This WILL create a potential spanning-tree loop - which may and may not create problems, depending on the capabilities and configuration of the switches involved. If they all are configured to handle spanning-tree properly (default for a Cisco workgroup switch often is), then the switches will realize this, and effectively block traffic on one of the links; but keeping it "on standby" for a situation where one of the OTHER links goes down. If the switches don't understand spanning-tree or it isn't configured, then wild and unwanted things will happen - connections will randomly drop off, and traffic levels may even reach capacity on the uplinks.

This WILL create a potential spanning-tree loop - which may and may not create problems, depending on the capabilities and configuration of the switches involved. If they all are configured to handle spanning-tree properly (default for a Cisco workgroup switch often is), then the switches will realize this, and effectively block traffic on one of the links; but keeping it "on standby" for a situation where one of the OTHER links goes down. If the switches don't understand spanning-tree or it isn't configured, then wild and unwanted things will happen - connections will randomly drop off, and traffic levels may even reach capacity on the uplinks.