08-14-2012 06:12 AM
Our management server is running consistantly at 100% with two instances of report.exe equally consuming 50% each. My quess is that is updating itself from the over night scans???
This is a windows VM with 2 cores, did I undersize the vm?
Looking to understand if I've misconfigured something, etc..
Thank you!
Solved! Go to Solution.
08-15-2012 02:54 PM
Hey Randy I am on a development machine, that was meant for me to confirm I had a new enough version to show you the command. By not coming back upon reboot do you think it may be a scheduled thing and return later or you think the topic can be closed and was an overloaded box using up the rest of the resources available to run the report component?
Rod
08-15-2012 12:09 PM
Randy:
I see from the case you had opened that you are installed at the 3.0.1 version which includes all the previous patches to optimize the reporting tool to run as efficiently as possible.
It could well be environmental or a cross query in a report. I would assume these are canned reports or paths, users or DFS you are running.
I would suggest we open a support call in this instance and not discuss your environment online.
Quickly with previously provided information and a generic response in nature:
The report.log would provide information on the running reports. You may have to sort it if you run multiple simultaneously. The scheduled reports will be present to investigate which one is running under the directory:
C:\DataInsight\data\console\reports\reportruns\342\engine\type.properties
That is C:\DataInsight\data\console\reports\reportruns sort by modified time and choosing the current report then checking the .properties or .xml file in the engine directory under engine or *puts (like inputs or outputs).
An example:
#
# Report Type attributes
# version 1.0
#
category=Dashboard Reports
type=summary_dashboard
description=Use this report to compute data displayed on device and msu summary dashboard.
fields=resource,usergroup,period,user_count,start_depth,end_depth,purge_count
name=Dashboard Summary
executor=DIMultiTierExecutor
report.merge.exe=merge_rpt.exe
hide=true
format_output.skip=true
There are also alerts and system reports which have a large number.
Example:
C:\Program Files\Symantec\DataInsight\bin>dir /P /OG /Q C:\DataInsight\data\console\reports\reportruns
Directory of C:\DataInsight\data\console\reports\reportruns
08/14/2012 12:03 AM <DIR> BUILTIN\Administrators 1000000000335
08/15/2012 12:03 AM <DIR> BUILTIN\Administrators 1000000000336
01/17/2012 02:37 PM <DIR> BUILTIN\Administrators 101
01/17/2012 02:39 PM <DIR> BUILTIN\Administrators 102
01/17/2012 02:42 PM <DIR> BUILTIN\Administrators 103
The long ones are system related and are generated for integration between SDI (Symantec Data Insight) and DLP (Data Loss Prevention).
Reports increment in number.
Check which jobs you have configured using:
C:\Program Files\Symantec\DataInsight\bin>configcli.exe version
3.0.1.2537 (Rev: 3687-4530-723) (Built: 06 Apr 2011 10:07 PM) (Patch: 30RU1)
C:\Program Files\Symantec\DataInsight\bin>configcli.exe list_jobs webserver
|
Your configclidump.txt file you sent in to Joe has the parameters you are operating the SDI product under.
Windows 2008R2 and 1GB (Xmx1024m) on a 64-Bit Server VM on AMD
I know of no known issues that would keep the CPU pegged but there are other environmental considerations.
note: long version ' configcli.exe dump webserver' pipe to a file to capture
I hope that leads you in the right direction.
Rod
08-15-2012 12:46 PM
Hi Rod, thanks for all the information and I appreciate knowing about configcli - very helpful.
I ending restarting the server and the condition did not come back and I was working with Cliff #600-832-865 and never sent a dump file, but I like the command.
And it looks like I have a different version of configcli.exe than you have.
3.0.1.2511 (Rev: 3687-4530-723) (Built: 06 Apr 2011 10:07 PM) (Patch: 30RU1)
08-15-2012 02:54 PM
Hey Randy I am on a development machine, that was meant for me to confirm I had a new enough version to show you the command. By not coming back upon reboot do you think it may be a scheduled thing and return later or you think the topic can be closed and was an overloaded box using up the rest of the resources available to run the report component?
Rod
08-16-2012 07:44 AM
I would agree with a scheduled thing..
08-16-2012 08:31 AM
In that case Randy you will see a recurrence upon hitting the next schedule and the job will be listed under rows:
ReportJob |
ReportJob_90 |
in the table above. You could continue with your support case and provide that information for review and explanation via email.
Rod
08-16-2012 08:40 AM
Okay, Thanks. I'll keep my eye on it.