Active IQ Unified Manager Discussions
Active IQ Unified Manager Discussions
Hi All,
After recently upgrading to WFA 2.0 and having a poke around the database I noticed a schema for 7-mode performance data. Unfortunately the tables for array_performance and aggr_performance are empty and don't seem to be updating when WFA hits up the DFM server. Is there any additional configuration I need to do in either WFA or DFM to get this working?
There is separate DataSourceTypes introduced in WFA 2.0 for PA 7-mode and C-mode.
You need to do PA data acquisition using these types with server(DFM) credentials, not the DB credentials.
Hope it helps.
Warm Regards
Sivaprasad K
Thanks Sivaprasad,
I've created the new Data Source for the performance data but when it tries to acquire the data it runs for about 8 minutes then returns with the following error message:
Error running data acquisition script: Exception calling "InvokeElem" with "1" argument(s): "host-list-info-iter-next: Iterator cannot be obtained."
I changed the data source to our other DFM server and the command successfully completed, however that particular DFM server doesn't collect performance data.
Anyone come across this issue before?
Hi,
I'm still trying to figure out your first case when it's start giving this failure..!
In 2nd case, can you please recheck the below PA options are enabled in your DFM server..!
C:\Documents and Settings\Administrator\Desktop>dfm option list perfAdvisorEnabled
Option Value
------------------ ------------------------------
perfAdvisorEnabled Enabled
and in, #dfm host diag <host> , the PA options are set like below,
Performance Advisor Checklist
perfAdvisorEnabled Passed
hostType Passed
hostRevision Passed
hostLogin Passed
perfAdvisorTransport Passed
once you make sure of all these settings, could you please retry the PA acquisition again in both cases..!
BTW, what acquisition you are trying, is it 7-mode PA or C-mode PA.
Warm Regards
Sivaprasad K
I kept trying and it finally managed to pull down the data, which turned out to be about 55,000 rows for both tables, so I'm guessing that was probably the cause of my issues earlier. I think going forward I won't bother with performance data given the sheer volume of information, but it's nice to get it working.
Thanks for your help.
Hi Tim,
I have a few queries to help me better understand your experience with the 7-Mode performance data source:
Is there any more performance related data you'd like available?
With respect to Point 1, WFA provides a neat feature of disabling cache acquisition for specified dictionary entries. This feature was primarily developed to address the issue of significant volume of data acquired per dictionary entry, which in your case happens to be Aggregate_Performance and Array_Performance.
However,in case you are working with any one of the objects, you can add the other to the "Disable acquisition for Dictionary Entries" list (Tools -> WFA Configuration -> Other). This would delete all current data acquired and disable acquiring data for the dictionary object in subsequent acquisitions.
Regards,
Tanzim
Regards,
Tanzim
Hi Tanzim,
The sheer volume of data is a big issue, we currently have 58 arrays and 280 aggregates, all of which are polled by our DFM server once an hour for performance data. But I think that the biggest reason I've deciding against implementing it is because the data acquisition seems spotty. More times than not it fails with the error:
Error running data acquisition script: Exception calling "InvokeElem" with "1" argument(s): "host-list-info-iter-next: Iterator cannot be obtained."
Originally I had planned to use the data in conjunction with our aggr selection finders to find the best aggregate to provision storage on based on the performance data. I find that the pre-canned data is more than enough for what I wanted.
Tim,
Thanks for providing the requested information.It has provided a good insight.
I'll get in touch with you offline regarding the performance error.
Regards,
Tanzim
Did anybody find a resolution to this? I'm getting the same error with a slightly modified version of the PA collection script.
Can you let us know the WFA version where this issue is hit for you ?
There was a fix done for this problem in WFA 2.0.1
Thanks,
Shailaja
I hit this error on a 2.0 instance. It seems to have gone away. I think it was just a timeout issue. The successful query was 412 seconds. The timeout was 600.