Wireless Access

Hi folks,

I have a problem after migration an A6000 in master mode, with nearly 300 APs mixing AP61 and AP125.

Migration consist in pass from 3.3.2.13 to 6.1.3.6. Upgrade was realize with an upgrade path to 3.3.2.13 to 3.4.4.1 to 6.1.3.6

After migration, i have 170 Aps up and 150 down. Ap up are a mix of AP61 and AP125, same for those are down...

All ap down retained old ip address of master controller... I try to clear Ap down in database, and poweroff /poweron Ap, but no result...

I have no idea... Please help :)

Thanks a lot.

Regards

I would go to one of those APS and console it to see whts the issue...

If APS are down and you cannot do anything... go to the AP 125 which is down console it and see if the values of the master ip address server ip address,  ip and default gateway everything is correct....

Are those Campus APs?

How do they find the master? do they get DHCP and find it with DNS? or you statically assigned it?

If you statically assigned the values, you can always console the aruba 125 and can correct it via console.  Now check one of those aps 125 through console... probably what happened to one of those happened to all the otheres

Cheers

Carlos

Hi and thanks for your answer.

I am connecting to the AP 125 throught serial port, and see, AP125 is still in 3.3.2.13...

Below log to the master controller : 

Jan 10 10:21:52 sapd[214]: <311013> <ERRS> |AP 00:1a:1e:c9:7e:c8@172.20.xxx.145 sapd| Image upgrade failed; details follow.
Jan 10 10:21:52 sapd[214]: <311014> <ERRS> |AP 00:1a:1e:c9:7e:c8@172.20.xxx.145 sapd| ERROR: File size exceeded 5242880 bytes
Jan 10 10:21:52 sapd[214]: <311014> <ERRS> |AP 00:1a:1e:c9:7e:c8@1xxxx..xxx.145 sapd| Local write failed: Success.
Jan 10 10:21:52 sapd[214]: <311014> <ERRS> |AP 00:1a:1e:c9:7e:c8@1xxx..xxx.145 sapd| ncftpget mips64.ari: no space.
Jan 10 10:21:52 sapd[214]: <311014> <ERRS> |AP 00:1a:1e:c9:7e:c8@1xxx...xxx.145 sapd| FTP Failed: 3. Trying TFTP.
Jan 10 10:21:52 sapd[214]: <311014> <ERRS> |AP 00:1a:1e:c9:7e:c8@1xxx...xxx.145 sapd| transfer timed out
Jan 10 10:21:52 sapd[214]: <311014> <ERRS> |AP 00:1a:1e:c9:7e:c8@1xxx...xxx.145 sapd| Flash Image upgrade failed: 255
Jan 10 10:21:53 sapd[214]: <311002> <WARN> |AP 00:1a:1e:c9:7e:c8@1xxx..x.xxx.145 sapd| Rebooting: SAPD: Reboot after image upgrade
Jan 10 10:21:53 nanny[196]: <303086> <ERRS> |AP 00:1a:1e:c9:7e:c8@1xxx..x.xxx145 nanny| Process Manager (nanny) shutting down - AP will reboot!

So i try to make a purgeenv, saveenv and reset, but no improvment ...

An idea ?

many thanks.

Is it in the same subnet or there are Firewall between controller and APs.

Please check if FW allow port TFTP.

It would be nice if you could show the output of what you see on the console... I want to see if i can see an error or where is failing

You can try doing

clearos

purgeenv

saveenv

reset

This will delete the OS and it will come with the OS that originally came when it was shipped...

Hi folks,

Thanks for reply, but it was a software version. 

I have validate upgrade process with ARUBA, in order to be sure to pass A6000 in 3.3.2.13 to 6.1.3.6.

Aruba tell me i have to pass 3.3.2.13 to 3.4.4.1 and pass 3.4.4.1 to 6.1.3.6.

But in this migration i have about 300 Aps, and 100 Aps stay in down status ... They don't flash and stay in 3.3.2.13...

So to have Ap up, i have to downgrade my A6000, in 5.0.4.10, to have all my AP flash in 5.0.4.10, and once it was ok, upgrade A6000 in 6.1.3.6, to reflash my AP in 6.1.3.6.

So to recover Ap down i have to pass controller in 5.0.4.10, and my goes up... 

Once in 5.0.4.10 upgrade in 6.1.3.6 and all works ...

Hope it help you in futur deployments :)

Regards ;)

thanks for info, i went through a similar proces recently from 3x, via 5.x to 6.2, reading this im happy i took that extra step.