As I may have mentioned before VMWare Data Recovery is not my favorite backup solution. I'm pleased to report that VDR2.0 is a LOT better than 1.x ever was. Since upgrading, my backups haven't failed once. And it even has a built-in email reporting system.
Unfortunately, this mail system isn't too customizable. Since I'd like my subjects to be clear they need extra attention, I made a little fix. This script is started by socat.
Socat handles incoming connections on port 25 for me.
rpm -i http://download.fedora.redhat.com/pub/epel/5/x86_64/socat-220.127.116.11-1.el5.x86_64.rpm
Get my script
cd /usr/local/bin ; wget https://www.geeklab.info/wp-content/uploads/2011/12/smtp-filter.sh
You should put your own mailserver in the script on line 7.
Start it automatically
echo "socat TCP4-LISTEN:25,fork EXEC:/usr/local/bin/smtp-filter.sh &" >> /etc/rc.d/rc.local
Now change your settings
Change the outgoing mailserver to localhost and test it. If everything's ok, you'll get a mail with 1 warning and 2 error-message, as the test-mail doesn't contain the fields that indicate no backups failed.
The iptables firewall does not allow incoming connections to port 25, so you will not become an open relay using this script.
Change VMWare Data Recovery 2.0 report mail is a post from GeekLab.info. You are free to copy materials from GeekLab.info, but you are required to link back to http://www.geeklab.info
VMWare Data Recovery is crap:
- VMWare Data Recovery cannot be monitored and does not send out any warnings.
- Running one instance of VMWare Data Recovery is too heavy for the appliance, causing the backups to be corrupted all the time
- When running multiple instances of VMWare Data Recovery, say one instance per host, the whole thing is still slow as hell
- Every few weeks, one backup gets corrupted. Of course the appliance does not notify you, no, instead you must login to the appliance to check the logs.
- When you find a backup that is corrupted, you need to manually flag the restore point as "marked for deletion" and then start the integrity check.
- When doing the integrity check, better pray that none of the other instances starts using the same nas/fileserver, as it gets so slow that the whole integrity check never gets completed.
- I just tried to split the backups to multiple destinations, so that we can backup each virtual machine to its very own destination, and reduce the shitload of work the integrity check has to do when another backup is broken. Appears that VMWare Data Recovery only supports 2 destination per appliance instance.
I seriously consider VMWare Data Recovery to be the worst backup solution I've ever used.
VMWare Data Recovery is crap is a post from GeekLab.info. You are free to copy materials from GeekLab.info, but you are required to link back to http://www.geeklab.info
If you surf to http://[vdr-host]:5480 you will be asked for a password.
Not very secure. Better make sure the VDR is firewalled.
VMWare Data Recovery default password is a post from GeekLab.info. You are free to copy materials from GeekLab.info, but you are required to link back to http://www.geeklab.info