I have need to change the IP address of a FAS2240-4 in both production and disaster recovery. The change needs to be made to the storage controllers, shares and replication ports. I believe that most of what I am looking for is contained in the /etc/rc and the /etc/hosts files.
Not really sure where to start or the order of operations as I am very new to NetApp. Any help would be greatly appreciated.
Assuming you are running 7mode since you mention the rc and hosts files, most of the IP related configuration is in those two files. There are a few other spots you may need to check as well.
rdfile /etc/resolv.conf - to see if your name servers need to be updated
rdfile /etc/snapmirror.allow - to see if any entries are using IP addresses
options timed.servers - to see if ntp servers need to be updated
options nis.servers - if using NIS
Repeat for each vfiler, if you have vfilers, and don't forget about the SP:
Thank you for the repsone to my question. The array is replicating from a production to a DR site. Were is this IP information kept and what needs to be changed there.
Thank you in advance for your assistance!
/etc/snapmirror.allow and /etc/hosts are the two to look at for replication purposes on both the source and destination, as well as any other mechanisms used for name resolution in your environment.
Thank you for your continued support/information. I really appreicate it.
My apologies but I do have a few addiitonal follow up questions if i could.
We will be making the IP address changes in the DR site one weekend and the produciton site the following weekend. Is there an order of operations that I should be following to perform these changes. As I read it, I will need to make changes to the following:
options timed servers - Can I display these settings ahead of time?
options nis.servers - Can I display these settings ahead of time?
sp setup - Can I display these settings ahead of time?
Should items above be performed in a specific order? Should I restart/reboot anything before or after any of these steps?
Lastly, can you confirm that I should be using the serial conection of the back of the array to perform these steps as I will lose connectivity during some of these activities?
Once again, thank you very much for your continued support!
I usually do that kind of work from either the console cable or the service processor. wrfile can be error prone so watch for paste buffer errors.
If you do the DR site first, you may need to update the hosts and snapmirror files on the source as well to keep replications updating. But those don't need a reboot on the source.
You can view all those settings ahead of time. Most, if not all, should be captured by config advisor as well. For the SP, use 'sp status'. On newer versions of 7mode it'll give you a message to use the ngsh style syntax instead but it should still work.
Once again, thank yo ufor the continued answers to my questions. They are certainly appreicated.
If I could trouble you with yet a few more questions...
Following is a high level list of the tasks to support the work being done to change the IP addresses. Please note that we are going to leave replication down for the week between the changes of the IP addresses in DR and Production:
Is there anything obvious in the above steps that I am missing?
Also, you mention that most of the settings that I am looking for will be captured by config advisor. How do I access that information?
Lastly, do I need to reboot/restart anything during this process?
Once again, I really appreicate your input on this. While I have deep experience in EMC, Pure and HP 3PAR is haven't done too much with NetApp so I am learning as I am going...
Sure, but the level of advice I can give you in a forum thread doesn't compare to what you could get by engaging a partner to help out on these two cutover events. If you're fairly new to NetApp its worthwhile to shadow someone who's done these things for a while.
ConfigAdvisor is a tool found in the utility toolchest on the support site. It collects configuration data and runs a number of best practice checks. It should be run before and after maintenance operations such as this. The data it collects can be used for planning and reviewed later if needed.
After you run sp setup to change the SP ip, you don't need to reboot the system. Just move the switch port it is connected to to the new network (if required), then log into it on its new ip. From there you can get back to the system console with the command 'system console', and perform the rest of the work from there. Remember it shares a physical port with the e0M insterface.
After updating the on-disk config (/etc/rc,/etc/hosts, etc), then the controllers need a takeover/giveback cycle for those changes to take effect. You want to confirm the changes persisted, worked as planned, and the interfaces are all still reacheable during a takeover/giveback. Plan on doing a second round so you know either node can takeover successfully with the updated configuration.
Also you don't necessarily need to keep your replication down for the week. As long as the 2 sites can communicate, and the destination systems can find/reach the sources, replication can resume.
There may be other elements in your configuration, such as IPspaces and vfilers, which could add additional touch points for this kind of work.
Also there are likely other things in the environment to account for. DNS records, iSCSI initiators, NFS automounters, management servers, switch port configurations, routers, etc etc.