Re: Bug (#3484) - Invalid page header again
| От | Zdenek Kotala |
|---|---|
| Тема | Re: Bug (#3484) - Invalid page header again |
| Дата | |
| Msg-id | 4768F648.2010009@sun.com обсуждение исходный текст |
| Ответ на | Re: Bug (#3484) - Invalid page header again (Gregory Stark <stark@enterprisedb.com>) |
| Список | pgsql-bugs |
Gregory Stark wrote:
> "Zdenek Kotala" <Zdenek.Kotala@Sun.COM> writes:
>
>> I got dump of affected two blocks from Alex and it seems that both blocks were
>> overwritten together with some 128bytes length structure (there some pattern)
>> and complete damaged size is 9728bytes (first block is overwritten completely
>> and second one only at the beginning), but another buffer from another relation
>> could be overwritten too.
>
> I don't understand this 9728 bytes. Postgres blocks are 8192 bytes. Are you
> saying one whole block is trashed and then part, but not all, of the next
> block?
>
> What's the block size of the ZFS filesystem? And what exactly does the trash
> data look like?
>
ZFS has dynamic block size (max 128kB), but this system runs on Linux probably
ext3 FS. Please, ale can you confirm used filesystem?
Zdenek
В списке pgsql-bugs по дате отправления: