Re: [HACKERS] Assertion failure when the non-exclusive pg_stop_backup aborted.
| От | Michael Paquier |
|---|---|
| Тема | Re: [HACKERS] Assertion failure when the non-exclusive pg_stop_backup aborted. |
| Дата | |
| Msg-id | CAB7nPqRLxYuu8ZwkRbX14+2DW30qJkBFOMJtMULsDCLtHFM6Kw@mail.gmail.com обсуждение исходный текст |
| Ответ на | Re: [HACKERS] Assertion failure when the non-exclusive pg_stop_backup aborted. (Masahiko Sawada <sawada.mshk@gmail.com>) |
| Ответы |
Re: [HACKERS] Assertion failure when the non-exclusive pg_stop_backup aborted.
|
| Список | pgsql-hackers |
On Thu, Nov 16, 2017 at 10:57 AM, Masahiko Sawada <sawada.mshk@gmail.com> wrote: > Agreed. Attached the updated patch, please review it. + /* + * Quick exit if session is not keeping around a non-exclusive backup + * already started. + */ + if (sessionBackupState != SESSION_BACKUP_NON_EXCLUSIVE) + return; I think that it would be more solid to use SESSION_BACKUP_NONE for the comparison, and complete the assertion after the quick exit as follows as this code path should never be taken for an exclusive backup: + Assert(XLogCtl->Insert.nonExclusiveBackups > 0 && + sessionBackupState == SESSION_BACKUP_NON_EXCLUSIVE); And your patch would discard both SESSION_BACKUP_EXCLUSIVE and SESSION_BACKUP_NONE. -- Michael
В списке pgsql-hackers по дате отправления: