Controllerless Networks

We have a campus with 233 Aruba AP's: a mix of 204s, 205s, and 275s.  On this campus, in AirWave, 30-50 AP's will show up in the down status, even though they are not actually down.  After a few minutes and a refresh, they will show in the up status.

 

Is this because of the large number of devices on one Instant Controller?  If so, would it help to add a second Instant Controller? 

 

I have already tried increasing the poll interval so it has time to get statuses on all the APs.

Hey, there could be a number of issues as to why this is a problem. I've seen deployments of 800+ IAP's in AirWave without issues. 

 

Each IAP individually contacts AirWave, so bare this in mind. When the AP is reporting as down in AirWave, what do you see in the logs on the IAP in question?

 

Does the AirWave server have enough resources (check the AirWave sizing guide) to handle the IAP's in your deployment? Do you have sufficient licenses?

Is there any latency issues or bandwith problems?

 

What version of AirWave and InstantOS are you running?

We have 1139 APs in AirWave, but we only have issues with one campus of 233 APs.  When they go "down" in AirWave, the device shows nothing in its logs, as it is not actually down.

 

Our server is pretty beefy with dual Xeons and 160GB of RAM.  Also, if it was a server issue, I would expect to see issues at some other campuses, besides the one.  We have plenty of licenses, and no network issues.

 

We are running AirWave Management Platform 8.2.3.1 and the APs are running version 6.4.4.8-4.2.4.4_57496.

 

 

Hi Matthew,

 

Is the VC reporting the IAP's to be down as well ?

 

Please check the output for the following commands:

 

VC:

 

#show aps

 

In the above command, check the uptime for the slaves which are reported as down.

 

Slave:

 

#show version.

 

The above command will list the actual uptime for the slave

 

Please compare the uptimes for both the comamnds & update the thread with your observation.

The VC shows that the slaves have stayed up. The only place where they are
reported down is in the Airwave server.

--


___________________________________

Anaheim Union High School District
E-mail Confidentiality Notice

This e-mail communication and any attachments, including documents, files,
or previous e-mail messages, constitute electronic communications within
the scope of the Electronic Communications Privacy Act, 18 U.S.C. § 2510 et
seq. This e-mail communication may contain non-public, confidential or
legally privileged information intended for the sole use of the designated
recipient(s). The unauthorized and intentional interception, use, copy or
disclosure of such information, or attempt to do so, is strictly prohibited
and may be unlawful under applicable laws. 18 U.S.C. § 2511. If you have
received this e-mail communication in error, please immediately notify the
sender by return e-mail and delete the original e-mail from your system.

Hi Matthew,

 

Do you mean the slave uptime seen on VC is same as the uptime seen from individual slave ?

Correct, both the slave and the VC say that they have been up the whole
time.

--


___________________________________

Anaheim Union High School District
E-mail Confidentiality Notice

This e-mail communication and any attachments, including documents, files,
or previous e-mail messages, constitute electronic communications within
the scope of the Electronic Communications Privacy Act, 18 U.S.C. § 2510 et
seq. This e-mail communication may contain non-public, confidential or
legally privileged information intended for the sole use of the designated
recipient(s). The unauthorized and intentional interception, use, copy or
disclosure of such information, or attempt to do so, is strictly prohibited
and may be unlawful under applicable laws. 18 U.S.C. § 2511. If you have
received this e-mail communication in error, please immediately notify the
sender by return e-mail and delete the original e-mail from your system.

Hi Matthew,

 

If the uptime seen for slaves on  VC & slaves themselves are similar, then it could be some communication issue between VC & AIrwave.

 

VC reports the count/uptime of slaves to Airwave.

 

Please open a TAC case for further investigation.

Mathew,

 

In Aiwave only slave IAPs status showing down though they are physically UP or even VCs shows status down sometime and slave IAPs UP?

 

In slave IAPs APs/Device>monitor page device event section, with what message it is marking as down?

 

Is this newly added cluster to Airwave, does VC or any slave IAP in this new cluster is part of any old cluster earlier in Airwave?

 

Regards,

Pavan