> I do expect the semi-real-time mirroring technology to be in by
> mid-year, because that will be directly tied into the UNDO fifo and the
> UNDO fifo is needed for crash recovery.
Cheers,
Is it possible at this stage to boot DragonFly kernel off a hammer
filesystem ? Maybe it would be a go
:Matthew Dillon wrote:
:> Here's what is left:
:> * Structural locking. The B-Tree is fine-grained locked but the
:>locks for the blockmap are just a hack (one big lock).
:
:Have you decided how to implement multi-master replication yet?
:
:--
:Jason Smethers
My current plan
:So -- in this instance you are both the establishment and the revolutionary at
:the same time. Could you explain in extra-bonehead language what problems you
:were solving with the cluster model, and if you are still solving those
problems
:with the newer model?
:
:Thanks!
The three major
Matthew Dillon wrote:
Here's what is left:
* Structural locking. The B-Tree is fine-grained locked but the
locks for the blockmap are just a hack (one big lock).
Have you decided how to implement multi-master replication yet?
--
Jason Smethers
:
:> Considering how few items are left on the list for HAMMER, it will
:> likely be production-ready long before the mid-year release.
:
:Do you think you'll get some of the clustering functionality in before the mid
:year release, like volumes spanning multiple hosts? What about booti
:Is it worth pushing release back, say, 1 month? Getting off the 1.1x
:series that much sooner would be nice.
No, we should release 1.12 this month as planned.
-Matt
Matthew Dillon
Matthew Dillon wrote:
HAMMER is really shaping up now...
This is far, FAR better then the cluster model I was using last week
and had to throw away. Massively better. Like night and day.
I'm fond of telling political hotheads (which you once were, but no longer ;o)
that,
> Considering how few items are left on the list for HAMMER, it will
> likely be production-ready long before the mid-year release.
Do you think you'll get some of the clustering functionality in before the mid
year release, like volumes spanning multiple hosts? What about booting
Dra
On Sun, February 10, 2008 4:00 pm, Matthew Dillon wrote:
> Ok, it's going to be 1.12. As much as I hate 2.0 being so late,
> there's no point calling it 2.0 without a production-ready HAMMER.
>
> Considering how few items are left on the list for HAMMER, it will
> likely be pro
HAMMER is really shaping up now. Here's what works now:
* All filesystem operations
* All historical operations
* All Pruning features
Here's what is left:
* freemap code (allocate and free big-blocks, which are 8MB blocks).
Currently a hack so everything else
Ok, it's going to be 1.12. As much as I hate 2.0 being so late,
there's no point calling it 2.0 without a production-ready HAMMER.
Considering how few items are left on the list for HAMMER, it will
likely be production-ready long before the mid-year release.
Garance A Drosihn wrote:
I'd go with 1.2 for now.
Too bad we can't reuse version numbers ;).
--
Thomas E. Spanjaard
[EMAIL PROTECTED]
signature.asc
Description: OpenPGP digital signature
Petr Janda wrote:
Make this a 1.12, because HAMMER is only in an alpha state. Then when HAMMER
is production ready release, change the numbering scheme altogether. The
roman numerals for example are nice, and would attract attention. DragonFly
XIII sure looks good, and the june release DragonFl
13 matches
Mail list logo
