Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Mute
- Printer Friendly Page
Using 7MTT 3.5.0 to perform migration from EOL hardware FAS3220 to AFF300
2023-02-22
10:25 AM
2,814 Views
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
To start I can't provide much insight as far as logs and such as this is a closed network. Although I know I need configuration help, I'm more looking for some information. Some backstories: The hardware is EOL (FAS3220) and has no maintenance and we are trying to migrate the data off so that we can setup our new FAS8300 and utilize the new equipment with support. Working with some smart guys at NetApp we came up with this migration plan of migrating to our new AFF300 first, decomming the FAS3220's, then setting up the FAS8300 and populating that with the FAS3220's data.
I have setup and install the 7MTT on my pc and have configured it so that I can see my filers, I select all the data I'm going to transfer and set up the schedules and the task gets created successfully. When run the precheck I get an error stating an intercluster-lif needs to be created, that's fine. I create it using the sporadic information available to me, on the AFF300. When I run the precheck then it successfully goes to start baseline. At which point I left for the weekend, when I came in Monday morning noticed that it had errored out and was still at the start baseline portion. I lost the error but the intercluster-lif seemed to be the point of contention.
What I'm asking for I guess is what is the proper procedure for this intercluster-lif creation? Do you create one on the remote 7-mode cluster and the Ontap cluster? Do the SVM's need one as well?
*Never received any Ontap training just picking up everything on the fly! XCP isn't bad but these FAS3220 are so bad if it's migrating with xcp and a developer hits the storage for a test it kicks everyone off and makes the shares inaccessible for like 15-20 mins.
Solved! See The Solution
1 ACCEPTED SOLUTION
netappbetatester has accepted the solution
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Data LIF = LIF that is used for NAS/SAN traffic
Management LIF = LIF used to manage ONTAP or a SVM
Cluster LIF = LIF for the private backend communication between cluster nodes (think intracluster LIF if that helps, like intranet vs internet if you can't keep intra vs inter straight)
Intercluster LIF = LIF used to communicate to a node on another cluster, typically for SnapMirror, FlexCache, etc.
7 REPLIES 7
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
XCP is a multi-threaded copy engine. It is powerful, and the 3220 may not have enough disk horsepower to keep up (BSAS 7200 RPM 3.5" disks right?). You could try turning down the concurrency or # of threads in XCP so it isn't such a hog on performance.
If you had the specific error for 7MTT that would help.
Another thought would be ndmpcopy.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Yeah I think you are on the money about the drives not being able to handle it. I tried at -parrallel 4 and it would still shut it down.
I'll try and create a intercluster lif on the AFF300 and see if I can get the error to pop up again.
Do I make a Data intercluster lif or a default? I get confused on their purposes.
netappbetatester has accepted the solution
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Data LIF = LIF that is used for NAS/SAN traffic
Management LIF = LIF used to manage ONTAP or a SVM
Cluster LIF = LIF for the private backend communication between cluster nodes (think intracluster LIF if that helps, like intranet vs internet if you can't keep intra vs inter straight)
Intercluster LIF = LIF used to communicate to a node on another cluster, typically for SnapMirror, FlexCache, etc.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Well after that breakdown I was able to get the 7MTT to work, I used the GUI on the Ontap to create an intercluster data lif. Trying to use the CLI was just to much for my smooth brain, but once that was done I went and ran the precheck again and it was successful. I have started the baseline now and everything seems to be going well! Thank you for the info!
To answer your question about ndmpcopy, thanks for telling me what that was! The engineers I have been working with suggested xcp, 7MTT, robocopy, and rsync. They weren't sure if 7MTT was still supported but looking at it, it was perfect for my hell hole of an environment.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Glad to hear! I'll create a KB "What are the different kinds of LIFs" or have it created. That may help too. Was it just the wrong LIF type was created?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Yes, initially I created a default intercluster lif using the CLI and it just would not work when I tried using it for the migration. But when I tried creating the same LIF type through the GUI it straight up told me that wasn't possible. So, using what you explained I created an intercluster data lif and it spun up just fine.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Ah ok sounds good. I hope it goes good. If you need more help we're on Discord as well. 🙂