cancel
Showing results for 
Search instead for 
Did you mean: 

how to build in reslience into 2 x ev9 servers

EVMan
Level 4

Hi

 

We have 2 x new physical (win 2008r2) EV9 servers. We have site reslience (physical data centres). One server is running the archiving task and the other the journaling task.

 

The index service is on each server - and the index locations for the journal server are to the jorunal indexes. And the index locations for the mailbox archiving server are mailbox index locations.

If mailbox and archiving tasks are on the same server (we are thinking of setting up VCS - active/passive config) does it share index locations - as you only have one index service and its locations?

With VCS in a 2 x data centre environment would you therefore zone both servers in each site to see the same set of storage?

 

thanks

James

1 ACCEPTED SOLUTION

Accepted Solutions

Wayne_Humphrey
Level 6
Partner Accredited Certified

Hi James,

WOW, a lot easier to understand thanks :)

Because you are using a UNC for writing your indexes, this is simple. I would not go the VCS route.  I would simply just use the Enterprise Vault built in USL (Update Service Locations)

Basically all you do is update the DNS entry for the server that has failed and point it to the working server and run USL from within the VAC and it will update all the new 

services with the information needed to host the failed servers services.

Take a look at the Enterprise Vault Administration Guide, it is covered off well in there and or search here for USL, I have discussed this many times here.

oh, its also known as Building Blocks. Its your way forward I am sure :)

View solution in original post

3 REPLIES 3

Wayne_Humphrey
Level 6
Partner Accredited Certified

This is more of a VCS question rather than Enterprise Vault.  Also you question is very Cryptic and Hard to understand.

Please try and elaborate a bit more on what you are trying to do?

EVMan
Level 4

Ok

 

Ignore the VCS.

 

We have 2 x ev servers. One runs the journaing task and the index service on that server has its locations set to \\nas\indexes\journal

The other ev server runs the mailbox archiving tasks and the index service on this server has its locations set to \\nas\indexes\mailbox

 

Say one of these servers went down. I could add the other task to the remaining server (ie journaling task to the archiving server if the journal server went down). However how would it work then with the index locations for the journaling??

Would it use \\nas\indexes\mailbox  ? and drop new indexes for the journaling task in here?

i.e. is the index service on a per server type basis (journal or mailbox archiving)

thanks

James

Wayne_Humphrey
Level 6
Partner Accredited Certified

Hi James,

WOW, a lot easier to understand thanks :)

Because you are using a UNC for writing your indexes, this is simple. I would not go the VCS route.  I would simply just use the Enterprise Vault built in USL (Update Service Locations)

Basically all you do is update the DNS entry for the server that has failed and point it to the working server and run USL from within the VAC and it will update all the new 

services with the information needed to host the failed servers services.

Take a look at the Enterprise Vault Administration Guide, it is covered off well in there and or search here for USL, I have discussed this many times here.

oh, its also known as Building Blocks. Its your way forward I am sure :)