vtic means the partner target port in 7-mode, how could it be difference in cDOT mode? how could the partner target port be represented in cD0T?
... View more
I remember where should be alway a "vtic" after logged in, and along with a port number, in (logged in). How come don't I see it in the following command? I am using 8.3.1 Thanks! cluster::> igroup show -v igroup_host Vserver Name: cluster-fcp-1 Igroup Name: igroup_host Protocol: fcp OS Type: aix Portset Binding Igroup: - Igroup UUID: 9fc9d6ee-1608-11e6-89b1-00a098666b19 ALUA: true Initiators: c0:50:76:07:2e:d9:00:1c (host-hba-1) (logged in) c0:50:76:07:2e:d9:00:1e (host-hba-2) (logged in)
... View more
I have a share and created on a path: share_name --> /volume_name/path_name 1. I know how to create a share on a volume or qtree, but don't know how to create a path and on top of that to create a share? 2. Under what circumstances, I would need to create a share on a path? 3. When I snapmirror the volume over DR site, after I create a clone on the mirror, I don't see the path has been snapmirrored, whey? Can anybody pleasehelp to shed light on this issue?
... View more
When we snapmirrior CIFS shares over to DR site, we will have to use SVM name on DR filer, which is different from Primary site. I am thinking this is not the right way, the application should not change the accessing name. Is there any way to achieve that? Thanks!
... View more
Hi Chris, To follow up the question, for 4x10Gbit ports, to balance out data across two different network interfaces, can I utilize DNS RR and make connections in turn between two interfaces? If yes, to ensure data will go through evenly between two, then two interfaces should be in two sperated routes. Correct?
... View more
I am trying to find a way to use either OnCommand Univified Manager or Performance Manager, or CLI to list SM status in daily basis over last year. Any way to achieve that? Thanks!
... View more
>With SLM, when a new LUN map is created, the LUN is accessible only through paths on the node owning the LUN and its HA partner. In a 4 nodes cluster (2 HA Pairs), what will it happen when HA pair owning the lun got all failed? Clients then could not access the lun at all, not even through indirect access, correct? Please share your ideas. Thanks!
... View more
Diff export, snapshot, retention could be achieved in a shared SVM, as you already indicated. Like I said, the only reason the customer needs a SVM is to put the volume in an isolcated network which is already the case in an existing SVM. Again, this will be just a <1TB volume, ~10 users. If none of those you said have anything to do with the need to create a new SVM, would that be then fair to say that we don't need a separated SVM in our case here? Creating a volume in a shared SVM should be good enough?
... View more
I did not make myself clear. I know a SVM is always needed or required in cDOT. We already have an existing SVM, and we have multiple SVMs already. We use one for NFS volume and in a secured network, so, we can use this one. The question is for this particular application, with < 1TB data, one volume, ~10 users, what would it be the critieria to have a separated SVM? In my opnion, we can just simply export the volume in the same secured network to clients. a SVM is a over-kill. Understand that the capacity is not the only consideration. Multi-tenancy: understand SVM is a logical unit, and can be indepedent on physical environ. delegation: Can you give me an example,pelase? compliance boundaries: this is the only reason that the customer wanted to have a seperate SVM, but the network the existing SVM is based on an already seperated network. In my opinon, we can create a volulme on the same isolated network, which could achive the same security. Isn't authentication requirements: AD authentication, same as all existing. untrusted domains: same AD as other volumes differing application requirements: can you please give me an example, like what type of applicatoin requirements DR requirements: we don't set up DR based on SVM Thank you, JGPSHNTAP , and SeanHatfield!
... View more
I know 1TB is small, that is why I raised the question of if a SVM is needed here in this case. So, are you saying that since this is a small size, and small envirioniment, SVM is not worth doing here?
... View more
We have a TB volume and will be accessed by around 20 users, 10 desktop, would that worth to create a SVM for this application, or should it be just a volume enough? My question is, when will it up to the point SVM should be created? any best practice and criteria? Thanks for you advice!
... View more
We have virtual appliance, and currently only have a few TB on the cache. Users are complaining the backup is running slower on NFS than NFS on the filer. There are not too manu attributes I can set on the data interface, for instance, full/half duplex, speed. Please help, where can I start with to check out the performance issue like this?
... View more
I wanted to transfer 10TB data from EMC VNX to NetApp FAS storage. Can I use two HBAs on the same server for each storage, and use robcopy to transfer data from one to the other? and also install Host Utility for two different purpose of HBAs? In general what speed would look like for 8Gbits HBAs? Thanks a lot for your help!
... View more
Thanks! My question mainly is on backups, not DR, or recovery. I did not make myself clear. Let's say I have 40TB total of backups, with a number of mount points, so, if we cannot selectively delete backup dumps, then 40TB amount of data will be staying in local cashe "forever" on AV appliance without eviction. if we only have total of 64TB cache, then there would be only 24TB space left for other data. Correct? Thanks again for your patience.
... View more
Hi Chris, I am testing on using AV for RMAN nfs backup dump, say, it would do full backup weekly, and Incr daily. Based on my understanding, when the local cache is getting full, then less accessed data would get ejected. In this sense, the backup dump will never get ejected, since the volume get r/w everyday. So, in this situation, I am thinking if I could purge volume/data selectively, I can then test out if I can download data from Cloud. Otherwise, data wlll be staying on AV appliance forever. Am I missing anything here?
... View more
We have an already productions AltaVault but ruining on a virtual appliance. Is there or what detailed procedures to keep all data on AltaVault and in the cloud to convert to a physical one? Thank you!
... View more
It sounds you are auggesting to use SV to replaced SM, then will SV play the same role as SM play on DR for DR purpose? Also, these SSs are created by SMVI not by OnTap Snapshots policy, so, not sure if SM / SV can recognize these SSs...
... View more
Thanks a lot! we are using 8.3p2. Can it support "version-flexible SnapMirror with sm_created label"? If not, Can we temporrarily use SV to replace SM? then I don't have to replicate all SSs? The issue here is that there are too many local SnapShots on Primary, and out of the space on DR, so, by not replicate SSs, I can save the space on DR. Looking forward to hear you again
... View more
When Snapmirror a volume, it will also replicate all SnapShots over to DR filer. Is there anyway I can ignore these SnapShots, or move them to somewhere else without replicating over? Thanks!
... View more
Thanks for your prompt message. I understand in general a clone is identical a flexvolume. But, can you please confirm with me if there will be performance impact when reads is happening on both clone and it's original volume, since they share the same data location before they split off?
... View more
As far as I know, write to the clone will have its own area on the storage that will be sperated from its oringal volume. What about in read situation, reading the original data from the clone and the origianl volume will come from the same location, in this kind of situation, will clone then have performance impact on its oringal volume. Thanks very much for your inputs!
... View more
We wanted to schedule a SnapMirror job accordingly based on other tasks. Is there any way to start the the job in Cron or a scheduling software. If yes, is this recommended? Thanks for your input!
... View more
When I run "qos statistics volume latency show", there are always some volumes like "volume_name-wid#digits" come out. The latency on this volume is very high. What is it? I know "#digits" is the workload ID, but, how could this workload be generated, and it has so high latency? Thaks for sharing!
... View more