nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Bad SmartZone UI Glitch, unable to access critical WLAN settings

donald_curry_58

https://youtu.be/2mMWKNllmvw

This is ridiculous. The web interface refreshes itself every time I attempt to click or double click on particular WLANs. Happens with all browsers from any computer I try. Only some WLAN's are doing it. Can't stop the browser before it reloads either.

Restarted controller, updated to the latest 3.6.2 build 250. No change. Can't recreate the WLAN because I can't get into the settings to pull the passphrase.

Find more posts tagged with

Accepted answers

All comments

paul_van_der_cr

is that controller located in-house or somewhere external in a datacenter?

latest 3.6.2 build 250 ? this is not the the latest build...

donald_curry_58

@paul_van_der_cruyssen In house. 250 is the latest build of 3.6.2. Not going to 5+ unless I need to.

paul_van_der_cr

is it possible to make a direct connection between the laptop and te controller? just to get rid of everything in between?

you are right that 250 is the latest build of 3.6.2, and not sure that upgrading to 5 would solve this problem (never have seen this problem before)

donald_curry_58

I think so, but I won't be back onsite until next week. Some of the WLAN it does allow me to click and edit. I can't see a difference between the ones that allow click and not.

syamantakomer

Hi Donald,

Please login into the controller CLI, run below command to sync the database and check if you still see the issue.

vsz_cli> enable
vsz_cli# debug
vsz_cli(debug)# reindex-elasticsearch-all

Command will not cause any network or service impact, it is just to sync the database.

Also attach a screenshot of the error or symptom.

eizens_putnins

It's definitely not a "vSZ glitch", but most probably issue with your installation - data corruption, it looks. Did you have power outage, HDD issue, or networking problem between nodes recently?

vSZ is quit resilient, but if DBs are not in good shape, it will not work properly.

If reindex-elasticsearch-all will not help, I would advice to arrange remote session with Ruckus support, to get internal DBs in synch. Of cause, if you have working backup, restoring it normally is the fastest way in virtual environment. I normally backup all VMs after any major configuration change or upgrade, and restoring in case of any failure takes about 30 minutes...

Was not useful that often, but twice it really saved my day .

If you don't have VMs backup, but have good cluster backup, you can reinstall everything from scratch, and apply backup, it takes about 3-5 hours.

donald_curry_58

Ruckus TAC has spent a couple hours trying to fix the database through the shell, and it's still broken. It is very disappointing that a problem like this requires jumping back to old backups until I hopefully find one that works. This erodes my faith in Ruckus wireless as product, the TAC, and it is overall extremely disappointing. As IT consultants, we rarely have control over battery backup, etc, and would have no way of knowing if power failures could occur. If a power failure can corrupt the database where I can't edit the SSID I have deployed to 1,000 chromebooks, Ruckus needs to be better prepared for fixing the database.

eizens_putnins

IT consultant should know that any critical installation must be protected by backups, because it is the only way to protect service in case of physical or logical failures, including files system corruption.

Cluster makes service resilient for host disconnections, but it is not replacement for backups (same as RAID array isn't).

If you have VM corrupted because of power or any other problems, and don't have backup handy, it is exactly your problem, not Ruckus. You must have backups for VMs as well as configuration backups.

When VM is damaged by power outage or any other external reason (for clusters it may include network faults), it can have any file corrupted -- it is completely random, and there is no safe way to fix it except restoring VM backup.

Actually Smart Zone is very resilient, and usually you don't have any issues even in case of blackout, but sometimes you are not so lucky - and if file system is broken too much, it just can't be fixed to restore all files to original conditions.

This why there are cloud solutions -- clouds are maintained by professionals, and have both resiliency and backup in place. But even than problems happen - it's real life!

donald_curry_58

Fortunately I know that in this case there was not a power failure as it is a primary client where I am familiar with the battery backup. I am familiar with recovering from those scenarios, but not familiar with my scenario where everything appears to work except for a UI glitch that locks me out of specific system settings.

As for your assertion that Ruckus SZ's are resilient, I respectfully disagree. I have 9 clients with smart zone controllers and 20 more with zone directors. In my 10+ years of working with them I would not define them as resilient, so given my current circumstances and history of recovering smart zones, seeing how slowly they boot up, restore, etc, you might understand why I am tilted by your use of the word resilient.

donald_curry_58

Also, this SZ was very recently recovered by RMA and restored from a cluster backup. I suspect the database corruption came along with that, so I feel like I would be wasting time doing cluster and config backups. I have plenty backups. I just don't want to accept that as Ruckus' official answer, because I don't like that answer.

So I have reached out to the Texas engineer rep for my area and linked him to this thread, if Ruckus can't pull up a real fix I'm going to have to re-create SSID's, and pray that the corrupted SSID's can be removed in the command line.