2021.2.3 class MES performance degradation?

Afternoon all,

We use the classic MES via thin clients RDP’ing to a term server. This has worked reasonably well since implementation ~9 months ago. We’re cloud and performance has always been “meh”. Entering an employee ID and hitting tab has generally taken ~7 - 8 seconds before additional actions can be performed. As of yesterday, I’ve been receiving reports of very poor MES performance. The same employee login task now take upwards of 15 seconds. This appears to impact other tasks as well, such as job clock-in and material issuance.

Yesterday I noticed that memory usage on our term server was very high. Hundreds of “Essential Objects Work Process” processes had been launched by each MES instance, eating hundreds of megabytes of memory each. Shutting these down had no effect on the MES client and freed up memory. I thought this might have been the issue, but memory use today does not appear to be an issue and performance is still very poor. CPU utilization and network/Internet performance do not appear to be an issue either.

The regular Epicor client doesn’t seem to be having a similar issue, so this issue appears to be MES specific.

Anyone having a similar experience?

Thanks!

Our MES stations are acting up too… I was thinking maybe someone just hasn’t clocked out of a job in a long time but seeing as we just got an update and you are also having a problem, something else might be afoot.

Like @Evan_Purdy said, an easy thing to check is that your operators are CLOCKING OUT, not logging out, every day.

Not clocking out can cause MES performance degradation over time.

I heard back from Epicor support. This is a reported issue and has been assigned problem PRB0240503. An excerpt from the problem description: “MESMenu.GetMenuRows is taking up to 60 seconds in MT environments”.

1 Like

Good to know. All of our “shop team” clocks out at the end of every day. Any stragglers that forgot are clocked out by shop management in the morning.

Yep, we got “known issue” for our case too

Anyone else having an issue where a previous jobs data is showing up when an employee tries to start production on a new job? Our only fix right now is an employee has to end activity, clock out, log in, start production, every time they start a new job. In some parts of our shop, an employee might have to do this 10 times a day… VERY frustrating. MES stations are also taking FOREVER to do anything…

I haven’t heard of this particular issue, just poor performance.

I’ve noticed some users times are being flagged as “Entered” and not “Submitted” after the update. I cant track down if this is some user error or update related. Is there something I’m overlooking? These users all have a Work Group assigned and nothing has changed from a configuration standpoint.

There is an issue with MES since the latest update. We have been back and forth with support all week trying to get it resolved. Our staff has not been able to report labor through our MES thin clients all week.

Our MES performance appears to be significantly improved. An employee login now only take 3 - 5 seconds; possibly even faster than before.