Re: [Firebird-devel] discussion about real fix for CORE-2348

2011-04-29 Thread Vlad Khorsun
I see that there this problem was analized completly More-or-less Is this really plan to fix this in FB3.0 None, AFAIK Vlad Korshun say that only transaction id will be unsigned int in FB3.0 - is this true or is in plan to fix this with two record header versions or other

[Firebird-devel] discussion about real fix for CORE-2348

2011-04-27 Thread liviuslivius
Hi discussion about http://tracker.firebirdsql.org/browse/CORE-2348 I have solution for this problem reuse transaction id feature.  When oldest active transaction id reach e.g. 1 500 000 001 value (shuld be big as possible but have also big difference to max integer value).  then:  1. set reset

Re: [Firebird-devel] discussion about real fix for CORE-2348

2011-04-27 Thread Dimitry Sibiryakov
27.04.2011 15:33, Kjell Rilbe wrote: Step 2 may be a big problem for 24/7 servers 24/7 systems should not be based on a single server, so, while one node is performing classic backup-restore or suggested procedure, all users can be served by other node(s). -- SY, SD.

Re: [Firebird-devel] discussion about real fix for CORE-2348

2011-04-27 Thread Dimitry Sibiryakov
27.04.2011 21:01, Mercea Paul wrote: Is Firebird server cluster ready? I mean for this scenario with multiple server without replication but cluster ! It sounds like you refuse to call shared-nothing architecture a cluster... -- SY, SD.

Re: [Firebird-devel] discussion about real fix for CORE-2348

2011-04-27 Thread Larry Baddock
discussion about http://tracker.firebirdsql.org/browse/CORE-2348 I have solution for this problem reuse transaction id feature. I guess you are referring to something like this solution, discussed in 2005: http://firebird.1100200.n4.nabble.com/High-Volume-Solution-Needed-td1107716.html Kind

Re: [Firebird-devel] discussion about real fix for CORE-2348

2011-04-27 Thread liviuslivius
Larry Baddock napisał(a): discussion about http://tracker.firebirdsql.org/browse/CORE-2348 I have solution for this problem reuse transaction id feature. I guess you are referring to something like this solution, discussed in 2005: