OnTAP 9.9.1 Issues

[TMADOCTHOMAS]

I'm aware of the following post which outlines many user's woes with the OnTAP 9.8 GUI (including my own). Now that many of us are on 9.9.1, I'm starting a thread to summarize what issues remain on this version. Hopefully this can 'go up the chain' to encourage further restoration of functionality that was stripped out of 9.8 and above.

https://community.netapp.com/t5/ONTAP-Discussions/Simplifying-ONTAP-User-Experience/m-p/163639 

Please chime in if you've had the same or additional issues. Also, if someone is aware, please let us know if any of these issues are resolved in 9.10.1. For the record, I really like the new GUI as a whole and find the new features useful, but the loss of very basic and critical functionality at the same time takes away from the enjoyment to say the least.

• Efficiency Details – cannot view compression and deduplication job status and statistics on a per volume basis.
• Event Log – you only get the full text of an event when you point to it, in a tool-tip format. Expanding it doesn’t give you the full text either. It is critical to be able to select/copy/paste events so they can be sent word for word to NetApp Support and searched on Google.
• SnapMirror Policy Choice – unlike before, you can’t select a snapmirror policy. We use network compression on all of our jobs so we have a single custom policy we use for every job. This means that for the first time in 10 years since we started using NetApp, I won’t be able to create SnapMirror jobs in the GUI because we need to use a different policy.
• Aggregate Percent Full – restore the % full column in the Aggregate (i.e. “Tier”) list. For that matter, restore the columnar Aggregate list! It was so easy for 10+ years to just glance and know % full for each aggregate, and now you have to eyeball it and guess. I use this stat frequently. I can get it from ActiveIQ Unified Manager but would like to have it here also. To me this is very basic. It’s a storage product and you want to know the % of storage used!
• Volume Percent Full – at least the Volume list can add a Percent Full column back, but you have to do it every time. There’s no way to save a view. Would request either restoring the column permanently or at least allowing us to save a view. I use the % full column frequently. Again, this is as basic as it gets for a storage product!
• Volume Resize Autosize Amount – previously, going back 10+ years, when you resize a volume the autosize amount automatically adjusts to 20% of the new volume size. In the new GUI, the autosize amount doesn’t change. You have to pull out the Calculator and manually recalculate it. Considering we have autosize turned on for EVERY VOLUME, this is a major pain point considering the fact that resizing is the most common task we perform on a weekly basis.

[TMADOCTHOMAS]

AND, there's no option to assign a drive letter to the LUN. The LUN shows up offline on the server. You have to manually assign a drive and online the LUN. Easy to do, but all of this was done automatically before. Why would NetApp switch from a clean automated process to a clunky manual one?

[pedro_rocha]

actually the LUN letter was always set at the server side when you format it

same for the online process. This seems the same as always

[TMADOCTHOMAS]

@pedro_rocha you are correct. With all my frustrations today, I will retract this one. As long as you add the LUN from SnapCenter it onlines it and adds a drive letter (although currently it always adds the WRONG drive letter since this is a failover cluster - that's another story; it's a bug we're waiting for a fix on from NetApp). Adding it from System Manager has always required onlining manually, you are right.

[TMADOCTHOMAS]

One more issue with this: if you pre-create the volumes as I just did, you have no option to create a LUN within the volumes you created. You have no choice but to create a new volume with the new LUN.

[TMADOCTHOMAS]

When patching OnTAP, I typically do NDUs rather than ANDUs but I'm trying ANDU via System Manager tonight. One point of feedback: it would be helpful to have feedback regarding progress on uploading a system image. I've been uploading for 10 minutes and don't know if I have 5 minutes left or an hour left. It would help for planning purposes. This is more of an enhancement request as I don't recall what the GUI showed under 9.7 (I used ANDU a few times on our DR system).

[TMADOCTHOMAS]

To update my last post, turns out we had new firewall rules in place that were causing this issue - however this came out when trying to do the same process through the command line. It instantly pointed to the issue, while the GUI gave us nothing. Therefore I still submit that there needs to be an improvement in the GUI.

[TMADOCTHOMAS]

This thread has gotten lengthy and there are many extra points added, so I thought I'd summarize everything up to this point:

• Efficiency Details – cannot view compression and deduplication job status and statistics on a per volume basis.
• Event Log – you only get the full text of an event when you point to it, in a tool-tip format. Expanding it doesn’t give you the full text either. It is critical to be able to select/copy/paste events so they can be sent word for word to NetApp Support and searched on Google.
• SnapMirror Policy Choice – unlike before, you can’t select a snapmirror policy. We use network compression on all of our jobs so we have a single custom policy we use for every job. This means that for the first time in 10 years since we started using NetApp, I won’t be able to create SnapMirror jobs in the GUI because we need to use a different policy.
• Aggregate Percent Full – restore the % full column in the Aggregate (i.e. “Tier”) list. For that matter, restore the columnar Aggregate list! It was so easy for 10+ years to just glance and know % full for each aggregate, and now you have to eyeball it and guess. I use this stat frequently. I can get it from ActiveIQ Unified Manager but would like to have it here also. To me this is very basic. It’s a storage product and you want to know the % of storage used!
• Volume Percent Full – at least the Volume list can add a Percent Full column back, but you have to do it every time. There’s no way to save a view. Would request either restoring the column permanently or at least allowing us to save a view. I use the % full column frequently. Again, this is as basic as it gets for a storage product!
• Volume Resize Autosize Amount – previously, going back 10+ years, when you resize a volume the autosize amount automatically adjusts to 20% of the new volume size. In the new GUI, the autosize amount doesn’t change. You have to pull out the Calculator and manually recalculate it. Considering we have autosize turned on for EVERY VOLUME, this is a major pain point considering the fact that resizing is the most common task we perform on a weekly basis.
• Subnets - restore the ability to manage subnets in the GUI.
• Scrolling - scrolling down long lists of objects (volumes, LUNs, etc.) can sometimes result in the list turning into grey boxes. The list then automatically refreshes and you scroll back to the same spot, which automatically turns the list into grey boxes again.
• Efficiency Policies - when editing a volume to change the efficiency policy, you don't see the list of policies available at the cluster level. The only choices are Default and None. Previously, all cluster and SVM level policies were shown as options. We have always configured our policies at the cluster level and used them across SVMs, so this is another case of having to go to the command prompt.
• Portsets - restore the ability to configure a portset in System Manager. (Resolved in 9.10.1)
• Creating Volumes in an SVM Hosting iSCSI LUNs - restore the ability to create volumes in an SVM hosting iSCSI LUNs in System Manager.
• Fix the Following Issues in the LUN Creation Wizard
  • The volume and LUN are created with the same size, leaving no space in the volume for snapshots.
  • No option to change autosize / autodelete options until after the fact. I strongly dislike the default options.
  • The phrase "Name Prefix" makes no sense. Should just be "Name".
  • Under initiator group, why isn't there an option to create a new initiator group that isn't part of an existing initiator group? That's what I would need over half of the time, and it's not an option.
  • The Enable SnapMirror functionality is awful. You have to choose the "Asynchronous and Synchronous policies could not be retrieved" "policy" to even see that you have the option of choosing the correct destination cluster and SVM. Significantly worse is that the drop-down protection policy list doesn't include custom policies. We use a custom policy for EVERY SnapMirror job, making this wizard useless.

[MouBou]

Thank you for the findings and the report.

It looks really disappointing and we still regret switching to the version 9.8+.
Seems like there will be no real improvement in the near future.

[TMADOCTHOMAS]

@MouBou , I should note that several NetApp developers involved with System Manager are paying attention and working on resolutions to many of these issues, however when they appear is anyone's guess. I've actually heard they may start putting some fixes in earlier versions via patches.

[TMADOCTHOMAS]

Just encountered another issue, this one specific to AFF systems:

• Efficiency Settings on AFF Systems – restore the ability to configure efficiency settings on AFF systems. While we leave the auto policy on most volumes, we’ve discovered that adding scheduled efficiency on NFS datastores and a few other volume types provides additional savings beyond inline/auto, so we would like the ability to schedule this to remain available.

[TMADOCTHOMAS]

Annnd .... yet another new issue.

• LUN Display in SAN Initiator Groups - restore the LUN ID column to the list of LUNs when you click a SAN Initiator Group. Also, fix the list of LUNs so complete LUN names appear instead of each one being truncated, with no ability to resize a column to see the entire LUN name.

[KBIT]

In the GUI - Edit Volume the autosize setting is not displayed at all, although it is activated for the volume. In the CLI we can make the setting.

Sometimes we have to set the volume size 2-3 times in the GUI.

[TMADOCTHOMAS]

@KBIT I can honestly say I've not experienced this in particular, but maybe you're on 9.8 or a different version that has those issues? I can see the autosize setting for all our volumes. My main issue is it doesn't automatically recalculate when resizing the volume anymore.

[TMADOCTHOMAS]

Another new one:

• Event Warnings - two additional issues with the Event Viewer related specifically to warnings.
  • Warnings don't automatically clear after 24 hours like they used to. I have one warning from a volume move that I cancelled a couple of days ago, and it's still here. This will obscure my noticing of any new warnings going forward.
  • When trying to filter the Events Log to show more details about the warning message, I noticed that there is no Warning severity level to filter with! It says "Warning" on the front page and on the top of the Events page, but when you filter the only options are Emergency, Alert, and Error.