We have two buildings within close distance around 500m. All of our Epicor users access to the servers which are located in other building. These two buildings are directly connected via dedicated gigabit WiFi connection and the connectivity is stable. Both buildings are in the same network. From the beginning of launching Epicor our users are facing slowness. We did not find any issue in network connectivity. Is this is could be the reason of slowness? If I ping the servers I get response 1ms.
Define “Slowness” Epicor is a lot of great things but it isn’t fast. How does it compare to running the client on the same building?
1 Like
Thanks Jose for your response. We allowed some users to work in Epicor on the same building but got mix output…some of them experienced the same slowness and some of them got it little faster.
ok so once again define slownes, what are you comparing against? what’s your base?
If you are comparing it to your existing non epicor solution then I’d say #MoveOn epicor isn’t fast
You need to compare apples to apples. Can you run epicor faster (significantly) than they (the people complaining) can.
Let me explain the slowness issue happening…it is too slow when user move from one screen to another, sometimes it gets stuck and shows Epicor offline. Sometimw it works fast sometimes slow. It is happening randomly for everyone.Specially it gets worse from the afternoon.
Does building 2 (the one without the App server) have an internal LAN? Or do all the users connect via the WiFi from Building 1 (the one with the server)?
What other communications are carried over the WiFi? Do people in building #2 access the internet via a connection to the ISP in building #1?
All users access to the network via wired connection. This Wi-Fi is a dedicated internal connection. This is not through internet or not from any service provider. There are not much other communication over the Wi-Fi.
Is there an internet connection to the LAN in building #2? Or does LAN in building #2 connect to the internet via the WiFi connection that joins the LANs in the two buildings?
I’m only trying to get a feel for how much other traffic there might be on the WiFi.
What if someone on Lan2 was to remote desktop into a computer on Lan1 (which has the client installed on it). Does the client (which is actually running on the computer in Lan1) seem slow?
There is internet connection in building#2. If any user do the remote desktop from building#2 to other building where is servers are located they face the same slowness in Epicor. I checked the utilization of bandwidth in Wi-Fi and the utilization is not high. Is there any minimum standard of network connectivity to get the Epicor work monthly?
If you do a tracert, are you seeing it routed over the WiFi link or is it going out through the Internet to get to your other building? If you do a Client Trace, what are the times showing in there?
1 Like
Then its not the WiFi
It does not go through internet, both building are on the same network. Here is tracert result.
C:\Users\mhossain>tracert 192.168.x.x
Tracing route to APPSSVR [192.168.x.x]
over a maximum of 30 hops:
1 1 ms <1 ms <1 ms 192.168.x.x
2 <1 ms 1 ms <1 ms APPSSVR [192.168.x.x]
Trace complete.
I also believe the same.
Calvin’s right. It’s not the WiFi. What’s the speed if you take a laptop into the building where the server is?
If you mean the Epicor speed if move the laptop to other building …some of them experience the same slowness and some of them got it little faster.
Find the “slowest thing” (what users say is really slow), and run a trace . Then repeat with that same user doing the same thing on a client on Lan1.
Look in the trace log for the <executionTime ...> entries, an compare the results
There is no specific module which is slow. The entire Epicor is overall slow…We already run the trace log but nothing found wrong…as well as shared the trace logs report with the Epicor support team and they also did not find anything wrong…that’s the reason I posted it here to get the opinion from the forum.
Did epicor support say that the execution times were slower than you should expect?
What is the longest execution times you see in the trace logs?
It’s not unheard of for some data fetches to take as long as 20 seconds.
Here’s just one BO while fetching a customized Report Data Defintion:
<tracePacket>
<businessObject>Ice.Proxy.BO.RptDataDefImpl</businessObject>
<methodName>GetDataDefUsedTables</methodName>
<appServerUri>net.tcp://xxxxxxxx/MATCOR/</appServerUri>
<returnType>DefUsedTablesTableset</returnType>
<localTime>8/23/2019 12:36:59:2916051 PM</localTime>
<executionTime>6753</executionTime>
<parameters>
<parameter name="rptDefID" type="System.String"><![CDATA[ARForm_001]]></parameter>
<parameter name="onlyDBTables" type="System.Boolean"><![CDATA[False]]></parameter>
</parameters>
</tracePacket>
That 6753 is 6.753 seconds. And there are 3 or for BO cals that take this long. So it takes upwards of 20 seconds to retrieve that RDD.
Here it is
Ice.Proxy.BO.UserTracingImpl GetNewOrGetByID net.tcp://jcaeic-erpapp01/EpicorLive/ Ice.Tablesets.UserTracingTableset 7/19/2019 11:04:49:2571595 AM 1 0Add ``` text above that part of the trace log you pasted, and then ``` after it.
like in the following pict
That will make the pasted text appear exactly as it is