AIX

Hi

I am currently testing SCOM2012 on my AIX systems for monitoring. I tested it on 3 systems and immediately i got the following errors:

Code:

Alert: Network Status Monitor daemon (rpcstat) is not running
Source: AIX 7.1
Path: (left blank)
Last modified by: (left blank)
Last modified time: (left blank)
description: The Network Status Monitor daemon on server (removed system name) is not running.

Is this related to rpc.ststd?
Isn't rpc.statd and lockd related to NFS?
Do these deamons run when NFS is invoked?
Is it neccessary that this daemon is running at all times?

Thanks

When the error message of the application says so, who am i to disagree?

Yes. "RPC" is short for "remote procedure call" and NFS builds upon this framework. For NFS to work "rpc.statd" (and a few other daemons as well) has to run.

Generally, daemons run when they are started. If NFS is to work: yes, they should be started. If not, then not necessarily. ;-)

No, only when it is needed: when NFS runs, it is needed indeed.

Counterquestion: what has NFS to do with "SCOM2012"? I don't know this software, but it is definitely not part of the NFS suite.

I hope this helps.

bakunin

@bakunin
Thank you for the reply bakunin.
I did not explain myself properly, appologies..... I installed SCOM2012, a microsoft product, to use as a monitoring tool for the AIX servers i have. In testing the installation, on the SCOM console it displays the following error.....

So to answer your question, SCOM has nothing to do with NFS in the fact that it is reporting an error with the rpc_statd daemon that is associated to teh NFS service. (whew!).

So my question is....
1 - Do i need to manually start this daemon or does it start automatically when NFS is invoked?
2 - If the answer to question 1 is 'No, i do not need to start the daemon manually' then is it an error I can ignore?

Thanks again

RPC (remote procedure call) is a framework of functions. It is used for all sorts of things, one (in fact the main) being NFS. "rpc.statd" implements part if this framework, "lockd" implements another.

Both these daemons are started and stopped va the SRC (system resource controller) and will have to be started manually if you invoke "nfsd" manually. As the SRC knows groups which cann be managed together and "nfsd", "rpc.statd" and "lockd" are all members of the group "nfs" you can start/stop/restart the complete NFS service with:

Code:

startsrc -g nfs
stopsrc -g nfs
refresh -g nfs

and don't have to start these daemons one after the other.

Again: i do not know SCOM2012, i have no idea for what it might need "rpc.statd" (or NFS) and hence i can't say if you can ignore the message or not. Maybe you can, maybe not.

IMHO - but that is just my personal opinion - a monitoring product which relies on NFS and/or RPC is not worth it, as it poses an inherent security threat. I know that RPCing back and forth is common with M$$, but their products aren't exactly the epitome of security either. I'd look for a tool which relies on SNMP, which is a protocol designed for exactly this purpose. There is an awful lot of them, from very big (HP-OpenView, BMC Patrol) to small, lean and freeware.

I hope this helps.

bakunin