the disk full error, while reporting it different since it is a non-standard machine.
 

noobie2tsm

ADSM.ORG Member
Joined
Sep 30, 2008
Messages
11
Reaction score
0
Points
0
resolution?

Was the resolution of this confirmed to be disk space?
 

The Spectrum Protect TLA (Three-Letter Acronym): ISP or something else?

  • Every product needs a TLA, Let's call it ISP (IBM Spectrum Protect).

    Votes: 22 19.5%
  • Keep using TSM for Spectrum Protect.

    Votes: 67 59.3%
  • Let's be formal and just say Spectrum Protect

    Votes: 15 13.3%
  • Other (please comement)

    Votes: 9 8.0%

Forum statistics

Threads
31,855
Messages
134,948
Members
21,832
Latest member
dataguy81
Top