See the About EZ-Backup page.
See the EZ-Backup Features page.
See the EZ-Backup Fees and Billing page.
See the subscriptions page.
See the mailing list page.
Unfortunately, at this time, we still do not offer the EZ-Backup service to students. There are two issues that we need to resolve before offering the service to students -- billing and support.
As far as support, we need to create a support structure for supporting student use of the service. Currently, the EZ-Backup team consists of three staff who support departmental IT staff who in turn support over 2100 systems. It would not be possible for the staff of three to directly support hundreds or thousands of students, so we need a support structure in place to do this. As far as billing, the billing algorithms for EZ-Backup are different enough from other services that we cannot simply use the same billing mechanism. Other services that are billed through your bursar account have a flat fee per month, whereas EZ-Backup has a variable rate based on how much data we back up per system.
We are also not sure how many students would be interested in the service if we did offer it to them, so we are not sure what priority to put on this effort. If you have any opinions on this, we would appreciate hearing them. Contact us via email at ezbackup@cornell.edu
Yes, we can define a "backup schedule" to suit your needs. There are many options, including choosing the starting time and changing the length of the backup window. If you would like to request a different backup schedule for your computer, email us at ezbackup@cornell.edu. Be sure to include your node name.
There is quite a bit of flexibility here. Here's how it works:
An EZ-Backup "management class" defines things such as how many extra versions of a file are kept, and for how long. This can be specified separately for files that exist on your workstation vs. files that have been deleted from your workstation. Files on a workstation can be mapped to different management classes, allowing you to have different levels of backup (or archive) protection for different types of files. This mapping is done using rules that are placed into the EZ-Backup options file (TSM Backup Preferences or dsm.opt) on the workstation. The rules can have wildcards in them, making it easier to map classes or collections of files with one rule. If you don't map a file to a management class, it is mapped to a default management class.
For backed up data (vs. archived data), the defaults:
If file exists on workstation:
If file no longer exists on workstation:
For archived data:
When a file is archived, by default, it does not expire. However, at the time a file is archived, a management class may be selected which designates a retention period for that file. For files that have a retention period of more than six months, we keep a mirror copy of that file. If the file is deleted from the workstation, the archive data will remain for the duration of the retention period associated with that file.
For departments with quite a few computers, we can set up a departmental "Policy Domain" which has its own private set of "management classes". It is strongly suggested that departments have their own policy domain if they have several computers. This way we can tailor management classes to your department's special requirements.
The list of management classes, and their definitions, can be viewed from the EZ-Backup software program running on your workstation (see Viewing Policy Information). To get a new management class set up, simply tell us what you need and we will set it up for you. Once set up, you can use it however you wish by specifying appropriate rules in the EZ-Backup client options file.
For more in-depth information on management classes, refer to the TSM/ADSM manuals for your platform.
The options are:
Send email to ezbackup@cornell.edu. Include the name of your EZ-Backup system (nodename) and also specify one of the following:
The EZ-Backup software makes distinctions between the "backup/restore" function and the "archive/retrieve" function.
When a file is backed up, it is stored on the EZ-Backup server for a specified number of days and a specified number of versions. For more info, see the version information. Backups are usually incremental, which means files are backed up only if they have changed since the last backup. For more info, see incremental backups. According to Tivoli, the EZ-Backup software vendor:
"Backup: The process of saving the files on your computer to a server. The server is usually in a different location. TSM can retain multiple versions of a file that you backed up to the server. For example, you backed up a file on Monday, updated the file on Tuesday and then backed it up again to the server. TSM retains the backup version you made on Monday and changes it to an inactive version. The file that you backed up to the server on Tuesday becomes the active backup version."
"Restore: The process of copying files that you backed up (backup versions) from the server to your computer. You can restore a backup version to its original location or a different location on your computer."
Archival is not incremental. When you archive a file, it will stay stored in the EZ-Backup server forever (or until the retention period of the associated management class is reached). If you delete the file from your computer, nothing would happen to the archived copy of that file (if it was archived). However, a backup copy of that file would expire 60 days (default) after the file was deleted from your workstation. By default, we maintain two copies of long-term archival data, but you are only charged for one copy. According to Tivoli, the EZ-Backup software vendor:
"Archive: The process of saving your files to a server in objects called active packages. You can assign a description to an archive package. Files that are archived are called archived copies. The server usually retains an archive copy for a long period of time, but keeps only one version on the server."
"Retrieve: The process of getting your archived copies back from the server. You can retrieve them to their original location or to a different location on your computer."
Backups are intended to protect data. Backups are usually automated. Archives are intended for long-term archival of data. Archival is usually done manually.
Yes, there are options to do all of the above. The first option is actually an extreme case of the third option, where the directory tree simply starts at the root level of the volume. For the second option, you can either specify the filename or select files individually from a directory-structured menu of your backup files.
Yes. See the EZ-Backup Publications and Documentation page for a complete list.
Your first backup will be a full backup of all included drives, directories, and files. Subsequent backups will be incremental; that is, only new or changed items will be backed up. (Deleted items are noted, as well.) Initial full backups will take much longer than subsequent incremental backups. Incremental backups usually take only a few minutes.
Yes. It is possible to use EZ-Backup over a modem. However, it is not recommended unless you are using a high-speed modem, and it is an incremental backup, not the initial backup. For more information on incremental backups, see incremental backups.
The EZ-Backup software processes the exclude and include options as follows:
Note: Includes and excludes apply to both manual and scheduled backups.
Note: Exclude.dir cannot be overridden.
For a more in-depth discussion, refer to the vendor's documentation at
12. Does EZ-Backup use file compression?
The EZ-Backup software has both client and server level compression options. At Cornell, we force compression ON at the server, so that the client (you) has no choice but to use compression. We do this to conserve network bandwidth and server storage utilization. If we wanted to, we could have configured the server so that the client could choose to compress data or not, at your choice. This choice would be specified in the "ADSM Preferences" file on your computer. Because we have forced compression ON, there is no need for the COMPRESS option in this file; it is overridden by server-level compression.
By using server-level compression, the data that you back up or archive is first compressed by your workstation before being stored on the EZ-Backup server. Compression efficiency depends quite a bit on the type of data that is being compressed. Executables may be less compressable than documents. (Note that the ADSM Scheduler does report the compression percentage in the summary statistics at the end of the scheduler log.) In general, we tend to see about a 2-1 ratio between what is stored on the workstation and what is stored on our EZ-Backup server.
Related notes:
To display your "Policy Information", i.e. tell how many versions EZ-Backup keeps on the server and for how long they are kept, you can check through the ADSM or TSM Backup client.
Unfortunately, it's not possible to check the storage space used from the client side. Our billing scripts do queries on the nodes each day, drop the five highest amounts (to accommodate temporary spikes), and average the rest over the entire month to come up with the storage amount billed. So the billing summaries are the best indicators of space used.
For related info, see:
While it may be technically possible, it is not a good idea to use more than one computer per EZ-Backup nodename. There are several reasons:
Yes. The EZ-Backup software is not dependent on the IP address of the client system. It uses a "nodename" and "password" scheme to authenticate the client system, not the IP address. Therefore, your computer can be backed up from anywhere, as long as you have a network connection (either over a LAN or a modem) and a valid IP address. (See the modem question for more info on backing up over a modem.)
No, EZ-Backup does not rely on Kerberos. The EZ-Backup software uses a "Kerberos-like", two-way authentication process. There is a unique password that goes with each EZ-Backup node name. Access to stored files is also password protected. Only the EZ-Backup software use this nodename/password scheme. Your Kerberos password is a separate process.
