Compatibility Issues with Arcserve UDP Agent (Linux)

Consider the following third-party software limitations that can affect the performance of Arcserve UDP Agent (Linux):

When you perform an instant BMR for RHEL 5.0 to 5.2 nodes and the firewall of the nodes is enabled by default, the instant BMR job may complete with poor performance because of system firewall issues.
The non-server edition Windows share cannot be used as a backup destination.
The backup job fails if the backup destination is an NFS server version 4 and the host name of the NFS server is not registered to DNS.
Symptom

You have provided an NFS server version 4 as your backup destination and the host name of the NFS server is not registered to DNS. When you run the backup job, the backup job fails.

Solution

Register the host name to DNS.
PXE boot fails on Citrix Xen virtual machine (VM). To avoid this problem, upgrade to gpxe version 1.0.1.
Some versions of Data Domain NAS do not support the file locking mechanism of NFS. Such NFS shares cannot be used as a backup destination.
Symptom

You have provided some version of Data Domain NAS using NFS as your backup destination. When you are performing a backup to that backup destination, the backup fails. The reason is that some versions of Data Domain NAS do not support the file locking mechanism of NFS. Such NFS shares cannot be used as a backup destination.

Solution

To avoid this problem, use CIFS as your backup destination. If the backup job still fails when you provide CIFS as the backup destination, follow these steps:
1. On the node that you want to back up, ensure no CIFS share is mounted.
2. Open the Command Prompt window on the node and enter the following command:
```
 echo "0"  >  /proc/fs/cifs/LinuxExtensionsEnabled
```
You can now submit the backup job for the node.
When you back up the RHEL 5.1 node, the setroubleshoot system service uses the system memory for the backup process.
Symptom

When you back up the RHEL 5.1 node, the setroubleshoot system service uses the system memory for the backup process. If you perform multiple backup jobs, the setroubleshoot service generates multiple alerts in the /var/log/messages file and it uses all the system memory. After many backups, the backup node memory is full and it does not have any free memory.

Solution

To avoid this problem, stop the setroubleshoot system service on the backup node using the following command:
```
service setroubleshoot stop
```