Re: [sqlite] transaction commit is successful despite I/O error

2012-10-25 Thread Pavel Ivanov
I guess this can stand as one more argument for not using SQLite over NFS. Another argument could be found here: http://www.sqlite.org/lockingv3.html#how_to_corrupt. Pavel On Wed, Oct 24, 2012 at 2:01 PM, Efim Dyadkin wrote: > I was probably not quite specific. So I

Re: [sqlite] transaction commit is successful despite I/O error

2012-10-25 Thread Efim Dyadkin
I was probably not quite specific. So I would like to rephrase the problem and give more details. I run a default configuration of Sqlite on Linux and I don't use WAL or anything that changes transaction behavior. A database is located on a remote file server and accessed via NFS. There is a

Re: [sqlite] transaction commit is successful despite I/O error

2012-10-20 Thread Ryan Johnson
On 19/10/2012 4:40 PM, Efim Dyadkin wrote: Hi Richard, You are right about the purpose of unlink but it is out of context. There are a transaction in progress and hot journal on disk. If journal can't be deleted by the end of transaction, the transaction can't be considered to be

Re: [sqlite] transaction commit is successful despite I/O error

2012-10-20 Thread Dan Kennedy
On 10/20/2012 09:14 PM, Simon Slavin wrote: On 19 Oct 2012, at 9:40pm, Efim Dyadkin wrote: You are right about the purpose of unlink but it is out of context. There are a transaction in progress and hot journal on disk. If journal can't be deleted by the end of

Re: [sqlite] transaction commit is successful despite I/O error

2012-10-20 Thread Simon Slavin
On 19 Oct 2012, at 9:40pm, Efim Dyadkin wrote: > You are right about the purpose of unlink but it is out of context. There are > a transaction in progress and hot journal on disk. If journal can't be > deleted by the end of transaction, the transaction can't be

Re: [sqlite] transaction commit is successful despite I/O error

2012-10-20 Thread Efim Dyadkin
Hi Richard, You are right about the purpose of unlink but it is out of context. There are a transaction in progress and hot journal on disk. If journal can't be deleted by the end of transaction, the transaction can't be considered to be successfully finished. But current implementation of

Re: [sqlite] transaction commit is successful despite I/O error

2012-10-18 Thread Ryan Johnson
On 18/10/2012 8:45 AM, Richard Hipp wrote: On Wed, Oct 17, 2012 at 4:11 PM, Efim Dyadkin wrote: Hi, I am testing loss of data in Sqlite database correlated to auto-mounter malfunction. I am running Sqlite on Linux and my database file is located on network disk. For a

Re: [sqlite] transaction commit is successful despite I/O error

2012-10-18 Thread Richard Hipp
On Wed, Oct 17, 2012 at 4:11 PM, Efim Dyadkin wrote: > Hi, > > I am testing loss of data in Sqlite database correlated to auto-mounter > malfunction. I am running Sqlite on Linux and my database file is located > on network disk. For a test I stop the auto-mounter right

[sqlite] transaction commit is successful despite I/O error

2012-10-18 Thread Efim Dyadkin
Hi, I am testing loss of data in Sqlite database correlated to auto-mounter malfunction. I am running Sqlite on Linux and my database file is located on network disk. For a test I stop the auto-mounter right before transaction is committed. Surprisingly commit succeeds without any error