Title:
Accessing your InstantDb from 2 JVM's at the same time causes this
problem. E.g. looking at the DB under the example browser provided and also from
an application. I make this mistake occasionally.
Setting recoveryPolicy=2 in the .prp
file can be used to stop multiple JVM's accessing
the DB. A recoveryPolicy=1 allows for automatic recovery and as a side effect
seems to allow more than one JVM. This seems a bit odd really.
If you have recoveryPolicy=2 and then you try and
access from 2 JVM's I believe you will get the message you are seeing below.
Basically the DB thinks/knows it is being accessed by another JVM and assumes
that if another JVM tries to attach that the DB must be in an inconsistent state
from a previous execution.
If the DB has not closed down correctly on a previous
occasion I would expect to see a similar problem. To avoid the message set
recoveryPolicy=1, the side effect of this is that multiple JVM's will really be
allowed to open the DB increasing the risk of real
corruption.
Hope this is helpful. Peter Hearty may tell me that
some of this is incorrect. But it what I believe is happening, without having
seen the code.
John.
Hi all,
Can anyone help me on this one. I got the following
message during statup of the application:
Enhydra InstantDB - Version 3.14 The Initial Developer of the Original Code is Lutris
Technologies Inc. Portions created by
Lutris are Copyright (C) 1997-2000 Lutris Technologies, Inc. All Rights Reserved. WARNING!!! - THE DATABASE JOURNAL FILE IS CORRUPT. SOME DATA MAY BE IN AN INCONSISTENT STATE.
My questions are:
- What could be the cause of a corrupt database
journal file? - What is the effect of this
corrupt file? - Is this a fatal fault? Is
the database lost? - How to prevent this
(besides recoveryPolicy)?
Thanks in advance. Jaco
|