I have wandered through both Sun's and Veritas's websites about
socket errors and bpbrm issues and have not found anything to explain
my problems that occured last night. I had bpbrm exit along with a
socket error, which killed off several of my database backups.  I am
wondering if anyone has seen these errors or understand exactly what
happened so I can prevent this from occuring in the future.

The error happened at 7:01am and killed off 6 clients all with the same
errors.

setup:
Solaris 8 280 box connected to 2 LTO standalone drives running
netbackup 4.5 feature pack 3 with appropriate Sun patches.

bpbrm error log output:
07:01:40.235 [12133] <16> bpbrm handle_backup: from client voyager3.gsuprv:
ERR - Cannot write to STDOUT. Errno = 32: Broken pipe
07:01:40.237 [12133] <2> logconnections: BPJOBD CONNECT FROM
192.168.100.106.60781 TO 131.96.3.38.13723
07:01:40.438 [12133] <2> job_monitoring: ACK disconnect
07:01:40.440 [12133] <2> bpbrm handle_backup: client voyager3.gsuprv EXIT
STATUS = 24: socket write failed
07:01:40.440 [12133] <2> inform_client_of_status: INF - Server status = 24
07:01:46.541 [11188] <16> bpbrm handle_backup: from client voyager1.gsuprv:
ERR - Cannot write to STDOUT. Errno = 32: Broken pipe
07:01:46.738 [11188] <2> job_monitoring: ACK disconnect


bpsched error log output:
07:01:47.628 [9826] <2> get_long: (2) premature end of file (byte 1)
07:01:47.628 [9826] <16> readstring: get_string failed:  I/O error (5),
premature end of file encountered (-9)
07:01:47.628 [9826] <8> read_bpbrm_stderr: readstring failed:  socket read
failed (23)
07:01:47.629 [9826] <2> send_runQ_msg: dpid=1 spid=9826 rqtyp=23 status=0
birth=0 stu=Drive0__L9-0 cls=NULL clnt=NULL sched=NULL
07:01:47.629 [9826] <2> clean_up_all: all active backups exiting with 23
07:01:47.629 [9826] <2> send_runQ_msg: dpid=1 spid=9826 rqtyp=14 status=23
birth=0 stu=NULL cls=NULL clnt=NULL sched=NULL
07:01:47.629 [9788] <2> recv_reqQ_msg: msgrcv(delay) interrupted by signal -->
sigcld=0 sigalrm=0 sigusr1=1 sigusr2=0
07:01:47.629 [9788] <2> recv_runQ_msg: msgrcv(nodelay) stat 532  errno 0
(Error 0).  sigcld=0 sigalrm=0 sigusr1=1 sigusr2=0
07:01:47.640 [9788] <16> log_in_errorDB: suspending further backup attempts
for client batman.gsuprv, policy db-unix, schedule full_daily because it has
exceeded the configured number of tries
07:01:47.641 [9788] <2> logconnections: BPJOBD CONNECT FROM
192.168.100.106.60787 TO 131.96.3.38.13723
07:01:47.838 [9788] <2> job_monitoring: ACK disconnect
07:01:47.840 [9788] <2> ?: batman.gsuprv exited with status 23 (socket read
failed)
07:01:47.842 [9788] <2> ?: /usr/openv/netbackup/bin/backup_exit_notify
batman.gsuprv db-unix full_daily FULL 23 1 &
07:01:47.863 [9788] <2> add_to_failure_history: successfully wrote to error
history file--> 09/08/03 07:01:47 batman.gsuprv db-unix full_daily 23 *NULL* 1
1063018907
07:01:47.864 [9788] <16> log_in_errorDB: backup of client batman.gsuprv exited
with status 23 (socket read failed)


Any help or pointers would be appreciated.

Thank you,
Chris
-- 
Christina Whitaker      Georgia State University
                        Systems Administrator
whitaker AT gsu DOT edu        Information Systems & Technology