Unsolved

This post is more than 5 years old

2 Intern

 • 

724 Posts

1406

July 24th, 2007 14:00

Service won't start on Solaris client

Hi,

I have one Solaris 9 (64bits) client here where I cannot start Nw software. It was working with 7.3.1, but after a reboot the service did not start. I tried the latest version in Powerlink (7.3.2 Update 1), but got the same results.

When trying to start the service using "/sbin/init.d/networker start", it waits for hours and does not finish. I can see that nsrexecd is running, but I cannot see the Nw port open using rpcinfo, and backup fails also.

On syslog and messages files I did not find any error messages. I guess Nw is trying to do something to open the port for communication, but I don't know where exactly I should look into. My Unix guys told me that there's no restriction on the client regarding ports.

Any ideas?
Thanks!

2 Intern

 • 

724 Posts

July 24th, 2007 14:00

root@snesun307 # /usr/sbin/nsrexecd -D9
nsrexecd: clu_is_cluster_mode:ENTRY
nsrexecd: app_name=NetWorker, app_dir=/usr/sbin/
nsrexecd: b_clusmode=0
nsrexecd: clu_is_cluster_mode:EXIT
07/24/07 18:39:45 nsrexecd: Origin is /usr/sbin/
07/24/07 18:39:45 nsrexecd: clu_is_cluster_host:ENTRY
07/24/07 18:39:45 nsrexecd: clu_is_cluster_host:EXIT:returning FALSE
07/24/07 18:39:45 nsrexecd: lg_statvfs(): Calling statvfs64().
07/24/07 18:39:45 nsrexecd: lg_statvfs(): Calling statvfs64().
07/24/07 18:39:45 nsrexecd: lg_statvfs(): Calling statvfs64().
07/24/07 18:39:45 nsrexecd: lg_statvfs(): Calling statvfs64().
07/24/07 18:39:45 nsrexecd: lg_statvfs(): Calling statvfs64().
07/24/07 18:39:45 nsrexecd: lg_statvfs(): Calling statvfs64().
07/24/07 18:39:45 nsrexecd: lg_statvfs(): Calling statvfs64().
07/24/07 18:39:45 nsrexecd: lg_statvfs(): Calling statvfs64().
07/24/07 18:39:45 nsrexecd: lg_statvfs(): Calling statvfs64().
07/24/07 18:39:45 nsrexecd: lg_statvfs(): Calling statvfs64().
07/24/07 18:39:45 nsrexecd: clu_is_cluster_host:ENTRY
07/24/07 18:39:45 nsrexecd: clu_is_cluster_host:EXIT:returning FALSE
07/24/07 18:39:45 nsrexecd: interface addr = 0.0.0.4
07/24/07 18:39:45 nsrexecd: interface addr = 10.170.50.21
07/24/07 18:39:45 nsrexecd: interface addr = 10.170.50.99
07/24/07 18:39:45 nsrexecd: interface addr = 10.170.50.25
07/24/07 18:39:45 nsrexecd: interface addr = 10.170.50.177
07/24/07 18:39:45 nsrexecd: interface addr = 172.16.1.1
07/24/07 18:39:45 nsrexecd: interface addr = 172.16.0.129
07/24/07 18:39:45 nsrexecd: interface addr = 172.16.193.1
07/24/07 18:39:45 nsrexecd: clus_get_hostname:ENTRY:
07/24/07 18:39:45 nsrexecd: hostname=null, path=/nsr/res
07/24/07 18:39:45 nsrexecd: clu_pathl_gethost:ENTRY
07/24/07 18:39:45 nsrexecd: input path=/nsr/res
07/24/07 18:39:45 nsrexecd: clu_pathl_gethost:EXIT:Clu_cluster_type is undefined
07/24/07 18:39:45 nsrexecd: Server_name=snesun307
07/24/07 18:39:45 nsrexecd: clus_get_hostname:EXIT:
07/24/07 18:39:45 nsrexecd: clus_get_hostname:ENTRY:
07/24/07 18:39:45 nsrexecd: hostname=null, path=/nsr/res
07/24/07 18:39:45 nsrexecd: Server_name=snesun307
07/24/07 18:39:45 nsrexecd: clus_get_hostname:EXIT:
07/24/07 18:39:45 nsrexecd: clus_get_hostname:ENTRY:
07/24/07 18:39:45 nsrexecd: hostname=null, path=/nsr/res
07/24/07 18:39:45 nsrexecd: Server_name=snesun307
07/24/07 18:39:45 nsrexecd: clus_get_hostname:EXIT:
07/24/07 18:39:45 nsrexecd: could not find GUI metadata directory
07/24/07 18:39:45 nsrexecd: Adding range 7937-9936 to "service" port type.
07/24/07 18:39:45 nsrexecd: socket bound to port 9933
07/24/07 18:39:45 nsrexecd: Calling clnttcp_create function
07/24/07 18:39:45 nsrexecd: socket bound to arbitrary port

The command stops responding after this last line. This server is in a Sun Cluster 3.1.

6 Operator

 • 

14.4K Posts

 • 

56.2K Points

July 24th, 2007 14:00

What do you see when doing /usr/sbin/nsrexecd -D9 ?

6 Operator

 • 

14.4K Posts

 • 

56.2K Points

July 24th, 2007 14:00

Not sure, but could be something has changed since last reboot. Try to reconfigure whole thing - I have a feeling this is quickest way towards having that client up and running.

2 Intern

 • 

724 Posts

July 31st, 2007 15:00

Well, finally we found the problem. Some wise guy changed the loopback address to another value other than 127.0.0.1, and did not update the /etc/hosts file. As far as I could verify, Nw uses this to when opening it's ports, and it was failling due to name resolution.

0 events found

No Events found!

Top