timallen
Well Known Member
My boss has asked me for a Disaster Recovery Plan for OneWorld. I looked in the Knowledge Garden and found that this term has a special meaning for World on the AS/400. It deals with a licensing code to be used to bring up the system in an emergency.
So for OneWorld, I assume that I just need to come up with a backup and recovery plan. Is that correct?
I've thought that it could be something like this, assuming an Enterprise Server acting as the Data Server and a single Deployment Server, with at least the path codes DV, PY and PD:
1) As a policy, development should not happen in production, and developments should be either saved to a save environment or checked into the server as often as possible.
2) A backup should be made of the database information on the Enterprise Server and stored separately.
3) A backup should be made of the entire JDE directory of the Enterprise Server and the Deployment Server and stored separately.
4) In event of a disaster, the system would be taken down and the appropriate parts of the system would be recovered from tape.
I also thought about how we could handle this according to how much downtime the client could handle. If the system had to be high availability, backups to tape would not be enough-- we would have to have hot spares for the ES and the DS, and maybe have journaling on the database (though this would effect performance adversely).
Has anyone done a Disaster Recovery plan for OneWorld? Is my plan basically sound, or are there problems?
Thanks in advance.
So for OneWorld, I assume that I just need to come up with a backup and recovery plan. Is that correct?
I've thought that it could be something like this, assuming an Enterprise Server acting as the Data Server and a single Deployment Server, with at least the path codes DV, PY and PD:
1) As a policy, development should not happen in production, and developments should be either saved to a save environment or checked into the server as often as possible.
2) A backup should be made of the database information on the Enterprise Server and stored separately.
3) A backup should be made of the entire JDE directory of the Enterprise Server and the Deployment Server and stored separately.
4) In event of a disaster, the system would be taken down and the appropriate parts of the system would be recovered from tape.
I also thought about how we could handle this according to how much downtime the client could handle. If the system had to be high availability, backups to tape would not be enough-- we would have to have hot spares for the ES and the DS, and maybe have journaling on the database (though this would effect performance adversely).
Has anyone done a Disaster Recovery plan for OneWorld? Is my plan basically sound, or are there problems?
Thanks in advance.