I see in my logs the following error.
SQLSTATE[HY000]: General error: 1205 Lock wait timeout exceeded; try restarting transaction (SQL: select * from
job_batches
whereid
= xxx limit 1 for update)
I found out that this query is made right before Laravel updates the number of processed jobs in the database. But have no clue why this timeout occurs (The code is inside the transaction, so the lock should be released immediately).
P.S. I use Amazon SQS to drive my queues
P.S.S. I suggested that there are some jobs with not committed nested Laravel transactions. But I tried to reproduce such a scenario, and it looks like this suggestion is wrong.