Replication left the database in a weird state

On my main DB Server (Running MSSQL Server 2000 latest SP) I'm
publishing the database, mybigdb (540Gb data file - my drive is
560Gb), using transactional and merging replication. Everything was
great until this morning, when I ran out of HDD space - so in order to
clear up some drive space I wanted to DBCC SHRINKFILE the Log
transactional file (which worked in the past), which requires for the
replications to be dropped, the Log transactions to be marked as
reset, and finally I could shrink the Log file. I started taking the
first step:

- Run the SQL Scripts to delete the two replications (Transactional
and Merging). During the dropping process the delete script started
hanging, I waited long enough, until I decided to stop SQL Agent and
SQL Server and restart it... This is when all the problems started...

When I restarted SQL Server...
- Enterprise Manager could not expand the "Databases" folder - it hung
- SQL Query Analyzer could not access the Databases catalog either (it
gave time outs)

I stopped, once more, SQL Server and moved the database 'mybigdb' to a
different directory outside the %SQLSERVER%\DATA directory to
somewhere else (this only required the inode to change without
requiring storage for copy). I then restarted the SQL Server service
once more, this time I could actually drill in the Enterprise Manager
to the Databases folder and the same worked for SQL Query Analyzer.
At this point the 'mybigdb' database was marked by Enterprise Manager
as a 'SUSPECT' database.

What can I do? I need to be able to bring this database back online -
I've taken the following steps, but I am not sure how to proceed to
bring the DB back - also there are a few 'bad' facts about this DB:

Bad facts:
1- No backup of the DB was ever made - so all I have is the MDF and
LDF file of mybigdb.
2- There are no additional drives on the system, so all I have is the
560Gb drive with 20Gb left of space.

New Facts:
1- Bought a 1Tb drive and copied the 560Gb data and log file...
2- Backed up all other DBs (msdb, myappdb,...) on the the server to
the new 1Tb drive

So here are my main questions:

1- Could it be that MSSQL Server, when it tries to start my 560Gb it
needs more than the available 20Gb storage left in the drive?
2- When I look at the publishing server it still believes that there
is a transactional publication of 'mybigdb', but I can not remove it
because the DB is under SUSPECT on Enterprise Manager (and the file is
no longer under the expected location - remember, I moved it).
3- I've tried to DELETE the database from Enterprise Manager (without
deleting the actual file), but Enterprise Manager complains that the
DB is under replication (which it shouldn't be the case)
4- I've Googled around and seen that people state to place the DB
under Safe Mode or Test Mode, so one can access it and stat taking
DBCC dump of it - which I don't want to think is the case yet - since
I have stored procedures and many other things.

I've learned my lesson, and I will NEVER allow for any other
deployment to go without a required backup plan. Any help or
assistance would be greatly appreciated.

.

Relevant Pages

  • Re: Btrieve upgrade options
    ... usually this is done through a drive letter connection or UNC pathname. ... and if drive K is mapped to the database server, ... management console where the database is, map up my drives and launch ...
    (comp.databases.btrieve)
  • Re: Is there an obvious way to do this in python?
    ... database, so why not use a database: ... some kind of update server to which you can periodically send Python ... machines, so that some can do a full range of transactions, and others a limited ...
    (comp.lang.python)
  • Re: Deleting All Forms
    ... The point that Martin was making is that since the FE is just a file on the users local PC, to update the FE its just a matter of the user copying the new version of the FE from a central location and saving it on top of the version that they have already. ... Also the statement 'I know I could just tell them to download the front end from the server but I wanted to avoid that and let them refresh in their own time' doesn't make any sense. ... I cannot access their local drives. ... At present I have the database split and the front end on ...
    (microsoft.public.access.modulesdaovba)
  • Re: Disaster recovery - Big Time!!
    ... > Original drives that were recovered were SCSI, ... > Copy the ExMoved mailboxes to the new server ... I'd then install Exchange 5.5 on a member server that has the same ... need to run ESEUTIL /P <path to database>. ...
    (microsoft.public.exchange2000.general)
  • Re: Newbie to Access Developer Extension
    ... drives goes back to my university days when they where hauling out the ... I've built a two-form database that feeds tables ... in fact a full blown version of ms-access. ... docs on the server. ...
    (microsoft.public.access.devtoolkits)