Re: [sqlite] Corrupt index == disk full?

2017-05-10 Thread Ward Willats
> On May 10, 2017, at 10:49 AM, Simon Slavin wrote: > > On 10 May 2017, at 6:31pm, Ward WIllats wrote: > >> I guess with corruption, all bets are off > > I see your results from "pragma integrity_check". As you write, "all bets > are off".

Re: [sqlite] Corrupt index == disk full?

2017-05-10 Thread Richard Hipp
On 5/10/17, Ward WIllats wrote: > > The unusual parts of our setup are: > > 1. The DB in questions is on a volatile filesystem and ATTACHED to another, > smaller one residing on a flash (JFFS) filesystem, and > 2. We use the shady SQLITE_SHM_DIRECTORY compile switch to

Re: [sqlite] Corrupt index == disk full?

2017-05-10 Thread Ward WIllats
> On May 10, 2017, at 2:20 PM, Simon Slavin wrote: > I’m concerned that this sort of corruption is happening repeatedly. Whether > or not there’s anything in howtocorrupt.html that rings any bells, you can > rely on us to help figure out what’s wrong. > Thanks. That's

Re: [sqlite] Corrupt index == disk full?

2017-05-10 Thread Simon Slavin
On 10 May 2017, at 9:26pm, Ward WIllats wrote: > But, we didn't do this in this case because a "disk full" error is not the > CORRUPT or NOTADB error we expected to see to trigger the panic. So I was > surprised, but... I suspect that your initial post is correct,

Re: [sqlite] Corrupt index == disk full?

2017-05-10 Thread Ward WIllats
> On May 10, 2017, at 10:49 AM, Simon Slavin wrote: > > On 10 May 2017, at 6:31pm, Ward WIllats wrote: > >> I guess with corruption, all bets are off > > I see your results from "pragma integrity_check". As you write, "all bets > are off".

Re: [sqlite] Corrupt index == disk full?

2017-05-10 Thread Simon Slavin
On 10 May 2017, at 6:31pm, Ward WIllats wrote: > I guess with corruption, all bets are off I see your results from "pragma integrity_check". As you write, "all bets are off". If your database is corrupt, any operation on it might silently lead to more corruption.

[sqlite] Corrupt index == disk full?

2017-05-10 Thread Ward WIllats
Hello. We have a DB with a corrupted index (see below). The database has a max_pages limit that works out to a 10 MB database size (we're embedded). We can insert into the table the index is on OK. When we do a SELECT out of the table that uses the index, we get a "disk or database full"