SQLite

Check-in [a8f9bd1e89]
Login

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

Overview
Comment:Call fsync() right after ftruncate() when in journal_mode=TRUNCATE and when synchronous=FULL in order to ensure that transactions are durable across a power loss that happens moments after the commit. Proposed fix for https://bugzilla.mozilla.org/show_bug.cgi?id=1072773. This is a cherrypick of [3e922208b68563489].
Downloads: Tarball | ZIP archive
Timelines: family | ancestors | descendants | both | branch-3.8.7
Files: files | file ages | folders
SHA1: a8f9bd1e890434fcffa02fcd2baf8b0fb5d4c9dd
User & Date: drh 2014-10-29 01:07:21.172
Context
2014-10-29
01:13
In the OP_Column opcode, when extracting a field that is past the end of a short record (because the row was originally inserted prior to ALTER TABLE ADD COLUMN) then make sure the output register is fully NULL and does not contain leftover flags (such as MEM_Ephem) from its previous use. Fix for ticket [43107840f1c02]. This is a cherrypick of check-in [24780f8ddc1683fc]. (check-in: 304ea6ba6f user: drh tags: branch-3.8.7)
01:07
Call fsync() right after ftruncate() when in journal_mode=TRUNCATE and when synchronous=FULL in order to ensure that transactions are durable across a power loss that happens moments after the commit. Proposed fix for https://bugzilla.mozilla.org/show_bug.cgi?id=1072773. This is a cherrypick of [3e922208b68563489]. (check-in: a8f9bd1e89 user: drh tags: branch-3.8.7)
2014-10-21
21:56
Call fsync() right after ftruncate() when in journal_mode=TRUNCATE and when synchronous=FULL in order to ensure that transactions are durable across a power loss that happens moments after the commit. Proposed fix for https://bugzilla.mozilla.org/show_bug.cgi?id=1072773. (check-in: 3e922208b6 user: drh tags: trunk)
2014-10-17
11:24
Version 3.8.7 (check-in: e4ab094f8a user: drh tags: trunk, release, version-3.8.7)
Changes
Unified Diff Ignore Whitespace Patch
Changes to src/pager.c.
︙︙
1937
1938
1939
1940
1941
1942
1943








1944
1945
1946
1947
1948
1949
1950

      assert( pPager->journalMode==PAGER_JOURNALMODE_MEMORY );
      sqlite3OsClose(pPager->jfd);
    }else if( pPager->journalMode==PAGER_JOURNALMODE_TRUNCATE ){
      if( pPager->journalOff==0 ){
        rc = SQLITE_OK;
      }else{
        rc = sqlite3OsTruncate(pPager->jfd, 0);








      }
      pPager->journalOff = 0;
    }else if( pPager->journalMode==PAGER_JOURNALMODE_PERSIST
      || (pPager->exclusiveMode && pPager->journalMode!=PAGER_JOURNALMODE_WAL)
    ){
      rc = zeroJournalHdr(pPager, hasMaster);
      pPager->journalOff = 0;








>
>
>
>
>
>
>
>








1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958

      assert( pPager->journalMode==PAGER_JOURNALMODE_MEMORY );
      sqlite3OsClose(pPager->jfd);
    }else if( pPager->journalMode==PAGER_JOURNALMODE_TRUNCATE ){
      if( pPager->journalOff==0 ){
        rc = SQLITE_OK;
      }else{
        rc = sqlite3OsTruncate(pPager->jfd, 0);
        if( rc==SQLITE_OK && pPager->fullSync ){
          /* Make sure the new file size is written into the inode right away.
          ** Otherwise the journal might resurrect following a power loss and
          ** cause the last transaction to roll back.  See
          ** https://bugzilla.mozilla.org/show_bug.cgi?id=1072773
          */
          rc = sqlite3OsSync(pPager->jfd, pPager->syncFlags);
        }
      }
      pPager->journalOff = 0;
    }else if( pPager->journalMode==PAGER_JOURNALMODE_PERSIST
      || (pPager->exclusiveMode && pPager->journalMode!=PAGER_JOURNALMODE_WAL)
    ){
      rc = zeroJournalHdr(pPager, hasMaster);
      pPager->journalOff = 0;
︙︙