[Tarantool-patches] [PATCH v14 1/6] qsync: track confirmed lsn number on reads

[Serge Petrenko]

10.09.2021 18:29, Cyrill Gorcunov пишет:
> We will use this lsn for requests validation
> in next patches for sake of split-brain detection.
>
> Part-of #6036
>
> Signed-off-by: Cyrill Gorcunov <gorcunov at gmail.com>
> ---
>   src/box/txn_limbo.c | 7 +++++++
>   1 file changed, 7 insertions(+)
>
> diff --git a/src/box/txn_limbo.c b/src/box/txn_limbo.c
> index 70447caaf..cca2ce493 100644
> --- a/src/box/txn_limbo.c
> +++ b/src/box/txn_limbo.c
> @@ -437,6 +437,13 @@ txn_limbo_read_confirm(struct txn_limbo *limbo, int64_t lsn)
>   		assert(e->txn->signature >= 0);
>   		txn_complete_success(e->txn);
>   	}
> +
> +	/*
> +	 * We use confirmed lsn number to verify requests and
> +	 * reject ones coming from split-brain cluster configurations,
> +	 * so update it even if there were no entries to process.
> +	 */
> +	limbo->confirmed_lsn = lsn;
>   }
>   
>   /**

I guess there'll be problems on master with this approach.

Say, a pair of CONFIRM requests is written, with lsns
N and N+1. So you first enter write_confirm(N), then
write_confirm(N+1). Now both fibers issuing the requests yield
waiting for the write to happen, and confirmed_lsn is N+1.

Once the first CONFIRM (N) is written, you reset confirmed_lsn to N
right in read_confirm.

So until the second CONFIRM (N+1) is written, there's a window
when confirmed_lsn is N, but it should be N+1.

I think read_confirm should set confirmed_lsn on replica only.
On master this task is performed by write_confirm.
You may split read_confirm in two parts:
set confirmed lsn (used only on replica) and
apply_confirm (everything read_confirm did before your patch)

-- 
Serge Petrenko