nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Disaster recovery

Jenni

Hi All,

We are trying to set up a DR system for the TeamSite server and wanted to know if any one can share any documents/guides on the same. Can any one list out the various considerations/tips before we can set up a DR system?

TS 6.5
OD 6.1 DAS enabled
Solaris 10

Thanks!
Jenni

Find more posts tagged with

Comments

Lozza

Here's some very high level information from what I've experienced.

Commonly I've seen DR implemented by having IWHOME and IWSTORE on remote storage, say a SAN or some kind of shared filer. These are mounted onto the Production box in a normal situation, and in the case of failure, they are remounted onto the DR box.

This is the simplest method I've seen. However consideration needs to be taken of config files in IWHOME which contain a hostname or IP address, such as iw.cfg, several OpenDeploy config files, SSL certificates and others.

For systems where IWHOME and IWSTORE are on local disk, you need to take backups / snapshots (freezing your stores first), and have a process for regularly mirroring them onto the DR storage, or restoring them to DR from backup in case of failure.

An idea I've discussed recently with colleagues is to test DR by using Virtual Machines. TeamSite 6.7.2 is available on Linux, and you can inexpensively set up a DR scenario by installing TS on two VMs. You can simulate the shared storage on IWHOME and IWSTORE with virtual disks on a "Prod" VM, move them into the "DR" VM and see what happens. This wouldn't be a thorough test of your DR strategy, but it would certainly highlight the main problems so you can plan accordingly. Caveat: I've not actually done this yet, but it feels like a good idea.

It's vital to test out your DR procedures in real life though, perhaps running production off your DR box for a couple of days during a non-critical period for your business (if such a time exists), backing everything up before hand.

Well these are my thoughts, I hope they're vaguely useful. Maybe some others will be able to share some more concrete plans.

Adam Stoller

...
An idea I've discussed recently with colleagues is to test DR by using Virtual Machines. TeamSite 6.7.2 is available on Linux, and you can inexpensively set up a DR scenario by installing TS on two VMs. ...

Inexpensively in terms of hardware and OS perhaps, but not in terms of licenses for TS and whatever other IWOV products you need to use for DR verification...

The use of remote storage devices and checking config files is generally the right track - though if you're dealing with DR - you probably need to deal with mirroring of the remote storage because it would have to exist within a different data center (otherwise you're talking about just failover of a TS server within the same data center as opposed to "disaster" recovery).

Jenni

Thanks for the responses guys. Is there a doc/KB with reference to DR implementation?

-Jenni

c911s

Jenni, did you ever find out if there are DR documents available? We are looking for DR documentation and haven't been able to find anything newer than 2002. Would appreciate any info you or anyone could give me.

Thanks,
Janet

Crsb

There are no published docs from Autonomy about a DR strategy or best practice.
In my opinion, you would have to go with what your organization has or wants to have in place for all of your applications not only TeamSite and probably for the entire datacenter. Whatever your organization decides that will work for all your applications is what you will implement for TeamSite, for example if they go with VMs, multiple datacenters, Network Storage vs local drives... If you know how TeamSite works you should be able to make it work within in DR implementation.

Binettoid

If anyone is interested, I can provide a plan for migration or recovery of 7.x.
PM me if you need this.