On 08/09/11 11:07, Kjell Rilbe wrote:
> Den 2011-08-09 08:54 skrev Alex Peshkoff såhär:
>> On 08/08/11 19:10, Dimitry Sibiryakov wrote:
>>> 08.08.2011 16:57, Kjell Rilbe wrote:
So, I ask you in the devel team: is the shadow file feature going to go
away or will you make sure it stays f
Den 2011-08-09 08:54 skrev Alex Peshkoff såhär:
> On 08/08/11 19:10, Dimitry Sibiryakov wrote:
>> 08.08.2011 16:57, Kjell Rilbe wrote:
>>> So, I ask you in the devel team: is the shadow file feature going to go
>>> away or will you make sure it stays functional?
>> If discussion about droppin
On 08/08/11 19:10, Dimitry Sibiryakov wrote:
> 08.08.2011 16:57, Kjell Rilbe wrote:
>> So, I ask you in the devel team: is the shadow file feature going to go
>> away or will you make sure it stays functional?
>If discussion about dropping of shadow is ever raised I will vote against.
>
+1
-
08.08.2011 22:14, Kjell Rilbe wrote:
> Any indication of how large the lag (N) could be? Can it grow without
> bounds if the server load is high?
I can say for sure only for replicator by IBPhoenix:
It depends on replication settings and timing. If crash occur after
replication finish
but
Den 2011-08-08 18:39 skrev Dimitry Sibiryakov såhär:
> 08.08.2011 18:33, Kjell Rilbe wrote:
>> This copy should be as up to date as possible,
>> transaction by transaction.
>>
>> Are there no replicators that can do this?
>
> Yep.
> All Firebird replicators that I know are asynchronous. It
08.08.2011 18:33, Kjell Rilbe wrote:
> This copy should be as up to date as possible,
> transaction by transaction.
>
> Are there no replicators that can do this?
Yep.
All Firebird replicators that I know are asynchronous. It means that there
is a
possibility to loose some data on crash. T
Den 2011-08-08 17:10 skrev Dimitry Sibiryakov såhär:
> 08.08.2011 16:57, Kjell Rilbe wrote:
>> So, I ask you in the devel team: is the shadow file feature going to go
>> away or will you make sure it stays functional?
>
> If discussion about dropping of shadow is ever raised I will vote against
08.08.2011 16:57, Kjell Rilbe wrote:
> So, I ask you in the devel team: is the shadow file feature going to go
> away or will you make sure it stays functional?
If discussion about dropping of shadow is ever raised I will vote against.
> Is replication a viable option? Perhaps even better? Why
Hi,
In a fb-support discussion it has occurred to me and Karol Bieniaszewski
that the old shadow file feature is actually useful for some purposes.
Karol runs the master on a RAM disk to make sure reads and searches are
as fast as possible, and has a shadow file on a regular disk to protect
ag
