Re: 2PC-induced lockup
| От | Tom Lane |
|---|---|
| Тема | Re: 2PC-induced lockup |
| Дата | |
| Msg-id | 916.1184183622@sss.pgh.pa.us обсуждение исходный текст |
| Ответ на | Re: 2PC-induced lockup (Heikki Linnakangas <heikki@enterprisedb.com>) |
| Список | pgsql-hackers |
Heikki Linnakangas <heikki@enterprisedb.com> writes:
> Peter Eisentraut wrote:
>> Heikki Linnakangas wrote:
>>> It's not? I agree with Tom here; this is just one of the numerous
>>> things you can do to screw up your database as a superuser. Why would
>>> you LOCK the pg_auth table, or any other system table for that
>>> matter, in the first place? Let alone in a distributed transaction.
>>
>> Well, my test case arose from a real application scenario, not an
>> attempt to destroy my database system.
> Why does the application LOCK pg_auth?
Even if there is a reason for a lock, surely it's not necessary to use
AccessExclusiveLock. A lesser lock would synchronize whatever the heck
it's doing without locking out readers.
regards, tom lane
В списке pgsql-hackers по дате отправления: