Just some more details to clarify the issue:
If I do a `BEGIN IMMEDIATE` on one connection, and then a `PRGAMA
journal_mode` on another, BUSY is not returned (as expected).
But if I have around 7 connections contending for a write lock via `BEGIN
IMMEDIATE`, and a different connection runs `PRGAMA journal_mode` (with no
`BEGIN`), it returns BUSY (not expected).
I also occasionally get a "SQLITE_BUSY_RECOVERY", but all API usages are
using finalize/close to give back any sqlite API resources.
*Questions:*
A. In which cases will BUSY be returned for read only queries when in WAL
mode?
- I assumed this was never, as WAL mode allows many concurrent readers.
- How can I find out exceptions to this rule?
B. Is there any method for determining lock transitions for connections?
- Is there an API?
- Would it be possible to use dtrace to instrument SQLite to detect
lock transitions?
- Where should I be looking?
Id really appreciate any pointers,
Thanks.
On Wed, Aug 28, 2019 at 9:46 PM test user <example.com.use...@gmail.com>
wrote:
> Hello,
>
> Im getting this message in the log:
>
> `SQLITE_BUSY, database is locked in "PRAGMA journal_mode"`
>
> I get this response when running the query `PRAGMA journal_mode`.
>
> The file is in journal_mode=WAL.
>
> Another connection holds a write transaction.
>
> Seeing as `PRAGMA journal_mode` is just a read, why would this return BUSY?
>
> I assumed that in WAL mode you can have many reads and a single writer at
> the same time?
>
> Thanks
>
_______________________________________________
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
