cancel
Showing results for 
Search instead for 
Did you mean: 

update_client fails on a Solaris 8 system..

Joe_Despres
Level 6
Partner
Having a problem with a Solaris 8 system at 5.1 mp5....

Attempting to push out the upgrade to 6.5.4

         prod5 Solaris/Solaris8 ...
Update of /tmp/bp_inst.14423/java/nbj.conf completed for client prod5
CLIENT_CMD_SOCK from bpcd = 4
CLIENT_STAT_SOCK from bpcd = 7
BPCD connected
Couldn't open /usr/openv/netbackup/version on client prod5
Version files do not match - updating.
bin_net.tar.14423 size = 160256
client_bin.tar.Z size = 59200021
NB-Java.tar.Z size = 96386019
extract_java size = 8374
JRE.tar.Z size = 35521373
prod5.14423 size = 20
move_libs size = 2972
version size = 25
tar size = 500256
.sizes size = 14
client_config size = 52207
x ., 0 bytes, 0 tape blocks
x ./help, 0 bytes, 0 tape blocks
x ./help/bp, 0 bytes, 0 tape blocks
x ./help/bp/Archive, 5649 bytes, 12 tape blocks
x ./help/bp/Backup, 5573 bytes, 11 tape blocks
x ./help/bp/Backup_Btrack, 5143 bytes, 11 tape blocks
x ./help/bp/Backup_Db2, 5085 bytes, 10 tape blocks
x ./help/bp/Backup_Infmx, 5113 bytes, 10 tape blocks
x ./help/bp/Backup_Oracle, 5109 bytes, 10 tape blocks
x ./help/bp/Backup_Sap, 5084 bytes, 10 tape blocks
x ./help/bp/Backup_Sybase, 5100 bytes, 10 tape blocks
x ./help/bp/Help, 1383 bytes, 3 tape blocks
x ./help/bp/Index, 486 bytes, 1 tape blocks
x ./help/bp/MBackup, 1362 bytes, 3 tape blocks
x ./help/bp/MRestore, 3050 bytes, 6 tape blocks
x ./help/bp/Main, 1203 bytes, 3 tape blocks
x ./help/bp/Restore_Btrack, 5196 bytes, 11 tape blocks
x ./help/bp/Restore_Db2, 5111 bytes, 10 tape blocks
x ./help/bp/Restore_Infmx, 5143 bytes, 11 tape blocks
x ./help/bp/Restore_Oracle, 5139 bytes, 11 tape blocks
x ./help/bp/Restore_Sap, 5114 bytes, 10 tape blocks
x ./help/bp/Restore_Sybase, 5139 bytes, 11 tape blocks
x ./help/bp/Restore_arch, 8795 bytes, 18 tape blocks
x ./help/bp/Restore_backup, 8908 bytes, 18 tape blocks
x ./help/bp/Restore_raw, 8500 bytes, 17 tape blocks
x ./help/bp/Restore_tir, 8952 bytes, 18 tape blocks
x ./help/bp/Tutorial, 15844 bytes, 31 tape blocks
x ./bpcd_service, 22 bytes, 1 tape blocks
x ./bpjava_service, 36 bytes, 1 tape blocks
x ./bprd_service, 22 bytes, 1 tape blocks
x ./vnetd_service, 24 bytes, 1 tape blocks
x ./vopied_service, 26 bytes, 1 tape blocks
x ./bpcd_inetd.cnf, 63 bytes, 1 tape blocks
x ./vopied_inetd.cnf, 59 bytes, 1 tape blocks
x ./bpjava_inetd.cnf, 95 bytes, 1 tape blocks
x ./vnetd_inetd.cnf, 56 bytes, 1 tape blocks

INF OTHER - EXIT STATUS = 27
EXIT STATUS = 27

it seems to dump all the files...  but this  "EXIT STATUS" pops up
in the middle of the log file..

Any Idea's?

I need to update this client by tomorrow


Thanks...

Joe Despres

7 REPLIES 7

Will_Restore
Level 6

from the Install Guide:

As a root user on the NetBackup master server, enter the following command to see whether bprd is running: /usr/openv/netbackup/bin/bpps

If bprd is running, terminate it with the following command:
/usr/openv/netbackup/bin/admincmd/bprdreq -terminate

Enter the following command to make sure that backups or restores are not in progress:
/usr/openv/netbackup/bin/admincmd/bpdbjobs

Update UNIX client software by running the update_clients script

After all servers and clients are updated, enter the following command as the root user on the master server to start the bprd daemon:
/usr/openv/netbackup/bin/initbprd
 

Joe_Despres
Level 6
Partner
The client had nothing running.....

Joe Despres

CRZ
Level 6
Employee Accredited Certified
Are we trying to write to an NFS mount which isn't normally read-only, but IS mounted read-only for some reason?

Did /tmp or /usr hit 100% during the initial untarring?

These things are really tricky to diagnose, but if you still have the scripts copied onto the client you can change the first line of the client script we call (name escapes me unfortunately) from "#!/bin/sh" to "#!/bin/sh -x" and usually that'll be enough debug output to clue you in to what's really going on.

Usually incomplete installs leave some cruft lying around that you'll have to delete first - look for "new" directories in /tmp.  (I think it's called "updating_client" or something equally original and obvious :) )

Claudio_Veronez
Level 6
Partner Accredited
Try

/etc/init.d/netbackup stop


CRZ
Level 6
Employee Accredited Certified
"netbackup stop" isn't really used for clients as there should be no daemons running.  Also, it may not even be ON the client.  :)

pylej
Level 4
I had a similar problem and had to clear out about 500MB of free space in the /tmp. Then it installed.

Will_Restore
Level 6
Please check on the Master.