Call for discussion on ZABBIX 1.8

[cjwallace]

I would like to see if this is not already part of Zabbix the following and it is rather critical.

When you have been alerted to something in your Windows Event Logs that you are emailed the actual item. I would like to see all the information that i would see if i looked at it in Windows or if i was to view the data in latest data as plain text.

The information is there already.

I would like to see more how to in the documentation and for the documentation to be more clear and not assume. for instance not once in the documentation does it say you might have to set a mysql password but all the how to documents that people write for example how to install on SUSE or Ubuntu that information is there.

We have a great product here but are very close to people not wantng to use it because of the lack of information. I keep time and time again walking away from Zabbix but i just cant find anything that ticks the all the box's for me. I keep chipping away at Zabbix and it is starting to do what i want but i am not sure how many people would have the time to play with it asmuch as i have.

Its a great product but somethings just dont add up

Craig

It may be a good idea to really get working what we have before making loads of new features.

just my 2p worth

[tmine]

Test function for trigger

Hi,

Suppose test function for each trigger items will be appreciated.
If each trigger can be tested not real value, but input by user.

[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.

I'd call it a function item and store it as a separate item type, let's say "functional". And in this items I'd like to do almoust all sorts of things I can do in triggers. For example store the item as delta and divide it by another items last(0) and so forth.

Another request is for multiple items being collected by 1 agent check. These items should be grouped(maybe they'll be of separate type - "grpcheck") and they should have one update interval for the whole group. With the option of
sending multiple values per connection this should be very handy.

[SergK]

I don't know is this major feature or not, but it will be very useful to have ability change SNMP community or password for SNMPv3 for items from template. Now you can only set password in template and it will be same for all all hosts used that template.

[kehall]

Excellent work chaps, and here's my list (in no particular order).

1) Ability to use graphs in screens with independent timescales (i.e. you could show last hour and last 24 hours as two separate graphs
2) Ability to 'roll over' legend to highlight a line in a graph (like PRTG) - appreciate this may be difficult as graphs are pre-generated aren't they?
3) Better queue handling (often find entries in > 5 minutes that sit there for ages, then eventually get processed), particularly SNMP. This leads to huge gaps in graphs, missing alerts etc.
4) Remember the 'all' view setting on monitoring tabs visited (only specific hosts are remembered)
5) Add 'Event Age' condition to action conditions
6) Add all the available action conditions to escalation conditions (only event ack is there at the moment)
7) Logfile to include more info on what the error relates to (item etc)
8) Stop logfile filling up with reports of 'truncating unsigned value to 32 bits (12)' for SNMP requests (on 64 bit Ubuntu 7.10 system)

Is it possible to alert based on Zabbix queue size?

Keith

[teferi]

and one more thing - do not remember if it was mentioned here or not.

A customisable legend (and title) for graphs would be great.

Often I do not want to see a line repeated 20 times if I have a graph that combines 20 similar items from 20 different hosts. And the name of the first host's item is not always a good title for a graph, that has items from 20 hosts.

[alixen]

Better IT services

It would be nice to :

- Be able to select a period of time for SLA calculation different from 7 days. At least 1 day, 1 week, 1 month, 1 year would be nice.

- Be able to display IT Services page and compute SLA for a specific date range. For instance, display SLA for last month or one week, 2 weeks ago.

- Filter IT Services shown based on user rights. Currently (in 1.4.6, I have not tested 1.6 yet) all IT Services are visible to everybody.

- Be able to display a subtree only in IT services page. So that only services relevant to the user are displayed.

- Don't compute SLA when there are no valid data (see my previous post http://www.zabbix.com/forum/showpost...9&postcount=15)

- Be able to export SLA results (CSV would be fine)

- Printable SLA reports (PDF ?)

- Be able to insert SLA results in screens. It would be nice to generate a screen that mix SLA results for a group of routers with bandwidth usage graphs.

- Document how IT Services work

Regards
Alixen

[DoubleP]

Highlighting Graph Items

I would like to be able to click a graph item in the legend and have its corresponding graph [line, point..etc] become highlighted as a way to easily distinguish a specific item. Would be great in a graph with many items...


Feedback?

[mactfines]

Requests for 1.8

1. In Slideshow mode, would like to have the typical "<< < O II > >>" (full rewind, back, stop, pause, play, full fast fwd) slideshow buttons.
2. Would like to have a 'random' slide show order option.
3. When adding Users to Groups, would like to be able to check box next to multiple users then click 'Add', like one may do with items & hosts on other parts of the system. Adding one-at-a-time gets tedious.
4. Would like to be able to set minimum refresh time a user may set, as some users will set high refresh rates that make no sense and put too much load on the server.
5. Would like to be able to have more control and customization options for how graphs are rendered.

[troffasky]

Make CfD on 1.8 a sub-forum

The Call for Discussion on 1.8 should be a sub-forum rather than a thread, because I see a lot of interesting ideas being raised that result in tangential discussions with further interesting ideas in them; these could easily get lost in a single thread. With a sub-forum, each idea could be its own thread, and related ideas could be kept together.

[alucard95]

Hello

Hello all!!

I use zabbix to monitoring my switchs park, i have a global screen and when i click on a switch i can see the screen of the switch with load CPU, memory, port up and down,...
But the problem is with the state of the port, if the port is up, it's true and if he's down it's false but we can't really consider that a port down is a problem.

So it can be very interesting to manage more than 2 states in the triggers, with my example it's allow us to have the state up, down and problem.

Congratulation to all the team for your work!!

[exkg]

Some ideas about web monitoring

Hi,

We have some web aplications and all things are going ok with web monitoring. But we have a application that create a 'ID' when user does login, and (to scenario continue) we need 'get' this value to post in others steps. Idont know if its possble to do with ZABBIX.


[]s,
Luciano

[exkg]

About IT Services permissions

Aly,


When we've created distinct users we need show only IT Services that have some bond whith their hosts (that user have read access). I mean: we dont want that all users can see all IT Services.



[]s,
Luciano

[n0name]

I would like to see in 1.8:

- Shedule reports;
- Export SLA in PDF (or any other format);
- Maintenance mode for host, group (with sheduling);
- Export graphs to jpeg or gif.

Thank you

[alixen]

Computed items

It would be nice to be able to define "computed items" as a function of other items.

For instance:
- Zabbix agent defines vm.memory.size[total] and vm.memory.size[free] but there is no "total memory used". It would be nice to define an item as used=vm.memory.size[total]-vm.memory.size[free] and be able to draw it or define triggers based on its value.

- When an item is defined, it can be stored "As is", "Delta (speed per second)" or "Delta (simple change)". It would be interesting to create one item stored "As is" and define two "computed" items as item_speed=delta_per_second(item) and item_variation=delta_simple(item).
This can be achieved now be creating 3 items but there are two drawbacks:
- it will generate 3 requests instead of one
- it will need two times more storage than necessary
- it must be planned from the beginning when a new host is setup. With computed items, it would be possible to define it afterwards and apply it to all stored values.

Alixen