Replication jobs failing with error 1 (4047879)

Replication jobs that have been working start to fail with error message: "Error 1"

Cause 1: One of the ESX(i) hosts involved in replications, be in the source or destination core, is disconnected either from the virtual center and/or has network or other connectivity issues.
Cause 2: The system disk on a Virtual Appliance has run out of free space. 

Resolution 1: Login into Virtual Center and make sure both source and destination ESX(i) hosts are showing in a connected state. If not, check for issues with the given hosts, get them back online, and/or connected then to the virtual center. Then, retry replication.

Resolution 2: 
1. Putty into the target VA or open the console in vCenter.
2. Verify that no replications that use this VA are currently running. If replications are currently running you must cancel them or wait for them to finish.
3. Enter the following command at the prompt to check if the /scratch/sdb mount is full:

df -h

If the usage is high, enter the following command to clean it up: 

find /scratch/sdb -name *.vzundo -exec rm -rf {} \;

4. Once all the temporary .vzundo files have been cleaned out, run the df command. There should be free space in the /scratch/sdb/ directory once again.
5. Please start running the replication jobs again.