Cisco AIR-CAP2802I-E-K9 isn't retaining/saving Static IP Configurtion

[Erfan7]

I have AIR-CAP2802I-E-K9 AP and it was working fine, suddenly it was disassociated with WLC, I found its static IP configuration was washed out, I tried to prime it through console and on that time it was associated with controller, as i powered off to install that permanently, its static IP again cleared by itself, I reset the AP through mode button and reprimed, on the same day it was retaining the IP but second day when I powered on it was again washed out, I have also changed its boot partition from part2 to part1 from Uboot mode but still same, please anyone here  help me out.

[marce1000]

 - When the AP boots , just after boot login and issue the command show logging  , you may find extra info's

 M.

[Erfan7]

Dear Marce thanks for reply, logging info as under..

AP4C77.6DED.6A9C#sh logging
System logging:
Jan 14 00:27:47 syslogd started: BusyBox v1.20.2
Jan 14 00:27:47 kernel: printkd started: BusyBox v1.20.2 (2017-03-14 12:51:51 PDT)
Jan 14 00:27:47 kernel: [01/14/2018 00:27:47.0600] Booting Linux on physical CPU 0x0
Jan 14 00:27:47 kernel: [01/14/2018 00:27:47.0600] Linux version 3.14.33 (kalairam@cheetah-build1) (gcc version 4.7.1 (GCC) ) #1 SMP Tue Mar 14 13:29:57 PDT 2017
Jan 14 00:27:47 kernel: [01/14/2018 00:27:47.0600] CPU: ARMv7 Processor [414fc091] revision 1 (ARMv7), cr=10c53c7d
Jan 14 00:27:47 kernel: [01/14/2018 00:27:47.0600] CPU: PIPT / VIPT nonaliasing data cache, VIPT aliasing instruction cache
Jan 14 00:27:47 kernel: [01/14/2018 00:27:47.0600] Machine model: Marvell Armada Barbados-2K
Jan 14 00:27:47 kernel: [01/14/2018 00:27:47.0600] Memory policy: Data cache writealloc
Jan 14 00:27:47 kernel: [01/14/2018 00:27:47.0600] On node 0 totalpages: 262144
Jan 14 00:27:47 kernel: [01/14/2018 00:27:47.0600] free_area_init_node: node 0, pgdat 8097dcc0, node_mem_map bf7fa000
Jan 14 00:27:47 kernel: [01/14/2018 00:27:47.0600] Normal zone: 2048 pages used for memmap
Jan 14 00:27:47 kernel: [01/14/2018 00:27:47.0600] Normal zone: 0 pages reserv--More-- [*01/14/2018 00:28:57.7337] waiting for uplink IP address and reachable default gateway
[*01/14/2018 00:28:57.7337]
[*01/14/2018 00:28:58.5156]
[*01/14/2018 00:28:58.5156] !!!!! {watchdogd} Process sync_log gone for 60s
[*01/14/2018 00:29:02.7346] waiting for uplink IP address and reachable default gateway
[*01/14/2018 00:29:02.7346]
[*01/14/2018 00:29:07.7355] waiting for uplink IP address and reachable default gateway
[*01/14/2018 00:29:07.7355]
[*01/14/2018 00:29:12.7363] waiting for uplink IP address and reachable default gateway
[*01/14/2018 00:29:12.7363]
[*01/14/2018 00:29:17.7372] waiting for uplink IP address and reachable default gateway
[*01/14/2018 00:29:17.7372]
[*01/14/2018 00:29:22.7380] waiting for uplink IP address and reachable default gateway
[*01/14/2018 00:29:22.7381]
[*01/14/2018 00:29:27.7389] waiting for uplink IP address and reachable default gateway
[*01/14/2018 00:29:27.7389]

[patoberli]

Is the LAN cable plugged into the correct port?

It looks like the AP can't reach the gateway. Can you check with another AP if that one works on the same cable? Maybe it's a cabling issue. 

[Erfan7]

patoberli everything is fine, only issue is AP is not saving the IP's, if I prime the AP it will hold the IP's for some time and after few hours or if I power off the AP, all cleared itself.

[marce1000]

  Hm. the issues noted by P. should be addressed and resolved , it looks like the AP is trying to use DHCP which is better then static IP and advisable at all times.

 M.

[Erfan7]

Marce we use static IP in our network, thousands of APs are primed, issue is why AP loses the static IP, even though other things are fine i.e. booting properly that means its flash is working fine and commands are properly executing why AP is not holding the IP's, as I said when I put the the static IP it joins the WLC and gets associated, if switch gets cold restart, AP will no longer hold the IP and gets disassociated.

[patoberli]

Is it just this one AP? Then a part of the flash might be broken, or as said, a cabling issue.

[Erfan7]

Im also suspecting so, and I will send it for RMA...

[marce1000]

                          ...>we use static IP in our network

 On the long run I still advised again it , DHCP makes address management central and more easy, you will have more options such as automatically assigning an address when there is a new AP, also for moving AP to other controller and or new subnet (e.g.)

  M.

[Erfan7]

Thanks Marce and Pat for reply...

[zeros66]

In my experience, set the dynamic IP address for the controller if you set the controller on an AP then, set the static IP for AP except the AP is running the controller on itself.

[Scott Fella]

The issue is in the ap join process. With capwap, if the ap all of a sudden looses connection to the controller, it will reboot and try again with the static, but if it still doesn’t join, the ap will revert to dhcp. So the issue is why did the ap not join back to the controller?  Could be an infrastructure issue, cabling or patch issue or maybe the ap.  Dhcp reservations would fix that is you really need to keep the same IP address.