Merge lp:~laurynas-biveinis/percona-server/bug1262500-5.1 into lp:percona-server/5.1
Proposed by
Laurynas Biveinis
Status: | Merged |
---|---|
Approved by: | Alexey Kopytov |
Approved revision: | no longer in the source branch. |
Merged at revision: | 604 |
Proposed branch: | lp:~laurynas-biveinis/percona-server/bug1262500-5.1 |
Merge into: | lp:percona-server/5.1 |
Diff against target: |
129 lines (+37/-5) (has conflicts) 5 files modified
Makefile (+5/-0) Percona-Server/storage/innodb_plugin/os/os0file.c (+7/-5) doc/source/conf.py (+13/-0) doc/source/faq.rst (+8/-0) doc/source/release-notes/release-notes_index.rst (+4/-0) Text conflict in Makefile Text conflict in doc/source/conf.py Text conflict in doc/source/faq.rst Text conflict in doc/source/release-notes/release-notes_index.rst |
To merge this branch: | bzr merge lp:~laurynas-biveinis/percona-server/bug1262500-5.1 |
Related bugs: |
Reviewer | Review Type | Date Requested | Status |
---|---|---|---|
Alexey Kopytov (community) | Approve | ||
Registry Administrators | Pending | ||
Review via email: mp+199699@code.launchpad.net |
Description of the change
Fix bug 1262500 (ibd files contain bogus trailing zeros and non-page
size file sizes): in the case pwrite() was interrupted, the retried
write would not advance the buffer pointer, resulting in data
corruption. Also, the case of a partial write writing 0 bytes was not
handled.
Fix by applying a patch by Andrew Gaul, and extending to handle
pread() as well, except that a read would not be restarted if pread()
returned 0 bytes read, as that indicates an EOF instead of a partial
read.
http://
To post a comment you must log in.
Asking for either G2 or KAlexey's as the original code author's review