I did an upgrade from 10.1.600.5 to 10.1.600.20. I ran through all of the conversions successfully, ran my local Client as Administrator and had it update, but as I’m verifying everything by opening various modules, I’m seeing error after error that looks like this (with many different *.svc errors):
Any idea what might be causing this type of error? I’ve bounced Application Pools, Recycled IIS, (Reinstalled) and bounced ICE Task Agent, Rebooted SQL and App, nothing seems to work. Any thoughts?
I found all of my RAM maxed out, as my EpicorTest10 w3wp.exe process had consumed over 10GB of memory, so it wasn’t letting my Live have anything. I didn’t take a “before” screenshot, but I killed the w3wp.exe process and it came back and stabilized around 2GB. Now nothing in my Live environment is throwing errors, and my Customizations are verifying successfully as we speak. crosses fingers
Next morning update: So far, so good. My nightly processes and automatic reports ran, RAM is still at 2 GB for Test. I’ll just leave this here in case someone comes across it in the future.
I’m glad things are working for you. When we moved from 10.1.400.(20 something) to 10.1.600.8 this same thing happened to us intermittently maybe 3-5 times over a month or two. We were not yet live at the time. We went live at the start of the year and have had 2 times now that live has locked up and kicked everyone out (this time without the service errors). We are hosted by Epicor (single tenant) and they added 8GB of ram to the app server and we haven’t had any issues for a few weeks now but are still very uneasy that we don’t know why this was happening.
This doesn’t really help you at all but I wanted to share since this was a recurring thing for us over a few months.
I wanted to comment that to resolve this issue for us I set the WCF throttle or minFreeMemoryPercentageToActivateService in our web.config file equal to “1” or 1% of memory before it would not allow new service calls. I believe the default is 5% and you can also set this to 0 which means no throttle. Here is where I got the information and suggestion to do this:
Since setting this, we have not run into any further “…requested service…blah blah blah…could not be activated…” issues.
It was a bear (<–it wasn’t that cute) trying to figure out exactly where to put this setting so I hope it’s okay to post this to help out.
PLEASE be careful making this change because, someone correct me if I’m wrong, but any changes to the web.config file might auto-trigger some IIS\AppServer resets or something to that effect…I think? I don’t make changes LIVE so I wouldn’t know. (cough) And I would definitely test this in a TEST environment first and make sure everything still runs correctly and then try making the change during a maintenance window and see how it works for you.
I saw that post when I was googling the issue and bookmarked it thinking it may come in handy. I’m glad to see it linked here with some validation, thank you!
We just had this similar problem today and I had to kill one of the w3wp.exe tasks to clear up the error message. I did notice that recycling the IIS App Pool spawned a new w3wp.exe task. Is this the expected behavior to have multiple w3wp.exe tasks or keep re-spawning upon each IIS recycle?
(<–it wasn’t that cute) trying to figure out exactly where to put this setting so I hope it’s okay to post this to help out.
(cough) And I would definitely test this in a TEST environment first and make sure everything still runs correctly and then try making the change during a maintenance window and see how it works for you.
All thanks and credit to 