Wireless Access

Reply
SBS
Contributor II

7240 local controllers

Hello,
I'm writing to try and get some additional ideas from the community.

We have (2) 7210 controllers (primary and master) and (3) 7240 local controllers where clients and WAPs (225,135,105) terminate. We got alerts from airwave that masses of WAPs at different locations were going down (I *think* this may have quietly been going on since December and has progressively gotten worse, I don't have solid evidence but a hunch).  I opened a case with TAC in July.   Various things, one person saw high STM process (104% CPU) and HTTPD process too.  One said he saw WAPs had been up 44 days yet the radios had rebooted recently.  I would montior cpuload and it seemed very high.  We were on version 6.4.4.4 and they recommended going to 6.4.4.9.  Completed that upgrade and a few days later.  Only issue had during the upgrade is that it took a couple hours for about the last 100 WAPs to fully register (they kept bootstrapping but eventually everything upgraded and seemed good). Now we are losing console even to local controllers, one is down completely (no ip connectivity, no console), the other pings but no console or ssh (beyond the login and password just sits, never gets you to prompt), the last one pings but no console or ssh either.  TAC had me pull the uplink to the network to see if that was impacting the console.  No change when removed the uplink, still no console.  I ended up hard booting all 3 controllers to restore services.  And here we are 12 hours later and I can't console into one of the controllers.  I can't ssh all the way in to 2 of them.  The HTTPD process seems high and I'm worried.  

 

Examples:

Tasks: 179 total,   2 running, 177 sleeping,   0 stopped,   0 zombie

Cpu(s): 25.9%us, 19.0%sy,  0.0%ni, 52.8%id,  0.0%wa,  0.0%hi,  2.3%si,  0.0%st

Mem:   5172096k total,  3087680k used,  2084416k free,    12928k buffers

Swap:        0k total,        0k used,        0k free,   905088k cached

   PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND

19017 nobody    20   0  292m  24m 7232 S  122  0.5   0:05.00 httpd  (122% CPU??)

3950 root      20   0  711m 479m  77m S   15  9.5 277:45.38 stm

3

 

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND

 3222 nobody    20   0     0    0    0 Z 9999  0.0   0:08.36 httpd <defunct> (9999% cpu and defunct??)

 3824 nobody    20   0  310m  24m 7168 S   58  0.5   0:03.06 httpd

 3747 root      20   0  356m 284m 281m S   34  5.6   2831:21 gsmmgr

 3950 root      20   0  727m 516m  98m S   14 10.2   1543:38 stm

 4111 root      20   0  271m 134m  48m S   11  2.7 102:13.90 arm

 

Thanks for any pointers,

Sarah

 

 

Guru Elite

Re: 7240 local controllers

Taken by itself, the bootstraps could mean there is congestion or you have connectivity issues in your network, in addition to taking and hour to upgrade.

I honestly have never seen a slow console port, but that could be a hardware issue if that was the case. Keep working with tac, because we would be just guessing here. You can PM me the case # and we can see if there is anything else that could be done


Colin Joseph
Aruba Customer Engineering

Looking for an Answer? Search the Community Knowledge Base Here: Community Knowledge Base

Search Airheads
cancel
Showing results for 
Search instead for 
Did you mean: