Routerconfigs Issue

[joltman]

I have migrated a CactiEZ install on a VM to a real server. It's CactiEZ 64 bit now. The migration worked (after a little help) but the routerconfigs plugin isn't working. I'm getting the following in the log when I try to run a manual backup:

05/15/2009 10:46:06 AM - CMDPHP: Poller[0] ERROR: SQL Row Failed!, Error:'1064', SQL:"SELECT * FROM plugin_routerconfigs_backups WHERE id = "
05/15/2009 10:46:04 AM - CMDPHP: Poller[0] ERROR: SQL Row Failed!, Error:'1064', SQL:"SELECT * FROM plugin_routerconfigs_backups WHERE id = "
05/15/2009 10:46:04 AM - CMDPHP: Poller[0] ERROR: SQL Row Failed!, Error:'1064', SQL:"SELECT * FROM plugin_routerconfigs_backups WHERE id = "

After some reading, I'm wondering if I need to import a SQL db? Is that correct? Can I just export it from my old server and import to the new one? If so, what are the commands? Thanks!

Jim

[joltman]

Just as an update, I tried to import the routerconfigs.sql file into the cacti DB. It gave this error:

ERROR 1050 (42S01) at line 2: Table 'plugin_routerconfigs_accounts' already exists

That's telling me that the routerconfigs.sql tables are already in the DB. What else can I do?

Jim

[joltman]

As another update, I looked at the debug info for one of the switches. This is the error:

ChestnutCore#copy run tftp
Address or name of remote host []? 192.168.10.37
Destination filename [chestnutcore-confg]? ChestnutCore
TFTP: error code 0 received - 20581

%Error opening tftp://192.168.10.37/ChestnutCore (Undefined error)
ChestnutCore#

Once I ran:

chmod 777 /var/routerconfigs

The backup ran. However, the configs are assigned these permissions:

[root@rnpadc612 routerconfigs]# ls -la
total 92
drwxrwxrwx 4 apache apache 4096 May 15 11:53 .
drwxr-xr-x 24 root root 4096 Apr 30 09:43 ..
-rw-rw-rw- 1 nobody nobody 65536 May 15 11:53 ChestnutCore
drwxr-xr-x 2 apache apache 4096 May 15 11:28 routers
drwxr-xr-x 2 apache apache 4096 May 15 11:53 switches

I have configured each device to dump into either a router or switch directory and they aren't doing that. What's going on?

[joltman]

Well this sucks...I re-created one of the devices in Cacti Web interface...and ran the backup, and it worked perfectly. That REALLY sucks as I'll have to re-create 20+ devices. Isn't there a better way to do this?

Jim

[rszebin]

i have 130 devices, and tftp just stopped working for some reason. All i did was "chkconfig xinetd off" and "chkconfig xinetd on" - game over tftp

btw, it was working somewhat ok until a week or so, when i wanted to look into automatic back-up, and until this post i thought i did something wrong. i have latest CactiEZ, and i love it, it's an amazing thing, but building it on CentOS? it is exactly what the name says: an OS wort one cent. and that would be a generous offer: it's like win32 - as long as you don't touch it, it's smooth and cool. as soon as you try anything that tampers with "out of the box" set-up - namely here routerconfigs workaround - crashes.
why go Red Crap based when Debian is out there? ubuntu is just as good .... *sigh*

[dieselboy]

I too are having the issue the OP posted with cactiEZ.. and yes why Centos? Ubuntu is more common and would have seemed logical but I am not complaining!

[rszebin]

i went Rancid + cvsweb and never looked back
if u like easy stuff,
http://openmaniak.com/rancid_tutorial.php

Njoy

[dieselboy]

i went Rancid + cvsweb and never looked back
if u like easy stuff,
http://openmaniak.com/rancid_tutorial.php

Njoy

Thanks for the link. I would rather have routerconfigs working.