IT31889: IF REPLICATION IS SLOW OR TAKES TOO LONG THEN THE OPERATION FAILS WITH ERROR MESSAGES.

APAR status

• Closed as program error.

Error description

• When a replication with IBM Spectrum Plus is running between two
  remote sites, if it takes longer than 10 days or runs over an e
  xtremely slow network, it can fail with one of the following err
  ors that can be seen in the job log and/or repl.log on the prima
  ry vSnap.
  No such file or directory:'/tmp/repld_send_<id>_error'
  Receive operation failed: packet_write_wait: Connection to port2
  2:Broken pipe.
  

Local fix

• If the replication runs for too long, some important files in /t
  mp get cleaned up. To avoid this, run the following command on t
  he vSnap:
  $ echo 'x /tmp/repld_*' | sudo tee -a /usr/lib/tmpfiles.d/tmp.co
  nf
  

Problem summary

• ****************************************************************
  * USERS AFFECTED:                                              *
  * IBM Spectrum Protect Plus level 10.1.5.                      *
  ****************************************************************
  * PROBLEM DESCRIPTION:                                         *
  * See Error Description                                        *
  ****************************************************************
  * RECOMMENDATION:                                              *
  * Apply fixing level when available. This problem is currently *
  * projected to be fixed in IBM Spectrum Protect Plus levels    *
  * 10.1.5 patch1 and 10.1.6. Note that this is subject to       *
  * change at the discretion of IBM.                             *
  ****************************************************************
  

Problem conclusion

• The replication process on vSnap makes use of some temporary
  files under /tmp. If a replication session runs very slowly and
  lasts for several days, the temporary files may remain
  unmodified for several days. This can cause the operating system
  to remove the temporary files which eventually results in a
  failure of the vSnap replication session. This problem has been
  resolved by modifying the system configuration to exclude the
  replication-specific temporary files from being cleaned up
  automatically.
  
  Additionally, when the replication sessions runs very slowly,
  the SSH connection between the vSnap servers may experience
  timeouts and disconnections if no data is transferred for a few
  minutes. This also results in a failure of the replication
  session. This problem has been resolved by ensuring that when
  vSnap creates an SSH connection during replication, it uses
  larger default values for the parameters 'ServerAliveInterval'
  and 'ServerAliveCountMax'.
  

Temporary fix

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

• APAR is sysrouted TO one or more of the following:

Fix information

• Fixed component name

  SP PLUS

• Fixed component ID

  5737SPLUS

Applicable component levels