Client connection refused (status 57)

[craig_chang]

I recently upgraded from Solaris 8 to solaris 10. I no longer can connect to the client. This is a new solaris 10 install and I do not have any clue what to do. I am currently running netbackup 5.1. Any help would be very much appreciated.

[Stumpr2]

Was there a hostname change? IP change? Have you troubleshot with the bpclntcmd?

[Nathan_Kippen]

The following technote might be of some use:

http://seer.entsupport.symantec.com/docs/315561.htm

[Karthikeyan_Sun]

status 57: client connection refused 

Details:
On each NetBackup client, the /etc/inetd.conf file contains the list of servers that inetd(1M) invokes when it receives an Internet request over a socket.  Each server entry is composed of a single line of the form:

    service-name endpoint-type protocol wait-status uid server-program server-arguments
    
For a NetBackup client to be able to transmit data via the bpcd daemon to a media server, when it receives a backup request over a socket, the following line must be in the /etc/inetd.conf:

bpcd    stream  tcp     nowait  root    /usr/openv/netbackup/bin/bpcd bpcd

Edit this line into the /etc/inetd.conf file on the client, or uncomment the line (i.e remove the # at the beginning of the line) in the file if it has been commented out.

Finally, stop and restart the "inetd" process to have the changes made to inetd.conf take effect:

1) obtain the process id of the "inetd" process:

     ps -ef | grep inetd

2) Use the process ID obtained to stop and restart "inetd":

   kill -HUP  "inetd process id obtained from ps -ef"

[craig_chang]

There was no hostname or ip address change. How do I use the bpclntcmd?

[craig_chang]

I used the install client feature from the admin console. Everything looks just fine on the client side.

[Stumpr2]

netstat -a | grep bpcd

are you listening?

[craig_chang]

It looks like the inetd.conf is setup correctly and the services has the right ports to listen to, but netstat doesn't show the bpcp running.

[Stumpr2]

correction. the bpcd should be "listening" on the client

[Ashz_Tomz]

check if the lookups are working from both ends..

u can pipe out the port status from bothe machines and see if it takes in,

enable robust logging and paste the status here...

[craig_chang]

This is from the server:

#  netstat -a | grep bpcd
      *.bpcd               *.*                0      0 24576      0 LISTEN
midsun1.bpcd         midsun1.827          32768      0 32768      0 TIME_WAIT
midsun1.bpcd         midsun1.738          32768      0 32768      0 TIME_WAIT
midsun1.bpcd         midsun1.931          32768      0 32768      0 TIME_WAIT
midsun1.bpcd         midsun1.921          32768      0 32768      0 TIME_WAIT
midsun1.bpcd         midsun1.968          32768      0 32768      0 TIME_WAIT
midsun1.bpcd         midsun1.639          32768      0 32768      0 TIME_WAIT
midsun1.767          midsun1.bpcd         32768      0 32768      0 FIN_WAIT_2
midsun1.bpcd         midsun1.767          32768      0 32768      0 CLOSE_WAIT
midsun1.bpcd         midsun1.994          32768      0 32768      0 TIME_WAIT
midsun1.bpcd         midsun1.660          32768      0 32768      0 TIME_WAIT
#

This is from the client:

# netstat -a | grep bpcd
#

[Karthikeyan_Sun]

http://seer.entsupport.symantec.com/docs/256685.htm

may be helpful ! 

[Stumpr2]

On the client:

the inetd.conf should contain

bpcd    stream  tcp     nowait  root    /usr/openv/netbackup/bin/bpcd bpcd
vnetd   stream  tcp     nowait  root    /usr/openv/bin/vnetd vnetd
vopied  stream  tcp     nowait  root    /usr/openv/bin/vopied vopied
bpjava-msvc     stream  tcp     nowait  root    /usr/openv/netbackup/bin/bpjava-msvc bpjava-msvc -transient

the /etc/services should contain

# NetBackup services #
bprd    13720/tcp       bprd
bpcd    13782/tcp       bpcd
vnetd   13724/tcp       vnetd
vopied  13783/tcp       vopied
bpjava-msvc     13722/tcp       bpjava-msvc

You will need to restart the inetd daemon

kill -HUP  `ps -ef | grep inetd`

[Stumpr2 · Accepted solution]

I found this on google

Netbackup 5.1 client on Solaris 10

If you push the netbackup client from the master server to a solaris 10 box it thinks it is solaris 8, so it writes entries to /etc/inetd.conf

Solaris 10 does not have inetd but SMF, so the bpcd port is not activated, you can test by doing a telnet localhost bpcd... it does not answer.

So you have to convert your /etc/inetd.conf file to SMF descriptors, you do so by running

# inetconv

You make sure it is imported by doing

# inetadm | grep -i bpcd
# telnet localhost bpcd

if you get a connection to the bpcd port you're all set and you're ready for backup/restore

[rishi_rich]

Hi Stumpr,

Thank you vrey much,you rocks