On Wed, Apr 13, 2011 at 11:36 AM, James Green <james.mk.gr...@gmail.com>wrote:
> On 13 April 2011 15:33, Simon Slavin <slav...@bigfraud.org> wrote: > > > > On 13 Apr 2011, at 12:14pm, James Green wrote: > > > >> sync=full does not work well for our app (no transactions). Far too > slow. > > > > If you're not syncing, then section 3.2 of the page Richard probably > indicates what's causing your corruption. > > Without a reset or power loss? This is the bit that's causing concern. > > > So for a while, leave that PRAGMA alone, accept that you're going to get > slow running, and see whether this makes the corruption go away. If it does > then you can tackle the speed problem separately without worrying about data > corruption. > > This is the approach we will be rolling out. More specifically we'll > be trying synchronous=normal. It will likely take a few days before we > can roll this out and evaluate. > synchronous=NORMAL in combination with journal_mode=WAL does syncing only on a checkpoint, not on a commit. That should give you good performance and safety from corruption due to power failure or reset. > > > It might be worth trying to figure out whether your storage system can > handle the speed you're expecting. If you have multiple threads all trying > to write at the same time can your hard disk handle that much data at that > speed ? If not, then that's probably why you're finding the use of 'sync' > to be too slow. > > We can say that SQL Server survives just fine. We are not hitting the > disk but hundreds of megabytes of changes. One of our operations has > the following times: > > sync=FULL: 3m 43s > sync=OFF: 10.5s > > We're talking approximately 2,000 rows consisting of 3x 32-char string > columns and 2x booleans. This is without WAL mode and without > transactions. > > Worth noting is that we only have one test environment in which we can > get corruption to occur and we cannot get it to happen at will. All > other test systems operate fine without corruption with sync=OFF. > > James > _______________________________________________ > sqlite-users mailing list > sqlite-users@sqlite.org > http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users > -- D. Richard Hipp d...@sqlite.org _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
