Ad blocker detected: Our website is made possible by displaying online advertisements to our visitors. Please consider supporting us by disabling your ad blocker on our website.
Here is the issue I face. The Data Queries on the Cacti console display "This data query returned 0 rows", but when I run the same queries on the Cacti server, I get the expected results from the EVA. This happens with all data queries. Can you please help me with that? Thank you in advance for your effort.
I have exactly the same problem on both our EVA's,
This data query returned 0 rows
but when i run the same query from the server console :
I had this exact same issue.
The problem in my case was:
- cacti poller running as cacti/cacti (u/g)
- apache running as httpd/httpd (u/g)
Now all the existing disks were queried fine, but adding new graphs always resulted in 0 rows.
How to debug:
- become httpd user and issue the same command (eg: query_wmi.py --host <hostname> --authentication-file /etc/cacti/<hostname>.pw --class Win32_PerfRawData_EVAPMEXT_HPEVAVirtualDisk index)
I got the error: ERROR: Could not write to logfile /var/log/cacti/query_eva.log
Make sure that the query_eva.log is writeable for the user cacti AND httpd (i set it to: -rw-rw---- 1 cacti apache 121907650 Dec 27 13:10 query_eva.log)
Now you should be able to create new graphs again.
Cacti has imported the following items:
+ Unmet Dependency: (Data Template Item)
...
The result is, that I have empty data queries, but of course the script (query_wmi.py) returns exptected and correct values.
First of all, thank you palli for an excellent contribution!
I get the same unmet dependencies on Template Import and I guess that's why some graphs aren't working.
There is also a problem with the management appliance server where WMI seems flakey.
WMI stops working after a while and I get no data for my wmi queries, the workaround so far is to restart the service Windows Management Instrumentation.
palli wrote:
You can try these (in the following order)
* Restart HP Evaperf service
* Run winmgmt /resyncperf
* Run wmiadap /f
* Restart Windows Management Infrastructure service
* Give HP a call, ask them to give you a hand
Nothing helps except Restarting Windows Management Infrastructure service.
Guess I'll try calling HP soon, has anyone else had any luck with this?
Symptom is that I have to restart the WMI service once every hour.
Everytime I've restarted the WMI service, the wmic client stops getting data after an hour.
mikaelf wrote:
Symptom is that I have to restart the WMI service once every hour.
Everytime I've restarted the WMI service, the wmic client stops getting data after an hour.
Hey mikaelf,
I think the problem of having to restart the service every hour is on quite every system running the eva monitoring.
I createt a batch script put it into planned tasks so that it runs every half hour.
net stop "Trend Micro Client/Server Security Agent Listener"
net stop "Vmware Authorization Service"
net stop Windows-Verwaltungsinstrumentation
reg import evaperf.reg
net start Windows-Verwaltungsinstrumentation
net start "Vmware Authorization Service"
net start "Trend Micro Client/Server Security Agent Listener"
lodctr /R
wmiadap /F
all you have to do is to resolve, which services DEPEND on your WMI service , start and restart it in the right order and then run lodctr /R (reloading the HPEVAPMEXT if it gets lost) and wmiadap /F
My file is currently in German but you have to rename the services to whatever your windows language is.
net stop "Trend Micro Client/Server Security Agent Listener"
net stop "Vmware Authorization Service"
net stop Windows-Verwaltungsinstrumentation
reg import evaperf.reg
net start Windows-Verwaltungsinstrumentation
net start "Vmware Authorization Service"
net start "Trend Micro Client/Server Security Agent Listener"
lodctr /R
wmiadap /F
all you have to do is to resolve, which services DEPEND on your WMI service , start and restart it in the right order and then run lodctr /R (reloading the HPEVAPMEXT if it gets lost) and wmiadap /F
Thanks Alpha, I've read your troubleshooting guide and your blog, very nice information.
So far I don't need to do anything else except restart the WMI service:
I'll do this every 30 minutes from now and see how it works out.
Sucks to get gaps in the graphs though, I'm going to open a case with HP anyway because I want this to simply work...
I also tried to get support form HP but all I got as answer was that getting performance via WMI isn't currently supportet by HP! and never will be supported, so they wont do anything for you ...
In my case the gaps in the graphs are very small (i dont even see them when going into half hour view)
But i don't know how big your polling intervall is (I poll every 15 seconds)
altough i have some unmet dependencies when importing the template.
in my case the EVA - HostPortStatistics - Throughput graphs are not working
any change that the someone (creator of this templates) could look at it as more people seem to have this problem..??
at the moment we're running cacti 0.8.7e
Alphadog wrote:I also tried to get support form HP but all I got as answer was that getting performance via WMI isn't currently supportet by HP! and never will be supported, so they wont do anything for you ...
I don't really expect anymore from them.
Alphadog wrote:
In my case the gaps in the graphs are very small (i dont even see them when going into half hour view).
But i don't know how big your polling intervall is (I poll every 15 seconds)
Same here, can't see any gaps, but I'm currently only doing 1 minute polling interval. wow, 15 seconds. How is your highest resolution rra configured?
I've been considering polling faster actually.
My RRAs:
Hello there palli. Do You updates of Your host template by any chance? The one from march gives some unmet dependency errors on 0.8.7g:
Can someone pretty please tell me how to resolve this. I am getting the same errors. Unmet dependency errors on .8.7g. This is my first cacti install on this box so I am unsure if I am able to "downgrade" to an earlier version that does not have this problem.
CDEF
[success] Multiply by 1024 [update]
[success] Total All Data Sources [update]
[success] Ratio, Multiply by 1024 [update]
GPRINT Preset
[success] Normal [update]
[success] Exact Numbers [update]
Data Input Method
[success] Get Script Data (Indexed) [update]
Data Template
[success] EVA - VirtualDisk [update]
[success] EVA - PhysicalDiskGroup [update]
[success] EVA - HostConnection [update]
[success] EVA - StorageArray [update]
[success] EVA - StorageController [update]
[success] EVA - HostPortStatistics [update]
Data Query
[success] Eva - Query new VirtualDisk [update]
[success] Eva - Query new PhysicalDiskGroup [update]
[success] Eva - Query new HostConnection [update]
[success] Eva - Query new StorageArray [update]
[success] Eva - Query new StorageController [update]
[success] Eva - Query HostPortStatistics [update]
+ Found Dependency: (Data Input Method) Get Script Data (Indexed)
+ Found Dependency: (Graph Template) EVA - HostPortStatistics - Requests / Second
+ Found Dependency: (Data Template) EVA - HostPortStatistics
+ Unmet Dependency: (Data Template Item)
+ Unmet Dependency: (Data Template Item)
+ Found Dependency: (Graph Template) EVA - HostPortStatistics - Throughput
+ Unmet Dependency: (Data Template Item)
+ Unmet Dependency: (Data Template Item)
+ Found Dependency: (Graph Template) EVA - HostPortStatistics - Queue Depth
+ Found Dependency: (Data Template Item) AvQueueDepth
+ Found Dependency: (Data Template) EVA - HostConnection
Host Template
[success] EVA Commandview [update]
Everything in this template works great except for the hostportstatistics and the unmet dependencies.
Can someone please lend a hand. THis is broken across our entire environment.
First of all, many thanks to Palli for creating this awesome plugin.
I got it working now for my two EVA's. What i have:
1) Installed Windows 2003 server with Commandview 9.2
2) Installed CentOS 5.6 with mysql/apache (using yum)
3) Installed 1x Cacti 0.9.7g with latest patches
