One graph display data one not

[zahir123]

Hi Guys!
I have been looking for a solution for a while now and i am confused a little bit, so looking for a help here.
I have two Cisco devices, exact same model, exact same configuration. After adding those two devices, one of them (Cisco 1) displays nice grpahs one not (Cisco 2). I can see that there is no RRD file for Cisco 2 device with apache error log:
ERROR: opening '/var/www/cacti/rra/cisco2_32.rrd': No such file or directory

What's more Cisco 2 device has status marked as "down" on device list. Thats probably because cacti error message:
CMDPHP: Poller[0] WARNING: SNMP Get Timeout for Host:'1.1.1.1', and OID:'.1.3.6.1.2.1.1.3.0'

But what is wired that i can do snmpwalk and snmpget directly from cacti server to my Cisco2 machine and it works just fine.

Using cacti 0.8.8a on Centos 2.6.18-308.13.1.el5.
As a poller i use cmd.php. Changing Timeout and max oid per one request doesn't change anything (cacti server and cisco devices are in the same network so latency is minimal).
Changing checking method from SNMP Uptime to UDP Ping create rrd file, but graph is empty anyway. When i run Realitime plugin for that graphs it works just fine. When UDP ping is in use then cacti error log change like that:

12/03/2012 01:55:15 PM - CMDPHP: Poller[0] Host[16] DS[43] WARNING: Result from SNMP not valid. Partial Result: U
12/03/2012 01:55:15 PM - CMDPHP: Poller[0] WARNING: SNMP Get Timeout for Host:'1.1.1.1', and OID:'.1.3.6.1.2.1.31.1.1.1.10.2'


thanks in advance.

[gandalf]

The second host is available under the IP@ of "1.1.1.1"???? Strange
R.

[zahir123]

i just made it up, it doesn't matter, does it?
Anyway, i haven't found the solution yet