Cheshire Cat Computing

The original files have hidden carriage returns in them "^M". either vi and search/delete "^M"; or if you use putty, cat the file; then cat > filename Ctrl + D

Ha, just found out command "dos2unix". Give that a go.

Hi,

I am using check_vmware 1.13 against multiple standalone ESX 4.x servers.

I works just fine for what I want on all servers, except one.

On that server(4.1.0) when I query for any report, except disk, I get the error:
Perf stats not available at required interval (300s) or invalid instance.

For a disk report I get results just fine.

The performance settings for that host seem to be the same as any other hosts'.

I understand that you have not done much testing on non-vCenter scenarios but could you help me troubleshoot this?

Thank you in advance.

After 2 years as IT Consultant, I'm back in business as Sysadmin.

I only just want to inform you, that your Plugin is working with vSphere 5 without any problems!

Currently I'm monitoring 1 VCS and 4 ESXi Servers with Icinga.

Best wishes from Hamburg and Thank you for your great work on this plugin!

Dennis

Thanks for the feedback! We dont have vsphere5 here yet so its good to know that it will support it when we do...

_________________
Steve Shipway
UNIX Systems, ITSS, University of Auckland, NZ
Woe unto them that rise up early in the morning... -- Isaiah 5:11

Sorry to say, but I do have problems with ESXi 5.0.
If i query our VCenter it works for all the reports.

CPU usage at 1.01%|cpu=1.015%;80;90;0;100
Network usage at 328Kbps|net=328;7680;10240;0;

But if i query the hosts, cpu and net goes through the roof.
CRIT: CPU usage at 557265.88% (need more CPU allocation?)|cpu=557265.88%;80;90;0;100
CRIT: Network usage at 2326066096Kbps (need more resource shares or physical interfaces?)|net=2326066096;7680;10240;0;

Memory and disk works on the hosts thou.
I know you don't have access to ESXi 5.0 but i thought i should inform you. I don't know how to fix this either.
I'm using version 1.13 of check_vmware.pl

You don't have these problems neo-mycin?

Thanks for a great plugin otherwise.

Now THIS problem I have replicated. It is a BUG in the VMWare API on some versions of VCentre -- the variable that SHOULD be returning a CPU percentage (0-100) is instead returning a MHz count (the MHz count variable is still returning MHz correctly). There is nothing I can do to work around this other than detect a ridiculously high value and simply not display CPU under these circumstances.

Ive tried to escalate this bug report to vmware but it disappears into a black hole

_________________
Steve Shipway
UNIX Systems, ITSS, University of Auckland, NZ
Woe unto them that rise up early in the morning... -- Isaiah 5:11

OK, thanks stevesh for your reply. I'll turn my ESXi servers to v5 tomorrow and will check if the behaviour changes.