On Wed, Sep 17, 2025 at 10:36 AM G.W. Haywood <[email protected]>
wrote:
> Firstly, I'd be inclined to check the numbers in some way other than
> by looking at the graph. :) It wouldn't be the first time that some
> graph or other didn't perfectly reflect the underlying data. It may
> be helpful to know exactly how much data we're talking about in the V3
> pool (as the graphs aren't terribly precise) and also how many files.
> It would be good to eliminate somehow any possibility that BackupPC
> isn't the culprit, since BackupPC's graph just shows you approximately
> how much data is in the pool directories, it doesn't actually know how
> it got there.
>
I can't find clear documentation about how to identify v3 pool files. It
seems to be sort of implied in the docs that they are the single-character
directories in cpool? I'm guessing based on the fact that the section on
v4 attrib files only mentions two-character directories at the top of the
tree.
If that's the case then it's about 15% of the total (1.2G out of 7.7G).
> Secondly, presumably you have
> $Conf{PoolV3Enabled} = 1;
>
I do.
In that case when it's doing a backup, BackupPC will of course use the
> V3 pool and matching data found there. Like you, I wouldn't expect
> *new* V3 pool/cpool data to be written but I haven't done any tests at
> all to see what might happen under oddball circumstances[*]. I see
> setting $Conf{PoolV3Enabled} = 1; in the nature of stop-gap measure,
> and I recommend setting it to 0 ASAP. Of course that supposes that
> all your V3 backups are expired or considered to be of no more use.
>
So I looked... the paths under `cpool/?` (single-character pool
directories) with the most recent mod-times are:
1) directories (e.g. `cpool/f/6/0`)
2) most recently modified 105 days ago
That is definitely within a few days of when I ran BackupPC_migrateV3toV4,
if not that very day. So I think that confirms I'm not getting new v3
backups.
Which really just leaves some kind of anomaly or variation in the data
collection for the RRD or data extraction from the RRD. Again, I'm not
really concerned about this, just curious what's up. If I find myself with
all kinds of free time (HA!) maybe I'll dig into the data tables just to
make sure which side of that line the issue is on.
Thanks for the assist!
Matt
>
_______________________________________________
BackupPC-users mailing list
[email protected]
List: https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki: https://github.com/backuppc/backuppc/wiki
Project: https://backuppc.github.io/backuppc/
