<tracePacket>
<businessObject>Ice.Proxy.BO.UserTracingImpl</businessObject>
<methodName>GetNewOrGetByID</methodName>
<appServerUri>net.tcp://jcaeic-erpapp01/EpicorLive/</appServerUri>
<returnType>Ice.Tablesets.UserTracingTableset</returnType>
<localTime>7/19/2019 11:04:49:2571595 AM</localTime>
<threadID>1</threadID>
<executionTime total="95" roundTrip="27" channel="0" bpm="0" other="68" />
<retries>0</retries>
<parameters>
<parameter name="UserID" type="System.String"><![CDATA[jperena]]></parameter>
</parameters>
So that function took 95 ms.
But that is just the time it took to start the tracing function. Scroll down through the log until you find a function tracePacket with a large execution time (something bigger than 3000) .
Just kind of skimming through this threadā¦Iām not a network gal but could it be your Anti-Virus Program running a scan that could be slowing it down? I think weāve run into this issue in the pastā¦I think. 
How does the client react when running on the application server itself? Is the application server and the database server different? Are they on the same network? How do they connect for database calls between the appserver and the database server? Is the client when run on the application server just as slow or faster?
Lots of good information going here, just read down the whole thread. Has it slowed down recently? @ERPSysAdmin has a good thought on anti-virus software. What type of hardware is supporting your server and clients? I also do not believe you have a network issue.
2 Likes
This is one of the larger execution time
<tracePacket>
<businessObject>Ice.Proxy.BO.ReportImpl</businessObject>
<methodName>GetReportStyleRules</methodName>
<appServerUri>net.tcp://jcaeic-erpapp01/EpicorLive/</appServerUri>
<returnType>System.String[]</returnType>
<localTime>7/19/2019 12:37:13:9914662 PM</localTime>
<threadID>1</threadID>
<executionTime total="21028" roundTrip="0" channel="42518" bpm="0" other="-21490" />
<retries>7</retries>
<parameters>
<parameter name="reportid" type="System.String"><![CDATA[JobPickl]]></parameter>
</parameters>
</tracePacket>
Yeah thatās not normal. 21 seconds just to get the list of style the user can select from is WAY too long. Should be a few hundred milliseconds at max.
Doing the same exact thing on a computer on Lan1, yields what kind of value?
It is slow even we run the client from server
We have separate apps and database server. These are on Hyper-v. The resource utilization is low.
It was slow from the launching of Epicor since last three months
We already tried by disabling the Anti-virus and the server backup
The server is DELL EMC PE440
We havenāt tried it on other LAN. Is this because of network?
I do not believe this to be the network. IMO
This looks like a communication issue between the appserver and the database server. It could also be a configuration issue with the database server. How much RAM do you have allocated to the virtual running the database server? Are the hard disk used in the server SSD or Spinners? If they are spinners are they 15K or 10K rated.
What report where you trying to run? I was going to look at my speeds in a trace on the same method call.
1 Like
Total RAM allocation is 48GB for each server. The hard disk is SSD. This slowness is not related with generate any specific reportsā¦if happen for anything user do in the Epicor.
Are both virtuals on the same host Windows server?
I just read through this and I think people missed this item.
I think your Server setup has issues.
There is a performance diagnostic tool to help find a few things, but your issue is not network IMO.
Weāre still on 10.1.400, and donāt see as much detail about execution time in the log file. Can someone elaborate what each param in the newer <execution time> entry means.
The OP had one of
<executionTime total="21028" roundTrip="0" channel="42518" bpm="0" other="-21490" />
Total appears to be the sum of the other 4 params.
What does channel represent?
How can the other value be negative?
Iād highly recommend you open the admin console on the app server, install the performance & diagnostics tool by clicking the button on the screen where it lists version etc, run the tests from the tool on the sever, and post the summary diagnostics results here after running the config check module. It wouldnāt hurt to run the network diag either. You can also perform a client trace from a computer that is having the issue, copy the text file to the server, and load that file into the perf and diag tool using the client trace analysis tool. The more information we can get from the available tool-set the more we will be able to help you here. Itās almost impossible to troubleshoot issues like this without knowledge of your network config, ERP platform config, being remote etc.
Great thanks for all that info, looks like your app server should be okay but may want to check out your SQL server config. Can you perform the performance tests with the client on that app server and post the results here? There is a section in help that will walk you through baseline performance tests like switching between sales orders etc.
Hi Jeff do you mean to run the test from any client PC?
Can you launch and run the client on the app server?
Yes, I can