When the engine can no longer open a database, recovery means extracting your data out of it regardless. Every job runs off a copy of the original files (the originals never get touched) as we repair the internal structure one page at a time, or build the database directly from the data file when the log is missing. SQL Server, MySQL, Oracle, PostgreSQL and Access all come back the same way: your tables and records, extracted and validated.
When a database is corrupt, stuck in Suspect mode or refuses to mount, you need to leave it alone. Do not run repair commands that allow data loss. Do not restore over the top of it. And whatever you do, don’t continue to write to it. Doing any of these things could turn a recoverable problem into a permanent one. Make a copy of the files as they stand, note down what happened, and let us assess it first. The safest recovery starts with an untouched copy of the files.
Corrupted MDF files, dropped tables, Suspect-mode databases, missing or damaged transaction logs and databases stranded on failed storage: the list below covers the database problems that come to us most, from corruption of individual pages up to complete failure of the underlying storage where the data resides.
We recover any database engine, including SQL Server, MySQL, Oracle, PostgreSQL, Microsoft Access and the rest, whether the files are corrupt, objects have been dropped or the entire database is sitting on a failed drive or array. The method depends on which engine you use, but the engine does not decide whether the data comes back.
MySQL and MariaDB (InnoDB, MyISAM), Microsoft SQL Server (MDF, NDF, LDF), Oracle (DBF), PostgreSQL, Microsoft Access (ACCDB, MDB), Exchange (EDB), SQLite and MongoDB · corrupt files, suspect databases, dropped objects and failed underlying storage.
Our approach is fixed: work only from a copy of your database, then either repair the database at page level or rebuild it from the data file when the transaction log is missing, and finally extract and verify your tables. If the storage underlying your database has failed, we address that first.
Let us know your database engine and its version, which files you have and a description of the problem. An engineer will review this information and send a written quote, usually within 48 hours.
First we make a read-only copy of the files (MDF, NDF, LDF, IBD, DBF or EDB), so every operation is run on the copy and your original files remain untouched.
Where the drive, RAID or server hosting your database has failed, we first repair and image it or rebuild the array, so that we have clean files to work from. Only then do we touch the database itself.
Repair happens at page level: corrupt pages, checksums and headers within your database are identified and corrected. And if the transaction log is missing or damaged, we rebuild your database to a consistent state without it.
We pull out your tables, records, schema and stored procedures (into a live database, SQL scripts or CSV), including data retrieved from pages the engine had already written off.
We check each recovered table and record for readability and consistency (row counts, key data and overall structure) prior to handing anything back to you.
You get the database back in an attachable state, or your data exported instead, whichever best suits how you need to bring it back into service.
SQL Server, MySQL, Oracle and the rest: we repair corrupted files down to page level and, where the storage has failed too, rebuild that first.
Tell us what happened and we’ll get back to you, usually within a working day.
We’ll be in touch shortly. If it’s urgent, call 01483 901310.
Our database recovery prices follow straightforward tiers, always with a free diagnostic and a written quote provided before starting any work.
A few recent database recoveries, spanning corruption, Suspect mode, dropped tables and failed storage. Identifying details have been removed and every result verified.
The transaction log was holding the database offline. We rebuilt the database to a consistent state without that log and then successfully recovered all the tables.
MySQL would not start because the ibdata file had become corrupted. We repaired the InnoDB tablespace and then extracted the databases.
The rows were still in the data file. We carved the dropped table out of the pages and put it back.
A stretch of blocks had become unreadable. At block level we repaired the datafile and got both the schema and data back.
A corrupt file header was the entire fault. We were able to rebuild it and get the tables and forms restored.
Two of the hard drives under the database had failed, so we rebuilt the RAID 5, then repaired the database on top of it and recovered it.
Our no-charge diagnostic lets you send us your database files along with the engine and version you’re using and what happened. An engineer will review the files and confirm the exact quote in writing prior to starting any work.
Your data recovery process begins when we receive your database files (or the failed storage they reside on). Send them to us safely, let us know which engine and version you’re running, and after completing the complimentary diagnostic we’ll confirm your exact quote in writing prior to commencing any work.
Sending it by post? We’d recommend a tracked, insured service. Dropping it in? No problem: Monday to Friday, 9am to 5:30pm. Please package the device as described above first.
Would you like a scoped quote first? Simply describe the database engine, the files and the fault on the form and an engineer will evaluate it and respond with a tailored quote.
We’ll be in touch shortly. If it’s urgent, call 01483 901310.
The most frequently asked questions about recovering databases.
Generally, yes. Suspect mode almost always means that SQL Server was unable to recover the database from its transaction log at start-up, not that the data has gone. Working from a copy, we repair the pages and rebuild the database to a consistent state, with or without the original transaction log.
Yes, that is the heart of what we do. We correct corruption at page level (pages, checksums and headers) in SQL Server, MySQL/MariaDB, Oracle, PostgreSQL and Access, and extract your tables and records to either a working database or exports.
Often, if you act fast enough. Dropped tables and deleted rows generally remain in the data file until they are overwritten by another write operation. So stop writing to the database and we can recover the records from those pages.
Usually, yes. A SQL Server database can be rebuilt from the MDF (and NDF) alone, and MySQL from its InnoDB or MyISAM files, even when there is no longer a log available. We repair what remains of the database and then recover the data from those files.
Recovery pricing follows a tiered structure. Repairing from readable database files starts at £500 + VAT. If the drive, RAID or server underneath has also failed, prices start at £800 + VAT. Either way, every job gets a complimentary diagnostic and a written quote prior to starting work.
On most jobs, yes. For database repair and any storage recovery, we take a deposit against the quote, and the remainder is only payable once we successfully recover your data. Every job receives a complimentary diagnostic and a written quote first.
All of them: Microsoft SQL Server (MDF, NDF and LDF), MySQL and MariaDB (InnoDB and MyISAM), Oracle (DBF), PostgreSQL and Microsoft Access are all supported. Irrespective of version, we repair the database files and recover your data.
Yes, and we deal with both aspects of the problem. The failed drive is repaired and imaged or, where applicable, the RAID array is rebuilt, and clean database files are produced first. Next, we repair the database itself and extract your data.
It depends. Strong modern encryption on a database cannot be reversed: no one can honestly do that without the decryption key. But we can recover earlier backups, unencrypted copies and any database files that the ransomware missed, so it is always worth asking.
Most jobs are completed within a few working days. Very large databases, severe corruption or an underlying storage failure that has to be resolved first will make it take longer.
Send us your database files on a drive or USB stick, or where the storage has failed, send the original drives or server. Post it fully insured to our Guildford address, or bring it to us Monday to Friday, 9am to 5:30pm.
There is a complimentary diagnostic, then tiered pricing from £500, and in most cases you only pay if we recover your data. We cover every database engine (SQL Server, MySQL, Oracle and the rest), files and storage alike. Begin your recovery today.