Re: svnsync crashes on a huge commit

On Wed, Jul 17, 2013 at 6:01 PM, Stefan Fuhrmann
<stefan.fuhrmann_at_wandisco.com> wrote:
>
>
> All that said, it may still be necessary or at least be useful
> to retry opening any FSFS file (in case of network glitches
> etc.). Because there opening files should never fail as part
> of the normal operation, retrying when it does should do no
> extra harm.
>

In this case the glitch (if it was a network glitch) was caused by
svnsync itself. System suffer from heavy swapping when svnsync
allocated more than 2Gb of ram, probably 3Gb or so. 32bit version just
can't mirror this repo in theory.

It also populate transaction folder with enormous number of files.
297652 files in single .txn folder is also a good stress test to file
system, network file system specifically. But it is not the cause of a
problem. Memory leak is a problem.
Received on 2013-07-18 09:26:21 CEST