Hi Sinclair,
I'd expect there are plenty tools out there that still run with MarkLogic 4, and enable you to pull out a subset of content. Tools like recordloader, xqsync and xmlsh come into mind. They might facilitate direct migration from one to the other. About configuration, depending on complexity, using Admin might not be as bad as you think. You can very easily do a back to back comparison. On other hand, if it involves many indexes, users, roles, etc. That can be error prone. In that case it might be worth looking into Roxy. Roxy uses an XML syntax to declare all such configurations that resembles the raw config XML inside MarkLogic Data dir. You could try recreate your server settings in that XML syntax, and do bits of copy past from the raw xml configs along the way. Once you have that XML, you can use Roxy bootstrap function to push the settings to any MarkLogic environment. It would both help migrate more quickly, and be better prepared for future. It could also be that someone in the Roxy team knows a trick to capture existing configs into Roxy XML? Cheers, Geert *Van**:* general-boun...@developer.marklogic.com [mailto: general-boun...@developer.marklogic.com] *Namens *Cram, Sinclair *Verzonden:* woensdag 19 februari 2014 11:16 *Aan:* general@developer.marklogic.com *CC:* Archer, Stephen X; Minford, William *Onderwerp:* [MarkLogic Dev General] Migration from version 4 to 7 Hi All, Over the next few months, we are migrating our test (UAT) and Production MarkLogic environments from version 4.2-6 running on older hardware, to 7.0-2.1 on new kit. Installing the software is straightforward enough, but the challenge we have is copying both configuration details and metadata/documents from the Forests. After consulting MarkLogic Support, we have been advised to upgrade from version 4 to 5, in order to use the MLCP tool. This would be extremely difficult as we only have a single UAT environment, which is heavily relied on and we would like to avoid disruption to that service if possible. The alternative suggested is manual configuration via the Admin console and to use backup/restore for the data. The latter would be particularly difficult as our new development server (an additional environment we didn't previously have) has significantly less disk space available (200MB in new Dev vs 3TB in old UAT), therefore we would be looking to extract a subset of the data. Has anyone overcome a similar problem or aware of other solutions? *Sinclair Cram* | Client Technology | J.P. Morgan | Alhambra House, 45 Waterloo Street, Glasgow | J.P. Morgan <http://www.jpmorgan.com/> | T: +44 141 228 0421 This email is confidential and subject to important disclaimers and conditions including on offers for the purchase or sale of securities, accuracy and completeness of information, viruses, confidentiality, legal privilege, and legal entity disclaimers, available at http://www.jpmorgan.com/pages/disclosures/email.
_______________________________________________ General mailing list General@developer.marklogic.com http://developer.marklogic.com/mailman/listinfo/general
