We use an open source graphing tool over the RRD's collected by Zenoss. Works great. If you would like details, email me.

- Laird Popkin, CTO, Pando Networks

Sent from my iPhone. Apologies in advance for typo's.

On Aug 2, 2007, at 7:27 PM, Eli Stair <[EMAIL PROTECTED]> wrote:


Ah, pity to reply to my own response like this but I thought of one more uber-useful feature that needs to be added, whether it be by me, the community,
or the ZenOSS organization:

8) Custom graphs/dashboards for aggregating+viewing RRD data.

Among the features that makes Cacti the best "monitoring" tool for viewing trending/graphed data (in spite of its horrible interface for creating custom templates/graphs ;) is the ability to create custom graphics pulling data from all available RRD's in the system. EXTREMELY useful for creating aggregate graphs for ingress/egress bandwidth via WAN/internet links, total filer ops for
a Netapp cluster, jobs in queue for a cluster, etc.

It's currently great that so MUCH info can be gathered with zenoss, one way or another, but the ability to do something useful with that is still lacking, IMO. Even alerting is fairly undeveloped, but in time surely will evolve as well.

Again, thanks for asking for input.

Cheers,

/eli



Eli Stair wrote:

Definitely. Sorry for the long email, I assume you're interested since you asked, if not skip to the bottom for my top-7 list in quick summary ;)

I'm currently migrating (and significantly expanding upon) my current very customized Nagios deployment. I've got >5,000 hosts, >20,000 frequent
service
checks across my "enterprise" composed of several international
locations and
WAN links, making extremely extensive use of service dependencies and event handlers, as well as being very dependent on the parent directive (among
others
obviously ;).  Nagios has some extreme limitations, but works very
dependably
as an alerting mechanism. The addition of trending/graphing, combined
with the
core infrastructure is one of the few ways to improve on it, IMO. Or to
make a
superior package.

In the process of testing and rolling out zenoss, I'm quite impressed by the number of things that have been included, but my biggest concern is that
there
are a large number of features that are not documented well (if at all),
or are
"included" but non-functional.  Specifically, WMI access,
event-handler/response script usage, the RPC/XML API, and a number of other items I've attempted to use are either downright broken or there is no
way to
access without doing serious digging through the underlying code to
glean some
knowledge... that's not a very efficient way of trying to implement
something.

That being said, I'm not trying to be negative, but give you some creative
criticism.  I'm making a serious effort to utilize what does work,
figure out
what doesn't and document it, etc. I've already written some scripts to
handle
integrating Zenoss with other tools using what IS known about the API and system to interact with it. I'm concurrently evaluating Zabbix and Hyperic, both of which are missing some core functionality that you have already (but
sometimes doesn't work), and flaws of their own to be sure.

What I'd like to see most out of the next few releases of zenoss:

1) Stability of the core daemons, and better (useful) alerting on their
failures.
2) Functional WMI system with documentation on use.
3) Xenpack documentation, and a system that works.
4) API documentation, abilities, examples, how to replicate GUI actions,
etc.
5) Better documentation of "event handler" equivalents.
6) Network parent device capabilities, needs to support >1 device
7) Service dependency support, integrated with event handlers.

So most of the groundwork for that appears to already be laid very
firmly.  I
know from the list that problems I've seen are seen by others, and are
likely
being worked on already.  The topic of documentation is a sticky one,
everyone
(including me) would rather be "doing" rather than writing for
posterity, but
it's hard for new adopters to get much use out of the product (and stick
with
it), and eventually buy something if it's TOO difficult ;)


Thanks for creating such an exemplary piece of code so far. Lots of people have "tried" to do better than Nagios, without much success or even hope... without contributing back anything of worth to the community. While you
guys
have gone on a tangent, it's obviously with skill and an actual goal.
Doing well.


Cheers,


/eli



mrhinkle wrote:
We have noticed that many Zenoss users are former Nagios users. To help
understand what features you liked in Nagios and replicate those
features in Zenoss we need your help. If you have thoughts please
respond to the poll and offer any other insights you might have to make your migration from Nagios to Zenoss more rewarding. If you have other thoughts not addressed in the questions on the poll please expand upon
them in the comments. Thank you for your participation.

Mark
The Zenoss Community Dude




-------------------- m2f --------------------

Read this topic online here:
http://community.zenoss.com/forums/viewtopic.php?p=9446#9446

-------------------- m2f --------------------



_______________________________________________
zenoss-users mailing list
[email protected]
http://lists.zenoss.org/mailman/listinfo/zenoss-users

_______________________________________________
zenoss-users mailing list
[email protected]
http://lists.zenoss.org/mailman/listinfo/zenoss-users

_______________________________________________
zenoss-users mailing list
[email protected]
http://lists.zenoss.org/mailman/listinfo/zenoss-users
_______________________________________________
zenoss-users mailing list
[email protected]
http://lists.zenoss.org/mailman/listinfo/zenoss-users

Reply via email to