dd strangeness

[abc]

I need to split a large stream of approx 600 GB that gets generated by
Solaris' zfs send command. I could try the whole chain with a big
enough file but I don't see how the zfs command could be the problem
as it only generates a stream.

These two commands work as expected:

# zfs send <backup> | wc gives the expected result, wc displays the
full size of the backup, well beyond 140 GB, as I'm interested in the
first 140 GB I think this is good.
# cat /dev/zero | dd bs=512 count=293601280 | wc gives also expected
results, dd reads exactly 293601280 blocks and wc sees 150323855360
characters, that is, 140 GB

The problem comes when I combine the former part of the first command
with the latter of the second, dd always reads short of the count
value:

# zfs send <backup> | dd bs=512 count=293601280 | wc

293590463+10817 records in
293590463+10817 records out

Admittedly, this is Solaris' dd but when I pipe the stream through nc
to an up-to-date Linux box and do the dd there, I also get short reads
on the last block.

I haven't tried BSD dd yet.

bs=512 is the last attempt I made but I've tried combinations of the
bs and count parameters to no avail, nothing seems to work with the
600 GB stream. I haven't tried bs=1 as I think it would take weeks to
go through but maybe I'm wrong. If I try with smaller files, up to
hundreds of MBs dd works fine, but I can't tell at what size it breaks
or under which circumstances or why. I welcome any pointers at this
time.

Thanks,
Lucia