Re: [Duplicity-talk] S3 connection timeout

[Peter Schuller]

>     raise S3ResponseError(response.status, response.reason, body)
> boto.exception.S3ResponseError: S3ResponseError: 400 Bad Request
> <?xml version="1.0" encoding="UTF-8"?>
> <Error><Code>RequestTimeout</Code><Message>Your socket connection to the
> server was not read from or written to within the timeout period. Idle
> connections will be
> closed.</Message><RequestId>xxx</RequestId><HostId>xxx</HostId></Error>

I am seeing this a lot recently, along with 'caught a socket error,
trying to recover" from s3/key.py in boto.

In fact right now I am completely and utterly unable to complete a
backup properly, and had the same problem yesterday. Even a single mb
chunk won't upload.

tracing the process indicates that the remote end is closing the
connection, but I don't know if any information is sent back.

tcpdumping traffic shows some a great deal of duplicate ACK:s (several
at a time), and traffic from the s3 server being very very bursty (a
bunch of packets comes through for a few hundre milliseconds, then a
pause up to perhaps a second, then some more, etc). The end result is
slow uploads (disregarding that the upload fails anyway).

This definitely seems to be an off-and-on thing. I have been backup up
to s3 for a while without difficulties in the past (well, sometimes
errors have happened and required retry, but as a rule it has been
stable).

Can anyone report similar problems? Or the reverse, is anyone
successfully using the s3 backend over time without difficulties?

-- 
/ Peter Schuller

PGP userID: 0xE9758B7D or 'Peter Schuller <address@hidden>'
Key retrieval: Send an E-Mail to address@hidden
E-Mail: address@hidden Web: http://www.scode.org

pgpPV17gg5Ca1.pgp
Description: PGP signature