apoc9009 wrote:
>
>> You've either not read 23.4.4 or haven't understood it. If the text is
>> unclear, documentation additions/changes are always welcome.
> I have read this:
>
> PostgreSQL directly supports file-based log shipping as described above.
> It is also possible to implement record-based log shipping, though this
> requires custom development.
>
> But thadt is not thadt what iam looking for!
>
> Filebased Logship backups having a big Problem for doing continous
> Backups. You have to wait until
> the Postmaster has written the WAL File, after this you can save it to
> the Backupserver. But 1 WAL
> has a size of 16 MByte ny default! (thadt is a big Datahole in your
> "Online Backup"!)
Which is why that entire section is about copying just the changed parts
of WAL files.
> It makes no sense to reduce the Filesize. If the Filesize is smaller
> then 16 MBytes for WAL Files
> you have still the same Problem, there are Data losses and thadt the
> Center of the Problem.
But in your original email you said:> All Users of Hugh Databases (Missioncritical and allways Online) can> bring up
its>Databases with the same information with differences 1-5 Sec. before> the Crash occurs!
That suggested to me that you didn't want per-transaction backup, just
one backup every second. OK, what you actually want is a continuous
backup with one copy made per transaction.
> I wish to have an Solution, thadt backup my Database DB wihout
> Datalosses, without locking Tables, without Shutdown
> and without any User must be forced for logging out (Backup in
> Production State Online without Datalosses).
So, if I understand, you want on of:
1. External RAID array. If main machine dies, turn backup machine on.
Both share the same disks.
2. Something like DRBD to copy individual disk blocks between machines.
You could do this just for WAL.
3. Replication.
-- Richard Huxton Archonet Ltd