| Commit message (Collapse) | Author | Files | Lines |
|---|
|
|
|
When running for the first time, people see migration failures because of sql statements that are trying to move data from old tables (e.g. users). The amended text attempts to calm their nerves.
|
|
|
|
to the log
|
|
This is for mantis 4783
|
|
Throwing an Ex. with SQL command in the message looks weird,
this is a bit better, but I'm still not sure if that's the
proper way to handle. Also, there is a catch one level up,
so is this one necessary?
|
|
Change migration error reporting to not truncate the statement when
reporting. It's a bit messier than the old error reporting, but at least
one gets an idea of what could be wrong again. And things look a lot
neater now.
|
|
|
|
prevented the full message from displaying.
|
|
migrations to continue in the face of an error
|
|
|
|
Scans for migration resources in either old-style "scattered" (one file per version)
or new-style "integrated" format (single file "Resources/{StoreName}.migrations[.nnn]") with ":VERSION nnn" sections).
In the new-style migrations it also recognizes ':GO' separators for parts of the SQL script
that must be sent to the server separately. The old-style migrations are loaded each in one piece
and don't support the ':GO' feature.
Status: TESTED and works fine in all modes!
|
|
Scans for migration resources in either old-style "scattered" (one file per version)
or new-style "integrated" format (single file "Resources/{StoreName}.migrations[.nnn]") with ":VERSION nnn" sections).
In the new-style migrations it also recognizes ':GO' separators for parts of the SQL script
that must be sent to the server separately. The old-style migrations are loaded each in one piece
and don't support the ':GO' feature.
Status: TESTED and works fine in all modes!
|
|
This ensures that 'cmd' gets disposed on errors
|
|
|
|
with it rather than within the loop
disposing of it within the loop causes Mono.Data.Sqlite.dll to get upset, and it's the wrong behaviour anyway
|
|
updating the Migrations table will not occur in a partial transaction, which
would be auto-rolled-back later.
|
|
missing from response, namely Library and Friends. Appearance service is also missing.
|
|
friends migration, which MAY error out if the old friends table is not
in the same database as the new one being created. This error is nonfatal,
it would only mean that friends will not be migrated automatically. It would
bite people with nonstandard configurations.
|
|
the grid server after running for several hours
|
|
Local member in MySQLAssetData
* First pass at applying the using(){} pattern to IDisposable objects. Always use the using pattern on IDisposable objects whenever possible, do not manually call .Close() or .Dispose() unless there is no other way to write the code. This pass mostly covers OpenSim.Data.MySQL, and should have no functional change (tests still pass)
|
|
LICENSE.txt.
|
|
|
|
causing problems with big migrations
|
|
* This patch gets NHibernate working
*** PLEASE NOTE: This patch now requires the libmono-oracle2.0-cil library to be installed on Ubuntu (to stop the System.Data.Oracle missing failure) ***
* Not sure what the dependency is on other distros. Adding this info to http://opensimulator.org/wiki/Build_Instructions would be most welcome
* Adds Castle.* libraries that were missing last time (note, dlls have been added from http://downloads.sourceforge.net/nhibernate/NHibernate-2.0.1.GA-bin.zip)
|
|
for now.
|
|
MsSql2005 to NHibernate.
* Patch courtesy of Adminotech and realXtend. Thanks!
|
|
|
|
|
|
Thank you, RuudL, for a complete adaptation of migration and estate
data to MSSQL, and the updating of the RegionData handling in MSSQL.
|
|
|
|
* commented out [Obsolete(....)] attributes where no replacement feature
was available: if we want to attribute code that we think needs to be
reworked, we should define a new attribute and use that instead
(together with a little tool to retrieve all the attributed code then)
* commenting out unused variables
|
|
migrations (like converting asset ids) don't run into the
normal 30 second kill switch.
|
|
side so that people don't think we're hung.
|
|
changes it wouldn't come up).
include embedded dialect specific sql files for nhibernate migrations
figure out how to get the raw db connection so that migrations can
work with nhibernate.
create initial migration for NHibernate + SQLite + Assets.
|
|
more clear
|
|
asset uuids no longer binary. I've tested this migration a
few times, and it seems working in all the scenarios I've found
but it wouldn't hurt to backup your asset db before running this
as it does touch a very sensitive part of our content system.
|
|
hard-coded port numbers.
|
|
is now using migrations instead of the old model to
create tables. Tested for existing old tables,
and for creating new ones.
|
|
migrating from the old to the new system. Silly legacy
code.
|
|
tracking, and support sub types that we'll need for
nhibernate.
|
|
|
|
|
|
Tomorrow I'll start trying to integrate them into sqlite
to see if this works right for table migration.
|
|
really turned out to be small enough to not need it's own
assembly
|
|
|
|
this becomes useful
|
|
intent is to create an easier way to manage database table versions
like the model used for ruby on rails migrations.
|
|
Fix spelling typo (Thanks ChrisDown for pointing this out)
|
|
them out.
* Fix for Scene.Inventory.cs - It assumes every entity at startup is a SceneObjectGroup. (Actually, this shouldn't have compiled[!] without a warning.)
* Fix for LandManager at startup - it assumes there's a land channel when perhaps there isnt. (Bug that needs another refactor to fix. [Mike - I've assigned a ticket to you about this])
|
