On Mon, 16 Feb 2015, Florian Kluender wrote:
Yes, I know that this are basics. But I am not too poor to
share my knowledge with other guys. And I hope, that other
specialists thinks like me???
Advance warning: This is going to ramble a bit, and there
are hints of rant to it as well. You have been warned.
The key here is to understand how the entire environment
fits together into a cohesive whole, and that'll take some
time to stitch together in one's mind if he isn't already
familiar with it or familiar with something substantially
similar.
For the database end of things, the schema can be updated
via the network so long as the (SQL) user you're using to do
the job has proper permissions granted by the DBAs. For the
most part, this tends to be an analogue of "GRANT ALL ON
icinga.* TO 'icinga'" or somesuch. The critical one is that
you need the permission to execute an "ALTER TABLE" statement.
One does not have to do it *on* the database system. That's
the simple bit.
Updating the actual software (the programs and executables)
likely means you'll need to have access to the "root" account
unless your system is using RBAC (Role-Based Access Control)
techniques. If packages are in use, the best practise at the
moment seems to be to remove the old one and then install the
new one in its place. While that's in flight, one would
update the database schema if required (always read, and
understand, the release notes), and only once all the moving
parts of the update have slowed to a halt, then start the
suite again and troubleshoot in place if the need arises.
I am very angry about that linux guys are mostly to proud
to do this. In the world of Microsoft you will never find
such admins or developers???
I have seen plenty in the Microsoft world, trust me. From
my own standpoint, I try not to be that way, but I understand
the genesis of the behaviour. You're dealing with people who,
in many cases, have spent a goodly chunk of their lives getting
to the point where they are -- and are frequently misunderstood.
I've been in the world of computing for more than a third of a
century, and have seen a lot of what goes on and, yes, some of
it is not pretty. Neither, frequently, is the technology. Put
politely, for all the "advancement" we've seen in the past
decades has been impressive, but the technology is still messy;
it has a lot of moving parts; many of those parts are brittle;
and computers have a habit of doing what you tell them, not
what you want them to do. Getting to the point of being a good
systems administrator likely takes the equivalent of a Master's
Degree because of the body of knowledge that we have to internalise.
This is not "certifications" and whatnot; this is knowing *why*
the dratted things work at all, and knowing how to leverage that
in the real world. And we still get looked down upon and slagged
off on when we get short with folks who don't seem to have read
the documentation that's provided (I am guilty of that myself;
sometimes I don't RTFM, and it's bitten me on the arse a few
times).
Did you ever get the ido2db process to stay running and actually
connect to -- and stay connected to -- the database?
Cheers!
+------------------------------------------------+---------------------+
| Carl Richard Friend (UNIX Sysadmin) | West Boylston |
| Minicomputer Collector / Enthusiast | Massachusetts, USA |
| mailto:[email protected] +---------------------+
| http://users.rcn.com/crfriend/museum | ICBM: 42:22N 71:47W |
+------------------------------------------------+---------------------+
_______________________________________________
icinga-users mailing list
[email protected]
https://lists.icinga.org/mailman/listinfo/icinga-users
