Hi all,

thank you for your feedback on my initial question.
Now I had the time to look at the Novell documentation.
Here's my personal conclusion to the question about wiki (WD) vs. Novell 
documentation (ND).
All my experiences are related to HAv2:

* ND is well structured. You just can start to read until the end. Even more, 
just the table of contents gives
you a nice overview of relevant issues. WD has no structure for me. 
Navigational elements are
missing (IMHO).
* If you look at the substance of ND, than I have also the feeling that you can 
find much of this stuff
at WD. I emphasis: You CAN find it.
* Worst thing of WD is searching and finding (IMHO). Many times the following 
happens to me:
- I know I read something interesting and valuable at WD.
- I put a search term to the search form to the right
- I get pages of mailing list stuff which hits perfectly the searched term but 
doesn't find the page
in WD.

That happens to me so often in the last weeks, that I just can't tell you how 
annoying this is.

What I'm wondering about is: The underlying MoinMoin does have search 
capabilities.
Why are they not used?

* ND and WS do have much information, BUT someone who is new to all that can't 
set it in a context. (IMHO)
So often I had to ask myself: Mhhh, nice piece of information, but I don't know 
what it means.

* The heartbeat documentation doesn't help you much to get the picture and 
relevant pieces. You
have to collect many many single mosaic tiles very hard to get a rough model of 
HAv2. Everyone
willing to go deeper than the provided GUI is interested in a deeper insight of 
the working of HAv2.

* I started with HAv2 because I had the feeling (only by reading, not 
experience) it is mature enough
to use it in a production environment. And I have to emphasis that someone who 
wants to use
HAv2 wants to enhance availablity of the hosted services. Manageability is a 
key factor
to get a system high available. And for that you have to have documentation 
that gives the user
the feeling always to understand what he's doing and influencing. In opposite 
of that I was
very often surprised about the cluster behaviour when I thought I do understand 
it now.

* Back to documentation though: I'm missing a flow chart or a description of 
actions which
describe the behaviour of all releted components. For example: What happens 
when I start
a cluster node via 'rcheartbeat start', which are the steps taken. Which steps 
are influenced by
which cluster properties set in ha.cf and cib.xml. What happens if I insert a 
primitive resource
to the cluster via cibadmin? What happens when I change a constraint on a 
running cluster.
Reading the logfile with a higher debug level givs you some insights, yes, but 
doesn't explain
anything.
I'm pretty sure that there are many guys (yes, no nice women until now) out 
there who have/had
the same questions. I think it's proved by the kind of questions repeating on 
the mailing list.

* Learning by doing is a good thing, but being forced to make try and error 
while having
a definite target and timeline is not so nice.

Conclusion: I would write down some pieces of information, give some snippets, 
explain
topics further, write "Blabla for dummies" kind of guides, when you out there 
tell me
where to write this down. And of course only if you're interested.

Best regards
Andreas Mock

_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Reply via email to