cpbackup failure

[djspark]

Having a hard time trying to figure this one out, anyone else ever run into this? It's on a FreeBSD 5 box.

srv# /scripts/cpbackup
pkgacct started.
pkgacct version 4.2 - running with uid 0
using time::hires for speedups
Copying Reseller Config...Done
Copying SSL Certificates, CSRS, and Keys.....yoursite.com.crt....yoursite.com.csr....yoursite.com.key....www.yoursite.com.crt....www.yoursite.com.csr....www.yoursite.com.key..Done
Copying Mail files....Done
Copying frontpage files....Done
Copying proftpd file....Done
Entering timeout safety mode
Grabbing mysql dbs............
.........
.........
.........
.........
.........
Done
Grabbing mysql privs...Done
Leaving timeout safety mode
Copying mailman lists....Done
Copying mailman archives....Done
Copying homedir....Done
Copying cpuser file.......Done
Copying crontab file.......Done
Copying quota info.......Done
Storing Subdomains....
Done
Storing Parked Domains....
Done
Storing Addon Domains....
Done
Copying password.......Done
Copying shell.......Done
pkgacctfile is: /home/cpbackuptmp/cpbackup/daily/yoursi.tar.gz
Creating Archive ....Waiting for load to go down to continue
Waiting for load to go down to continue
Waiting for load to go down to continue
Waiting for load to go down to continue
Done

md5sum is: 1047de6b0e702a9a373f6e70949bd4db
Can't call method "login" on an undefined value at /scripts/cpbackup line 405, <PWCACHEENT> line 29.

[djspark]

problem solved, if there is a connection error (my problem was I had a spelling error in teh hostname and dns couldnt find it) it returns this error...

[SoftDux]

I have to say, the message generated is very, very misleading. I had a similar problem recently and looked at every possibility why the server load would suddenly be too high for the backup to run. The funny thing is, the load wasn't very high, peaking at about 2.2 - 3. In fact, two weeks prior to my problem, the server had higher loads and the backup still run.

Funny thing is, now that I see this thread, I did have problems with the backup server, the IP changed, and I didn't know about it. So, in the meantime I was looking for problems on the server load.

Odd though, if you use the wrong username / password, it tells you that it's incorrect. Maybe this error messge needs to be revised so that if the IP / hostname of the FTP destination is incorrect it could rather say something like "Host not found / Connection to FTP host broken / etc"?

[djspark]

I did some reading, and it's the perl ftp modules that spits out this error. So it would be a hard one to trap in the script to produce a better error. None the less, when you receive this error it's almost certain that there is a dns related problem, because if the login was wrong, it does return a proper error message for this.

I have to say, the message generated is very, very misleading. I had a similar problem recently and looked at every possibility why the server load would suddenly be too high for the backup to run. The funny thing is, the load wasn't very high, peaking at about 2.2 - 3. In fact, two weeks prior to my problem, the server had higher loads and the backup still run.

Funny thing is, now that I see this thread, I did have problems with the backup server, the IP changed, and I didn't know about it. So, in the meantime I was looking for problems on the server load.

Odd though, if you use the wrong username / password, it tells you that it's incorrect. Maybe this error messge needs to be revised so that if the IP / hostname of the FTP destination is incorrect it could rather say something like "Host not found / Connection to FTP host broken / etc"?