WinXP-Win8, Win2000-Server 2012 x32/x64 Templates

[candy]

hi G0dSmack,
can you send me a copy of you test software?it's look so cool
thanks


_________________
lv handbags

[G0dSmack]

Anyone found any issues with these? I am open to any ideas as well.

I like the look of the graph templates. I may have to borrow some of the ideas and port them to use WMI

Feel free man....Let me know how they work and I may switch over to using WMI as well.

[G0dSmack]

When I try to import the templates I got the following error:

Code: Select all

Error: XML: Hash version does not exist.

Same problem here... Cacti 0.8.7b

Solved by upgrading to 0.8.7d.

Now it is another issue....

Code: Select all

ERROR: opening '/usr/home/cacti/rra/10_0_12_12_yosh_515.rrd': No such file or directory 

Looks like .rrd files aren't generated...

I am actually using version 0.8.7c without issues. Did you guys get this figured out. Sorry I have been away for a while again.

[G0dSmack]

hi G0dSmack,
can you send me a copy of you test software?it's look so cool
thanks


_________________
lv handbags

You get the templates to work?

[spike100]

Hiya,
Downloaded the templates the other day. Very straight foward to set up and use. Had them up and running in no time. Great work!

They've only been running for just over a day and I've spotted a strange graph... On certain disk Percent Usage graphs the Disk Read time plus the disk write time is more than 100%. The Read and Write values appear to be 'stacked', which is great as long ay they don't add up to more than 100

Either I've got a problem with my values

or

The fact these are SAN drives and the rear/write cache on the SAN is skewing the figures...

Anyone have any ideas?

Spike

[G0dSmack]

which templates did you grab? The last ones that I posted?

[spike100]

which templates did you grab? The last ones that I posted?

Yep, I used the ones from your 5th March posting.

I've been thinking about this over night and I think I now understand why it's happening. The SAN drives in question are dual ported so therefor are full-duplex. This would permit read and write requests down both channels and could generate the situation where the combined read totals add up to more than 100% when drawn as 'stacked' (ie added together) graphs. This wouldn't happen for traditional, single ported drives.

I'm not very familiar with Cacti yet, I'm new to it from MRTG. A fix feels a bit tricky to implement as I have a mixture of single and dual ported drives. If the vaules were plotted non-stacked (ie not added together) single ported drives vaules would have a maximum of only 50% for each value. Alternatively the read write values for dual ported drives could be divided by two, but that would only apply to a certain number of drives.

Your clearly have far more experience with Cacti than me, do you have any ideas on a solution? I've yet to delve too deeply into configuration of the graphs within Cacti

Cheers

Spike

[ahtshun83]

hi godsmack,

i tried to import the template from Cacti web console but nothing happen...how do i know whether it is succesful or failed...no error on log...

pls help..i intend to use to monitor Win2k3 memory usage....

[G0dSmack]

hi godsmack,

i tried to import the template from Cacti web console but nothing happen...how do i know whether it is succesful or failed...no error on log...

pls help..i intend to use to monitor Win2k3 memory usage....

Did you have any luck? Follow the steps a few pages back.

[G0dSmack]

which templates did you grab? The last ones that I posted?

Yep, I used the ones from your 5th March posting.

I've been thinking about this over night and I think I now understand why it's happening. The SAN drives in question are dual ported so therefor are full-duplex. This would permit read and write requests down both channels and could generate the situation where the combined read totals add up to more than 100% when drawn as 'stacked' (ie added together) graphs. This wouldn't happen for traditional, single ported drives.

I'm not very familiar with Cacti yet, I'm new to it from MRTG. A fix feels a bit tricky to implement as I have a mixture of single and dual ported drives. If the vaules were plotted non-stacked (ie not added together) single ported drives vaules would have a maximum of only 50% for each value. Alternatively the read write values for dual ported drives could be divided by two, but that would only apply to a certain number of drives.

Your clearly have far more experience with Cacti than me, do you have any ideas on a solution? I've yet to delve too deeply into configuration of the graphs within Cacti

Cheers

Spike

You get it figured out. I actually ended up seeing something the same as you did. The Reads were almost 100% and so were the writes. So it was almost 200%. This has only occurred one time for me out of about 50 servers that I monitor using these templates.

[Lien]

I've used your templates to get information from my P.C servers, but it looks like some of the servers are reporting the wrong information back.

I have deleted the RRA files & removed, re-added, re-booted the servers & cacti box, but they still give strange information back.

Example of this is the C drive on one of the servers is show as 40G & 0GB used. This machine only has a 15GB C drive & 8GB is used. The graph for the D drive is fine.

The thing I find strange is that I am monitoring 16 P.C's & its only 4/5 of them that are reporting the wrong information. Where do I need to look to start de-bugging this problem? There are no errors in the cacti.log file. I would assume the problem is with the servers & not with cacti or your templates. If the problem was your templates or cacti I would have a problem with all of the machines

[G0dSmack]

Post a screenshot and let me see what you are looking at so I can try to figure out if I can duplicate it.

[Lien]

Post a screenshot and let me see what you are looking at so I can try to figure out if I can duplicate it.

I was using the 'Host MIB - Available Disk Space' template & not yours. So I think the problem was between the chair & keyboard....

I'll run your template for few hours & see if I get the same result. You can tell after a few polls it was returning the wrong information back with the other template.

[G0dSmack]

Cool. Let me know if you have any issues with my templates. And good luck learning this great monitoring tool.

[neroZ]

Same problem here... Cacti 0.8.7b

Solved by upgrading to 0.8.7d.

Now it is another issue....

Code: Select all

ERROR: opening '/usr/home/cacti/rra/10_0_12_12_yosh_515.rrd': No such file or directory 

Looks like .rrd files aren't generated...

I am actually using version 0.8.7c without issues. Did you guys get this figured out. Sorry I have been away for a while again.

Hi, G0dSmack.

I see... But last stable version is 0.8.7d, and a lot of other templates/plugins REQUIRE updating to last version. For example, in my system I have couple of plugins, which working ONLY with 0.8.7d. So, there is no possibility to downgrade, and actually - there is no reason. Your templates are very good, but, please, let them work with last stable version.

thank you for your contribution!