Unsolved
This post is more than 5 years old
24 Posts
0
11368
How to start Networker server?
I installed Networker 8.2.0.1 (lgtoclnt, lgtonode, lgtoserv and lgtonmc) for Centos 7.2.1511 64bit and I don't know how to start Networker server.
I run '/etc/init.d/networker start'.
When I tried to connec to to server localhost from Management Console I got an error 'Unable to connect to server'.
When I run 'nsradmin' I got 'There does not appear to be a NetWorker server running on localhost'.
When I run 'ps -ef | grep nsr' I saw that nsrexecd was running, but nsrd was not.
When I tried to run '/usr/sbin/nsrd I got:
0 1455629804 1 5 0 3935561536 22284 0 localhost nsrd NSR notice 5 %s %s 2 0 24 02/16/16 08:36:44.881414 0 27 Shutdown_delay_interval: 10
83271 1455629804 1 5 0 3935561536 22284 0 localhost nsrd NSR notice 76 Startup in process (step %d of %d); opening NetWorker server RAP database... 2 1 1 1 1 1 5
83272 1455629804 1 5 0 3935561536 22284 0 localhost nsrd NSR notice 64 Startup in process (step %d of %d); initializing RAP handlers... 2 1 1 2 1 1 5
71193 1455629804 0 0 0 3935561536 22284 0 localhost nsrd NSR info 9 %s %s: %s 3 11 11 2646:Server 11 12 19362:Notice 49 14 7324 7 started
88940 1455629804 5 5 0 3935561536 22284 0 localhost nsrd NSR critical 61 %s: Cannot get server credentials from authentication daemon. 1 0 10 GSS Legato
12546 1455629804 1 5 0 3935561536 22284 0 localhost nsrd NSR notice 13 shutting down 0
0 1455629804 1 5 0 3935561536 22284 0 localhost nsrd NSR notice 5 %s %s 2 0 24 02/16/16 08:36:44.911391 0 225 An unexpected error occured in file: rpc/lib/gssglue.c, line: 159, message = "Daemon credential was not properly acquired at startup. gsslgtov1 authentication is disabled. Please call get_gss_server_cred at daemon startup."
Are there any post-installation steps that should be done?
bingo.1
2.4K Posts
0
February 16th, 2016 10:00
What you did so far is correct. It seems as if there is a problem with the authentication db (/nsr/res/nsrladb).
- Stop the daemons (/etc/init.d/networker stop or nsr_shutdown)
- Rename the directory
- Restart the daemons
tolejniczak
24 Posts
0
February 17th, 2016 02:00
Unfortunately it didn't help. Nsrd still won't start.
bingo.1
2.4K Posts
0
February 17th, 2016 03:00
Hmmm ... look which daemons have really been started. (ps -ef | grep nsr)
Your first message made me think that there is an issue with the client.
However, if there is an nsrexecd process then the server is failing to start (nsrd). This is due to either a corrupt res file or/and the media index.
As it is a new install ...
- Make sure all nsr daemons have been stopped. If necessary, kill them
- Delete
/nsr/res/
/nsr/mm/
/nsr/index/
/nsr/logs/
/nsr/tmp/
- Now start NW again.
Verify which NW daemons are running.
tolejniczak
24 Posts
0
February 17th, 2016 03:00
Only nsrexecd was started.
I removed the directories you listed from /nsr and started Networker again, however only nsrexecd is running. When I then tried to start nsrd manually, the same error as above (daemon credentials) happened.
tolejniczak
24 Posts
0
February 17th, 2016 03:00
The same result:
[root@localhost to]# /usr/sbin/nsrexecd
[root@localhost to]# ps -ef | grep nsrexecd
root 25361 1 1 09:26 ? 00:00:00 /usr/sbin/nsrexecd
root 25371 24550 0 09:27 pts/1 00:00:00 grep --color=auto nsrexecd
[root@localhost linux_x86_64]# /usr/sbin/nsrd -D1
0 1455632858 1 5 0 792016704 25426 0 localhost nsrd NSR notice 5 %s %s 2 0 24 02/16/16 09:27:38.191963 0 27 Shutdown_delay_interval: 10
83271 1455632858 1 5 0 792016704 25426 0 localhost nsrd NSR notice 76 Startup in process (step %d of %d); opening NetWorker server RAP database... 2 1 1 1 1 1 5
83272 1455632858 1 5 0 792016704 25426 0 localhost nsrd NSR notice 64 Startup in process (step %d of %d); initializing RAP handlers... 2 1 1 2 1 1 5
71193 1455632858 0 0 0 792016704 25426 0 localhost nsrd NSR info 9 %s %s: %s 3 11 11 2646:Server 11 12 19362:Notice 49 14 7324 7 started
88940 1455632858 5 5 0 792016704 25426 0 localhost nsrd NSR critical 61 %s: Cannot get server credentials from authentication daemon. 1 0 10 GSS Legato
12546 1455632858 1 5 0 792016704 25426 0 localhost nsrd NSR notice 13 shutting down 0
0 1455632858 1 5 0 792016704 25426 0 localhost nsrd NSR notice 5 %s %s 2 0 24 02/16/16 09:27:38.388915 0 225 An unexpected error occured in file: rpc/lib/gssglue.c, line: 159, message = "Daemon credential was not properly acquired at startup. gsslgtov1 authentication is disabled. Please call get_gss_server_cred at daemon startup."
crazyrov
4 Operator
4 Operator
•
1.3K Posts
0
February 17th, 2016 03:00
Can you run nsrexecd in one ssh session. once that is running open another ssh session and run nsrd manually? Share the output here. I know you have done this earlier but i assume at that moment the nsrd was already trying to come up but failed.
bingo.1
2.4K Posts
0
February 17th, 2016 09:00
At this point I do not know anything else but to
- Deinstall NW
- Remove /nsr completely
- Reinstall NW
tolejniczak
24 Posts
0
February 17th, 2016 13:00
The same behavior as before.
Maybe I should try on Centos 6?
bingo.1
2.4K Posts
0
February 17th, 2016 15:00
I don't think so. I have successfully installed NW 8.2.2 on Centos 7.
Please see the Compatibility Guide to verify your configuration.
Do not forget
- beware of dependencies - let Centos install missing packages
- Disable the firewall (for test purposes only)
tolejniczak
24 Posts
1
February 17th, 2016 23:00
I'm using Virtualbox - maybe that's the problem?
Compatibility Guide requires kernel 2.6 - I assume that means also all higher versions (for example 3.10)?
I installed ksh, glibc.i686, nss-softokn-freebl.i686 and compat-libstdc++-33 manually. I assumed yum would install any missing dependencies?
I stopped firewalld, but it didn't help.
I noticed that I skipped SELinux configuration step because I assumed it won't be enabled if I don't do it manually. However it seems CentOS enables it by default. I executed commands from installation guide ('semanage fcontext -a -t textrel_shlib_t "/usr/lib/nsr/lib.*\.so' and 'restorecon -R /usr/lib/nsr', but only after installing Networker).
bingo.1
2.4K Posts
0
February 18th, 2016 06:00
I just repeated my installation and provide you with the script of my steps.
Maybe this can help.
1 Attachment
networker_8222_with_centos_7.txt
wlee
263 Posts
0
May 15th, 2018 09:00
The best way to install software is to use the yum command.
Before using yum, you need to:
- If you do not have one, create a Redhat account
https://www.redhat.com/wapps/ugc/register.html;jsessionid=txwh3urniE-EhmxOKnN7OatH.1fa0d0df?_flowId=register-flow&_flowExecutionKey=e1s1
- On your RedHat host,. register your redhat system using: # subscription-manager register. It will ask you for your RedHat account and password
- next use: # subscription-manager attach
- download , unzip and untar the NetWorker package
- use yum to install, eg... "# yum install lgtoclnt-8.2.4.11-1.x86_64.rpm". This example will install the client, and it will try to install any dependent packages it needs.
In my first attempt yum still fails with :
Error: Package: lgtoclnt-8.2.4.11-1.x86_64 (/lgtoclnt-8.2.4.11-1.x86_64)
Requires: libstdc++.so.5()(64bit)
...
Packages skipped because of dependency problems:
lgtoclnt-8.2.4.11-1.x86_64 from /lgtoclnt-8.2.4.11-1.x86_64
...
warning: /var/cache/yum/x86_64/7Server/rhel-7-server-rpms/packages/glibc-2.17-222.el7.x86_64.rpm: Header V3 RSA/SHA256 Signature, key ID fd431d51: NOKEY:-- ETA
Public key for glibc-2.17-222.el7.x86_64.rpm is not installed
To resolve this, I use this:
# yum-config-manager --enable rhel-7-server-optional-rpms
# yum install compat-libstdc++-33
# yum install lgtoclnt-8.2.4.11-1.x86_64.rpm
wlee
263 Posts
0
May 15th, 2018 12:00
I agree if the server was running CentOS.
But @rajeev_kishore said his server was RedHat O/S
Kleinenbroich
86 Posts
0
May 15th, 2018 12:00
I don't see a need for an RedHat account, Centos is an Open Source Linux build out of the available sources from the RedHat project and it will not resolve and download dependencies from the RHEL repository.
I excpect, that all packages should be available on the Centos DVD image and should be resolved automaticly. If not, yum will raise a related alert, in that case the dependencies need to be resolved and not to be ignored.
Regards
Michael