This post is more than 5 years old
3 Posts
0
775
Seeding at local site and moving to DR site afterwared
Hello,
I understand that DM will assist in changing the IP addresses, but how about in cases where we need to change IP on the remote cluster due to site move; total shutdown of the remote cluster and array? Is the DM process made for a situation like that?
We plan on seeding a RP/SE CRR setup locally, shutting down the VNX array and RP cluster that are to be shipped to their ultimate remote DR sites and re-ip. Would something like this cause a full sweep (which we are trying to avoid)? The BW between the two sites is pretty small (7mbps), hence the seeding but I am afraid that while in transit, and depending on change rates, we may end up full sweeping anyhow. At what point, if the re-ip is doable above, would I have to worry about a full sweep being needed anyhow due to so many changes on the production site while target environment is in transit? At what point does RP say there are too many dirty blocks and therefore need to do a full sweep?
Thank you.
forshr
1.1K Posts
0
June 19th, 2013 06:00
That all depends on the level of writes changes and to which blocks, i.e. the same block could be written to multiple times in the downtime window whilst other blocks are never written to. When you pause a CG the splitter will mark the writes as metadata in the local journal and when the CG is resumed this data is read from the journal and subsequnetly read from the array and transferred. RP will not default to a full sweep requirement based on any amount of writes.
nnaser1
3 Posts
0
June 19th, 2013 06:00
Thank you Richard for the response and the procedure.
Do you have insight into the second part of the question?
At what point, if the re-ip is doable above, would I have to worry about a full sweep being needed anyhow due to so many changes on the production site while target environment is in transit? At what point does RP say there are too many dirty blocks and therefore need to do a full sweep?
Regards,
Nader
forshr
1.1K Posts
1
June 19th, 2013 06:00
Hi there,
Here is the procedure to follow:
1. Pause all CGs.
2. Before shutting down the equipment make sure the journal lag is zero.
3. Make the IP changes to the RPAs and VNX SPs - the RPA IP address changes will need to propogate to all RPAs. Use DM and check this for all RPAs using boxmgmt after making the changes.
3. Move the equipment and perform connectivity testing once back up.
5. Resume the CGs.
This should avoid the full sweep. Let me know how it goes.
Regards,
Richard
nnaser1
3 Posts
0
June 19th, 2013 07:00
Thank you!!
Steve0_2012
73 Posts
0
June 19th, 2013 08:00
I am out of the office until 24/06/2013.
Note: This is an automated response to your message "Re: -
Seeding at local site and moving to DR site afterwared" sent on 19/06/2013
14:40:53.
This is the only notification you will receive while this person is away.
Steve0_2012
73 Posts
0
June 19th, 2013 08:00
I am out of the office until 24/06/2013.
Note: This is an automated response to your message "Re: -
Seeding at local site and moving to DR site afterwared" sent on 19/06/2013
14:26:54.
This is the only notification you will receive while this person is away.
Steve0_2012
73 Posts
0
June 19th, 2013 08:00
I am out of the office until 24/06/2013.
Note: This is an automated response to your message "Re: -
Seeding at local site and moving to DR site afterwared" sent on 19/06/2013
15:18:54.
This is the only notification you will receive while this person is away.
Steve0_2012
73 Posts
0
June 19th, 2013 08:00
I am out of the office until 24/06/2013.
Note: This is an automated response to your message "Re: -
Seeding at local site and moving to DR site afterwared" sent on 19/06/2013
14:05:52.
This is the only notification you will receive while this person is away.