This will make it easier to change RIBucketWriter in the future to reduce the wire
protocol to just open/write(offset,data)/close, and do all the structuring on the
client end. The ultimate goal is to store each bucket in a single file, to reduce
the considerable filesystem-quantization/inode overhead on the storage servers.
Also include the encoder portion of Bob Ippolito's simplejson-1.7.1 as
allmydata.util.json_encoder . simplejson is distributed under a more liberal
license than Tahoe (looks to be modified BSD), so redistributing it should be ok.
To use this, write a number like 10MB or 5Gb or 5000000000 to a file
named 'sizelimit' in the client's base directory. The node will not grant
leases for shares that would take it much beyond this many bytes of
storage. Note that metadata is not included in the allocation count until
a restart, so the actual space consumed may grow beyond the limit if
the node is not restarted very frequently and the amount of metadata is
significant.
If the error occurs before any data has been sent, we can give a sensible
error message (code 500, stack trace, etc). This will cover most of the error
cases. The ones that aren't covered are when we run out of good peers after
successfully decoding the first segment, either because they go away or
because their shares are corrupt.
Previously, exceptions during a web download caused a hang rather than some
kind of exception or error message. This patch improves the situation by
terminating the HTTP download rather than letting it hang forever. The
behavior still isn't ideal, however, because the error can occur too late to
abort the HTTP request cleanly (i.e. with an error code). In fact, the
Content-Type header and response code have already been set by the time any
download errors have been detected, so the browser is committed to displaying
an image or whatever (thus any error message we put into the stream is
unlikely to be displayed in a meaningful way).
