From: | jiye <jiye_sw(at)126(dot)com> |
---|---|
To: | depesz(at)depesz(dot)com |
Cc: | pgsql-hackers(at)lists(dot)postgresql(dot)org |
Subject: | Re:Re: Is there any plan to support online schem change in postgresql? |
Date: | 2022-10-11 12:31:53 |
Message-ID: | 7e9e506b.4168.183c707b0b4.Coremail.jiye_sw@126.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
But, as follow, if txn1 not commit (just like long term readonly txn), it will block txn2's ddl job, why alt add/drop column can not concurrently with read only access?
txn1: long term txn not commit access t1.
txn2 waiting txn1 to commit or abort.
txn3 wait txn2...
At 2022-10-11 18:05:01, "hubert depesz lubaczewski" <depesz(at)depesz(dot)com> wrote:
>On Tue, Oct 11, 2022 at 05:43:03PM +0800, jiye wrote:
>> As we know postgres using high level lock when do alter table or other ddl commands,
>> It will block any dml operation, while it also will block by long term dml operation.
>
>Most of the things can be already done in non-blocking (or almost
>non-blocking way) if you just do it in a way that takes concurrency into
>account.
>
>Specifically - I have no problem adding/deleting columns.
>
>Best regards,
>
>depesz
From | Date | Subject | |
---|---|---|---|
Next Message | Laurenz Albe | 2022-10-11 12:37:25 | Make EXPLAIN generate a generic plan for a parameterized query |
Previous Message | Peter Eisentraut | 2022-10-11 11:51:50 | Re: Allow tests to pass in OpenSSL FIPS mode |