Subject | Re: Long Running Transactions...again... |
---|---|
Author | beru7 |
Post date | 2005-04-28T13:17:24Z |
--- In firebird-support@yahoogroups.com, "Ann W. Harrison"
<aharrison@i...> wrote:
The test I did was:
T1 ReadCommitted R/W
T2 Snapshot
T1 Open
T2 Open
Loop many many times:
Write something in T2 and commit
Write something in T1 and commitretaining from time to time.
Now, checking GSTAT, I see the oldest snapshot is stuck to where it
was when our transactions started.
Which is not what Craig Stuntz said would happen: that it was OK to
keep a transaction open using ReadCommitted/CommitRetaining, but ONLY
in IB7.1 because of some new functionality.
Am I getting something wrong or do FB and IB7.1 behave differently in
this situation ? Or does the fact that "oldest snapshot" doesn't move
forward not mean anything is going wrong ?
Thanks
Berenger
<aharrison@i...> wrote:
> Pierre Y. wrote:Craig Stuntz
> >
> > Are the advices concerning Long Running Transaction, posted by
> > in his blog, applicable to Firebird ?I've done some tests to verify this and it's not what I saw !
> >
>
> Yes.
The test I did was:
T1 ReadCommitted R/W
T2 Snapshot
T1 Open
T2 Open
Loop many many times:
Write something in T2 and commit
Write something in T1 and commitretaining from time to time.
Now, checking GSTAT, I see the oldest snapshot is stuck to where it
was when our transactions started.
Which is not what Craig Stuntz said would happen: that it was OK to
keep a transaction open using ReadCommitted/CommitRetaining, but ONLY
in IB7.1 because of some new functionality.
Am I getting something wrong or do FB and IB7.1 behave differently in
this situation ? Or does the fact that "oldest snapshot" doesn't move
forward not mean anything is going wrong ?
Thanks
Berenger