I have a recollection this was covered earlier, but some searching of
archived ADSM-L postings and some search of IBMLINK has not been
fruitful, so ...

I have an OS/2 client (I don't know it's level ... it won't tell me
(I don't have direct access to the machine, but do have the schedule
and error logs and the server log)) whose scheduled backups to a 0.8/1.8
ADSM/VM server are failing.  The schedule log says, in part...

>Incremental backup of drive 'D:' with label 'OS2-D'
>Normal File-->           11,469 D:\ADSM\DSMERROR.LOG   Sent
>Normal File-->          498,521 D:\ADSM\DSMSCHED.LOG   Sent
>Normal File-->        1,687,040 D:\IVR\VOICE\FNAID002.VLB   Compressed Data 
>Grew
>Retry # 1   Normal File-->        1,687,040 D:\IVR\VOICE\FNAID002.VLB   ** 
>Unsuc
>ANS4017E Session rejected: TCP/IP connection failure
>Command will be retried  in 20 minutes.

The error log says, in part...

>commtcp.(1290): TcpFlush: Error -1 sending data on Tcp/Ip socket 47.
>session.(1425): sessSendVerb: Error sending Verb, rc=-50
>commtcp.(1290): TcpFlush: Error -1 sending data on Tcp/Ip socket 47.
>commtcp.(1290): TcpFlush: Error -1 sending data on Tcp/Ip socket 47.
>session.(1425): sessSendVerb: Error sending Verb, rc=-50
>ANS4017E Session rejected: TCP/IP connection failure
>ANS4847E Scheduled event 'NOON' failed.  Return code = 1.

The dsm.opt file is set for the maximum tcpbuffsize and
tcpwindowsize.  Is this a timeout problem?  Is there a bypass?

Thanks in advance,

Wayne T. Smith
Systems Group -- CAPS        internet: wts AT maine.maine DOT edu
University of Maine System   BITNET/CREN: WTS@MAINE