Call for discussion on ZABBIX 1.8

[Koichi Morigami]

well, i'd love to see most of the features listed http://www.zabicom.com/english/index-01.html#pro0503 to propogate back upstream

(though i'm unsure what "NTT ComTechnology License" at some items implies)

As Alexei answered to you on another thread
(http://www.zabbix.com/forum/showthread.php?t=10673),
some of the features may be integrated into base product at some point.

Acknowledge function for events has already been integrated into base
product.

Please note that following three functions ware developed by
NTT ComTechnology and released under NTT ComTechnology's License.
These functions are not ZABBIX product.

1. Sync function of the configuration for backup
2. Supports MIB import from SNMP-mounted equipment
3. Information Extract Function for Report Creation

I am very happy to say that NTT ComTechnology will contribute actively to
ZABBIX SIA continuously for improvement of ZABBIX.

Kind regards,
Koichi

__________________
Koichi Morigami
Service Planning Division
NTT ComTechnology Corporation
Tokyo, Japan

[alixen]

Hi,

with 1.6, you can run a specific shell/batch script for an hoste through the web interface. You can't do an interactive script.

Regards,

[teferi]

I'd ask for a more flexible way of storeing items.

I'd like to see a way to store items more flexible. A bit closer to triggers flexibility. There are often values that I want to multiply or divide by other items latest values and so on.

Will write a more detailed request here or in a separate thread soon.

[Sjeik]

Host hierarchy

I'd like to see a feature to configure a hierarchy in hosts.

In most cases, the Zabbix server is in a certain network (lets call it "net1") which contains a router to the internet ("router1") and some other hosts to monitor.
On some other location there is another network (lets call it "data center") which also has a router to the internet ("router2"). Of course, Zabbix also monitors hosts in the data center.

Now, a graph can be extracted from this information with a load of dependency information. Clearly, if router1 or router2 goes down, there is no need to send notifications about downtime in the data center, simply because the hosts in the data center have an unknown status. Unless there exists a redundant connection, of course.

Bottom line: if a system administrator configures the "parent-child" relations of multiple servers, a load of information and auto configuration can be extracted from that!

Furthermore, this graph can automatically be drawn (just like Nagios does that, but in a more sophisticated way) using standard graph-drawing techniques. I'd be pleased to add such a drawing algorithm to the Zabbix interface, just let me know (by e-mail) if that is desired.

[SergK]

I wish to have items which values calculated from expressions with other items. Something like aggregated checks but with expressions as in triggers.
This is somewhat useful with snmp. For example I have snmp items related to memory:
UCD-SNMP-MIB::memTotalReal.0
UCD-SNMP-MIB::memAvailReal.0
UCD-SNMP-MIB::memBuffer.0
UCD-SNMP-MIB::memCached.0
And there is no item that show how much memory really used by processes on system. But it can be calculated as:
memTotalReal - memBuffer - memCached

[onslo]

More efficient auto-discovery using distributed arpwatch agents to prevent blind subnet scanning as per this thread :

http://www.zabbix.com/forum/showthread.php?t=10173&highlight=arpwatch

This would make bring Zabbix usage into areas where frequent changes to the hosts and topology occur. Such as support labs and research labs where monitoring is still a requirement and useful as a diagnostics tool.

Onslo

[tchjts1]

Screens. We rely heavily on them to identify under/over utilized servers, compare like-servers for usage, trend data, etc.

As it is now with the use of dynamic items, it leaves a lot to be desired and we have even stopped using screens with dynamic items and are back to doing each screen manually for each server.

Dynamic items really limit what you can graph out from host to host as it is all dependent on a default host. Also, with dyanmic items, you can no longer click on a graph for an "expanded" view.

It would be great to be able to create a template for your screens, or to even be able to clone a screen from server to server.

[kyi]

• Acknowledge function
• Maintenace function
• Check Item at specific time

[arli]

Ability to define graph names using ifName or ifDescr grabbed from SNMP

http://www.zabbix.com/forum/showthread.php?t=8130

"All graphs" option for each host in graphs. At the moment I'd have to make a template for every host

http://www.zabbix.com/forum/showthread.php?t=3064

[ronnyromero]

1- Add 95%, 90% and 25% averages to the traffic utilization graphs (view image004.gif attached)
2- +1 on hierarchy in hosts
3- +1 ifName or ifDescr grabbed from SNMP
4- Maintenance button
5- NETFLOW ANALYZER (v10, IPFIX IETF standard)...see:
Views (by router, protocol, interface, application, endpoint, ip address gorup, country, domain)
Blocks Inside Each View (all include graph, list, interface, bytes used, utilization of configured speed to the interface):
a) top endpoints
b) top conversations
c) top countries
d) top sources

[byronsmith]

Can't believe no one has mentioned this.

I would really love to see some better SNMP trapping. i.e the ability to import MIB's which can translate traps and then link them to hosts within ZABBIX for triggering etc.

The trap functionality currently is a little limited.

Thanks
Byron

[gmicalzoma]

Wish List after evaluation of several Open Source NMS

Alexei & Community, 1st Congratulations on such a great project. The following wish list was composed after installing and testing several Open Source NMS projects for two Enterprise clients:

1- A simple way to install, a binary installer with necessary dependencies for several platforms can help automate the install and make Updates from Installer version to another easier. Examples: Zenoss InstallStack, Osmius Binary Installer. The install instructions on the Zabbix Manual are not all inclusive, I had to post on this forum to finish my install on CenOS 5.2 and the instructions from Mucknet site someone suggested enabled me to finish.

2 - Wizard for SNMP or integrate the Community Name and Version (v1,v2,v3 or Unknown=test all) on the form to add a Host (a little button to test SNMP connection will not hurt either). These info could be used for the SNMP Walk. Support to get info from SNMP Informant Agent if it is installed in the machine by default

3- Utilize WMI for Windows. This can expose most info on a Windows (Installs, apps, services, contact info, etc.) and part of these info can be used to gather automatically the entries for Inventory. Actually this can create an ITIL Change Management Database (CMDB) as the Inventory!

4- Reports! The ability to create SLA, Availability, etc. HTML and PDF reports in a Human readable and printable format, A4. For a good Example you can check OpsView's or OSSIM Reports. They are formatted as Executive Summay Style so that Functional Managers as well as Business Process Owners/Users can understand them.

5- More AJAX and customization of the Dashboard. The use of configurable Widgets (applets or panels) that you can add and setup on your Dashboard as your default View. Maybe you want to have a panel to the top right with Events and under that a panel with Network Maps. On the top left a panel with a Service Group and under that a couple of graphics for devices you are closely watching. As an example you can check Zenoss.

All of the testers at both clients as well as our group, were enchanted with the easiness of use an logical display. You can configure every Widget to show what you want to monitor the way you want it. Another improvement we saw on Centreon 2 (Nagios based) where by means of AJAX you could get an ID Card of most necessary info about a host by hovering the mouse over its name much like you do on the top alerts summary in Zabbix 1.6.

I know I have mentioned several names of more than the dozen projects I evaluated, but I can assure You and anyone reading this post that Zabbix features, granularity of configuration and openness is hard to match. Hats off to You and the members of the community that help make this such a great project.

[troffasky]

Interface improvements

It would be great to have on every configuration sub-tab and sub-menu:

- Tick boxes for multi-object operations, like Copy Selected To, Mass Update, Poll Host Now
- Search filters

So for example, when creating a new graph I'd like to be able to, in the Add Items sub-menu, search through the items so I can enter, say if*01, and add ifInOctets01, ifOutOctets01, ifInDiscards01, ifOutDiscards01, ifInErrors01 etc to the graph all at once. Right now in the Items sub-menu I can only select one item at once, and this makes adding multiple items to a graph a long and tedious process.

- Search patterns

Preferably regexps, but failing that, at least some rudimentary support for wildcards, not, and, or, less than, greater than, etc. I'm sure PHP provides a regexp library :-)

- Exportability, clonability, copyability, templatability

It should be possible to export/import, clone, copy and template every object through the web interface.

[troffasky]

Programmatic operations through the web interface

This is a hard one to get right. Anyone who's used a Mitel 3300 web interface will know what I'm talking about. What I'd like to be able to do is add multiple objects at once, and increment the values. For example, I'd like to add ifHCInOctets1 - ifHCInOctets48 to a template. This would involve incrementing at least three objects per item added [Description, OID, Key], which makes it something of a tedious data-entry job. Obviously it doesn't have to be limited to incrementing - I'd want to be adding In and Out values for all the given keys, so being able to do that at the same time would be useful. In bash, I can express it fairly simply:

for ((i=1; i <=48; i++))
     do echo ifHC{In,Out}Octets$i
done

When executed this will expand to print all the strings I want. So, how do we translate this to a web interface? Right now, if I want to do things like this, I

1. - create the first objects of the sequence
2. - export it to XML
3. - pull out the fragment of code I want [eg the <graph></graph> bit]
4. - slap it in a bash script [ie put it after the 'echo' in the script above]
5. - substitute the bits I want to be incremented with $i
6. - run it, redirecting output to a new .xml file
7. - paste the rest of the <zabbix_export ...> code around it
8. - re-import it into Zabbix

Phew! Tedious, but nowhere near as tedious as manually creating every single one. The 'Clone' button is a step in the right direction, but unfortunately it's only one step, and some operations like this could take thousands of steps [48 ports on a switch, 12 objects per port, you get the picture]. Lets just call it an anti-RSI tool :-)

[Zebedee0026]

Agent Utils

1. Being able to push agents from the web interface
2. Being able to pull (click to download) the agents from the main webpage,
3. A web agent configuration page, to customize the zabbix_agentd.conf then push or pull to the monitored computer.