At some point in the future is all I can say, not anytime soon. My
point was more that MVCC on PageStore is a dead end for us, so we're
not going to expend any effort debugging problems with it.
So if you want reliability, use PageStore without MVCC.
--
You received this message because you are
http://h2database.com/html/advanced.html#file_system
You'll need to examine the source to see how the actual classes involved work.
--
You received this message because you are subscribed to the Google Groups "H2
Database" group.
To unsubscribe from this group and stop receiving emails from
Ah, I see. Can I get linked to that part of the API? I see mentions of
pluggable interfaces on the website, but I couldn't find leads to where the
docs for them are, so I didn't investigate further when I came across the
topics.
Thanks.
On Tuesday, February 16, 2016 at 10:20:57 AM UTC-8,
Do you know of any timelines to disable MVCC support in PageStore? I don't
see it listed in the roadmap http://www.h2database.com/html/roadmap.html
but I do see that the removal of at least one PageStore feature is being
called out.
--
You received this message because you are subscribed to
I spent some time trying to recreate the issue with a simple table set up
but was not able to. There must be something else in my database schema
that is contributing to the problem.
Interestingly enough, when I use the org.h2.tools.Shell to view my h2
database, I am able to connect to it
Yeah, we don't have anything like that ATM.
You might be able to implement it by using our pluggable file system API.
--
You received this message because you are subscribed to the Google Groups "H2
Database" group.
To unsubscribe from this group and stop receiving emails from it, send an email
It's likely that I did misunderstand. I want to use off-heap storage when
using MVStore. I want the file specified in the Builder to live on, even
after the application exits. I hope that makes sense.
On Tuesday, August 25, 2015 at 9:03:00 AM UTC-7, Thomas Mueller wrote:
>
> Hi,
>
> Is it
Really sorry. I never got the notification of a reply in my email, and
thought everyone was busy. It was an error in the version I was using
1.4.186. Using 1.4.191, the problem is gone.
While on the topic of "errors" in 1.4.186 that aren't in 1.4.191 but didn't
seem listed on the changelog,
I'm working on creating a standalone test and will update once I have one.
Thanks!
On Tuesday, February 16, 2016 at 12:30:45 AM UTC-6, Noel Grandin wrote:
>
> Of course,it could also be that you've uncovered another tricky corner
> case - in which case we need some kind of
> standalone test
Thanks for the hints, but my main concern was not to recover the data.
It turns out that only a few records out of almost 20 million in the
original database was causing NPE.
Those records could not be selected or deleted without causing NPE.
I therefore copied all other records using a new
Once a DB has been corrupted, there is not much that can be done.
You can run the Recovery tool to generate a SQL dump, but its not at all
guaranteed that it will recover all of the data.
--
You received this message because you are subscribed to the Google Groups "H2
Database" group.
To
Hi Peter,
The wildcard '*' in "java -cp h2***.jar org.h2.tools.Server" should be a
"Filename Expansion" in Bash not a classpath wildcard. A classpath wildcard
should be '*' not "*.jar" or "xx*.jar".
So you can execute "java -cp ./* org.h2.tools.Server" in Windows cmd:
> java -cp ./*
12 matches
Mail list logo
