rdiff-backup-users
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [rdiff-backup-users] OverflowError when backing up remote server


From: Ben Escoto
Subject: Re: [rdiff-backup-users] OverflowError when backing up remote server
Date: Sun, 8 Feb 2004 14:35:31 -0800

>>>>> Ryan Castle <address@hidden>
>>>>> wrote the following on Fri, 06 Feb 2004 10:46:47 +1100

> I don't know much about Python, but it appears that FileObject.read()
> expects an integer, and in your particular the instance the size data is
> large enough that it won't fit into an int. I think that must mean that
> the data length is over 4gb cause that's where python ints seem to end.
> 
> Even if the data should never get to this size, if would be good if the
> code didn't depend on the implicit cast from long to int. 

Well in this case it's not just that rdiff-backup doesn't like large
files or something.  rdiff-backup breaks up large files into chunks as
determined by Globals.conn_bufsize (default is 96Kb).  So unless you
change conn_bufsize to 4Gb (probably a bad idea), it should be pretty
safe.

The buffers are overloading in this case because the other side is
dying, and then random garbage is being sent over, which is causing the
error.


-- 
Ben Escoto

Attachment: pgpOzovEphcZe.pgp
Description: PGP signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]