WPM Player Suddenly Slow on AWS

On Tuesday, 11 March around 8am EDT all of my WPM Transactions running on AWS m1.small instances in Virginia, Oregon and Ireland began to report back wildly high timings.  This happened to be just before or after an upgrade to NPM 10.7 which I would not expect to be related.  I destroyed all of the AWS instances and recreated from scratch.  Same issue.  Today I created an m1.large and the problem does not manifest there.  Even with a single transaction, all of my m1.small instances report very high timings.  This was Microsoft patch Tuesday and I am wondering if that is related?  I'm not sure of the AMI is configured to install updates automatically.

All of my internal players are working fine.  It's just the AWS instances.  Anyone else have this?


  • Hey kahmeelyon, Saw your issue today and without understanding your internal jargon, this does ring similar to an issue I am seeing. Let me speak the issue back to you so you can confirm if I am seeing the issue correctly. You run the same transactions from locations on your internal network (LAN), and from locations on your (WAN). The local transactions run ok, but anything outside of that is running very slow... and this has not changed from the date you reported the problem? Paul (Joseph)

  • So after a month's hiatus working on other things, I decided to revisit this issue.  I spun up small instances in AWS East, AWS West and AWS EU West (Ireland).  None of these players has more than 3 transactions.  The result is just as in the original post.  The transactions run but much more slowly than more an ordinary VM.  Is anyone else using this feature with m1.small?


  • 1. Did you configure the players with the same number of worker processes? 2. Are you using the default user accounts (SEUM-User-1)?