Using --set-vars of course is much better and I agree with you, there is no need to introduce yet another variable now that I see your point.
I have one another request regarding the variable --retries. I see that currently the documentation says that it does the following:
"Retry a chunk this many times when there is a nonfatal error. Nonfatal errors are problems such as a lock wait timeout or the query being killed. This option applies to the data copy operation."
However, in the context of this bug report, I believe that the end user might want to retry metadata locking part many times as well, for example suppose a rename fails because of metadata locking and low value of lock_wait_timeout, then the user might want to have that retried x number of times. So I was wondering if we could have this retry behavior for more parts of the online-schema-change?
Hi Daniel,
Using --set-vars of course is much better and I agree with you, there is no need to introduce yet another variable now that I see your point.
I have one another request regarding the variable --retries. I see that currently the documentation says that it does the following:
"Retry a chunk this many times when there is a nonfatal error. Nonfatal errors are problems such as a lock wait timeout or the query being killed. This option applies to the data copy operation."
However, in the context of this bug report, I believe that the end user might want to retry metadata locking part many times as well, for example suppose a rename fails because of metadata locking and low value of lock_wait_timeout, then the user might want to have that retried x number of times. So I was wondering if we could have this retry behavior for more parts of the online- schema- change?