We've been using s3sync for months now - it's a great script. Thanks for the hard work.
I have seen this error occur a few times recently as well. I am using v1.2.4, executing the following command:
s3sync.rb -v --ssl -r /backups/daily server-backups:
A sample error message:
/usr/lib/ruby/1.8/net/http.rb:586:in `connect': Connection reset by peer (Errno::ECONNRESET)
from /usr/lib/ruby/1.8/net/http.rb:586:in `connect'
from /usr/lib/ruby/1.8/net/http.rb:553:in `do_start'
from /usr/lib/ruby/1.8/net/http.rb:548:in `start'
from /home/bin/s3sync/S3_s3sync_mod.rb:55:in `make_http'
from /home/bin/s3sync/s3try.rb:62:in `S3tryConnect'
from /home/bin/s3sync/s3try.rb:69:in `S3try'
from /home/bin/s3sync/s3sync.rb:284:in `s3TreeRecurse'
from /home/bin/s3sync/s3sync.rb:345:in `main'
from /home/bin/s3sync/thread_generator.rb:79:in `call'
from /home/bin/s3sync/thread_generator.rb:79:in `initialize'
from /home/bin/s3sync/thread_generator.rb:76:in `new'
from /home/bin/s3sync/thread_generator.rb:76:in `initialize'
from /home/bin/s3sync/s3sync.rb:266:in `new'
from /home/bin/s3sync/s3sync.rb:266:in `main'
from /home/bin/s3sync/s3sync.rb:724
Sync Complete.
I've just upgraded to v1.2.5 but as far as I can tell this new version includes the no md5 option with no other fixes.
I found this thread:
http://developer.amazonwebservices.com/connect/thread.jspa?threadID=11975#45396, where greg says:
The first one seems self-explanatory. I'm not catching connection resets.
Greg, would it be possible to implement a fix so if this error occurs, the sync script attempts to retry (as it does when it encounters 500 Internal Server errors)?
Thanks.