Re: Latest behaviour for write-behind cache on node crash

[Ilya Kasnacheev <il...@gmail.com>]

Hello!

There is no failover for write-behind, it is lost if node is stopped
abruptly.

Regards,
-- 
Ilya Kasnacheev


пт, 31 янв. 2020 г. в 07:03, crypto_ricklee <ri...@crypto.com>:

> Dear all,
>
> I'm trying to get Ignite as a read-through/write-behind cache on top of
> postgres. If I have 2 nodes setup (replication) and the primary node down
> before writing to postgres, will the backup node take over the DB
> write/update for queued data OR the data will lose forever?
>
> Thanks & Regards,
> Rick
>
>
>
> --
> Sent from: http://apache-ignite-users.70518.x6.nabble.com/
>

Re: Latest behaviour for write-behind cache on node crash

["kimec.ethome.sk" <ki...@ethome.sk>]

There is a ticket opened for this for some 4+ years

https://issues.apache.org/jira/browse/IGNITE-1897

Since Ignite got its own native persistence implementation some years 
after the ticket was opened, chances that 3rd party persistence write 
behind failover gets implemented any time soon seem pretty low to me.

---
S pozdravom,

Kamil Mišúth

On 2020-02-03 14:19, Ilya Kasnacheev wrote:
> Hello!
> 
> There is no failover for write-behind, it is lost if node is stopped
> abruptly.
> 
> Regards,
> 
> --
> 
> Ilya Kasnacheev
> 
> пт, 31 янв. 2020 г. в 07:03, crypto_ricklee
> <ri...@crypto.com>:
> 
>> Dear all,
>> 
>> I'm trying to get Ignite as a read-through/write-behind cache on top
>> of
>> postgres. If I have 2 nodes setup (replication) and the primary node
>> down
>> before writing to postgres, will the backup node take over the DB
>> write/update for queued data OR the data will lose forever?
>> 
>> Thanks & Regards,
>> Rick
>> 
>> --
>> Sent from: http://apache-ignite-users.70518.x6.nabble.com/