When launching a backup with the following command I'm
getting:
$ dibs.py auto_check
[...]
Exception of type '<class
dibs_lib.dibs_exceptions.NoPeersToTradeWith at
0x82c1234>': 'DIBSException: No peers exist in
database with sufficient trading room.Traceback:
'.
My peer however says:
{some.peer@some.domain, {
email='some.peer@some.domain',
comment='no comment here',
talk='passive',
listen='active',
host='some.domain',
port=6363,
remoteQuota=50000000 (bytes),
remoteStorage=0 (bytes),
localQuota=500000000 (bytes),
localStorage=52282390L (bytes),
[...]
localQuota is 500MBytes of which 52 have been used.
The total of what I'm trying to backup is:
autoBackup$ for i in `ls`; do readlink $i; done | xargs
du --total -k
[...]
354844 total
(which is 354MBytes)
It needs to be said however, that I had localQuota set
to 50MBytes initialy and raised that to 500MBytes after
the first run out of space. After that I've restarted
everything. To me it looks like though, as if the
previous value had been remembered somewhere?
*t
Logged In: YES
user_id=665772
Backing up 354 MB takes roughly 3*354 MB of space for a
number of reasons. First, DIBS makes redundant copies
depending upon various parameters. Second the encryption
adds some overhead as well. Thus, the amount of space shown
by the show_database command is how much room you have left
on your peer *NOT* how much more data you can backup.
I'll accept the bug but I'm giving it a low priority for now
since it's not really a bug but instead it's just the way
the remaining space is reported.
Of course, I realize that you probably want this to work
differently but since the overhead added by the encryption
is non-deterministic, that is not straightforward at this point.
-Emin
Logged In: YES
user_id=15179
Emin wrote:
> Thus, the amount of space shown by the show_database command
> is how much room you have left on your peer *NOT* how much
> more data you can backup.
Currently the total amount of used diskspace on the peer is:
$ du -s -m /var/lib/dibs/
56 /var/lib/dibs
That is 56MBytes. I've reserved 500MBytes for the peer in
the database though. The biggest file to backup is under a
megabyte:
autoBackup$ find . -follow -printf "%k\n"|sort|tail -1
996
So anyway one calculates it, it won't sum up to 500MB.
I maintain, that there's something wrong. Am I
missunderstanding something?
*t
Logged In: YES
user_id=665772
When you upped the backup space from 50M to 500M, did you do
that on both your end and the other end? You need to do it
on both ends.
Specifically, the output of the show_database command in
your bug report is for the peer. Does the output of
show_database on your local machine show remoteQuota = 500 MB?
-Emin