NetBackup media eject timeout has changed

Problem

As of NetBackup version 7.x, there is now a timeout of 300 seconds on an eject which gives 5 minutes to remove tapes from the Media Access Port, otherwise a timeout error is generated.

Error Message

from robots log:

09:44:05.252 [636] <6> tldcd:SendEjectNotice: Using map_continue_timeout value of 300 seconds<snip!>09:49:05.182 [636] <2> get_long: (1) cannot read (byte 1) from network: Interrupted system call (4)09:49:05.182 [636] <3> tldcd:SendEjectNotice: TLD(0) timed out awaiting CONTINUE for EJECT, retracting media access port09:49:05.182 [636] <5> tldcd:MultipleEjectLoop: scsi_command_error09:49:05.253 [636] <6> tldcd:MultipleEjectLoop: scsi_command_error, errno = 409:49:05.254 [636] <3> tldcd:SendMailslotStatusStrings: delete = 1, finished = 1... deleting file09:49:06.287 [636] <5> tldcd:tld_main: TLD(0) closing/unlocking robotic path09:49:06.316 [636] <2> TAO: ACE_Select_Reactor_Notify::notify [handle=259]: write to notification pipe handle failed: Timer expired (62)

Cause

Tapes were not removed from MAP  in less than 5 minutes.

Solution

Add a directive to the /usr/openv/volmgr/vm.conf file on the robot control host, and restart MM  daemons for the change to take effect. The directive is

MAP_CONTINUE_TIMEOUT = <timeout in seconds>

For example, to set the amount of time to remove tapes from the MAP  to 15 minutes, add the directive

MAP_CONTINUE_TIMEOUT = 900

and restart MM  processes on the robot control host.