|
When
your server goes down, do you have a disaster plan? Most companies
don't. There are some companies that can afford to be down for a day or
two while replacement parts are ordered and techies patch the system
back together. But if you are like most companies, and can't
afford this down time, you should think about this subject now while you
have the luxury of time.
With that said, here's what we
have to offer you to keep you one step ahead of disaster. The program
has two elements: prevention and recovery.
1. PREVENTION OF DISASTER
Prevention by monitoring your file systems. For several years, we have
been offering our DUKE program, which checks your Unix file systems each
night, compiles a report of it's findings, and sends the results to our
computer. Our programmers check the reports each morning, alerting us to
any obvious problems within your system.
Prevention by monitoring your
site. We offer a preventative program that will help to avert disaster
caused by hardware problems and by inadequate procedures at your site.
This is an annual on-site inspection which covers many important points.
The intimate
details are listed on this link.
2. RECOVERY FROM DISASTER
Disaster recovery plan. When you crash (notice I didn't say "if"),
you want to be up and running again in the quickest time possible. If
you have a functioning tape drive, good backup tapes, a listing of part
numbers for reordering, and good current sets of root/boot floppies or
boot CD, then we can get you up and running in the shortest time
possible with minimum expense. If any of these elements are missing and
we have to restore from scratch, the cost to restore, not to mention
employee down time, will skyrocket. As part of this package, we will
discuss and write up a plan with you for recovery, making
recommendations specific to your site.
DETAILS
OF THE ANNUAL SITE VISIT FOR DISASTER PLANNING
As part
of this program, we will:
Check the condition of your
Unix hardware: make sure all fans are working, clean out floppy drives,
CD-ROM drive, and tape drive. We have seen several systems fail this
year, destroying the hard drives because a ten dollar case fan seized
up.
Make current root/boot
floppies or boot CD, an absolutely necessary step for emergency
recovery, and test the floppies by booting up under them. Floppies go
bad over time and system information changes, so root/boot floppies
should be updated annually. While the system is up under the emergency
floppies, we will:
Test the tape drive to make
sure it will pull a file off of the tape. Many times people religiously
make backups, only to discover at a critical time that the backups are
no good. The only true way to confirm the integrity of your backups is
to pull data off of them. Also, we will check the condition and age of
your backup tapes.
Make sure your backup software
is at the latest version. If not, we will offer to do an upgrade.
Make sure all of your media is
accessible and in order and that you have all licenses, install media,
and device driver disks, and that they are current.
Check file systems for
excessive junk files taking up space unnecessarily, check log files
carefully.
Update hardware parts list, so
that reordering parts will be quicker and more efficient.
Create or update your site map
so that we can better support you when workstation specific problems
arise. The site map gives us a chart of all port names for printers,
modems, terminals, hubs, and port servers on your system, as well as IP
addresses for any component on the network. This gives us the
information we need when we receive a support request without having to
barrage you with questions to obtain this data. |
COST
The cost of the DUKE program
which does a daily file system report to us is $400 per year plus a one
time installation fee.
The cost of performing the on-site disaster preventative monitoring
will be time and materials, by sending a programmer to your site. Since
every one of our customers' sites differ, we can not set a price for the
package. It should take from several hours to one full day to complete,
less if all is well and if we have your full cooperation when it is time
to reboot the system. In addition to the labor charges, this analysis
will cause a slight disruption to your site since we will have to reboot
several times to test the root/boot floppies, test the UPS software,
open the case to clean and catalog parts, and possibly other reasons.
Total down time should not exceed one hour. However, compare this
scheduled hour to several days down time if your system crashes without
preparation.
Don't delay; schedule a site
inspection soon. Our recent experience with several customer disasters
has prompted us to take these measures to protect our customer base with
this program.
Print out all system specific specs. Having this information will
maximize recovery speed in cases where a component of the system fails
not requiring a complete system reinstall. This list includes kernel
tuning parameters, printer and modem configurations, SCSI settings, and
other hardware parameters.
Test your UPS (uninterruptible
power supply) to make sure it is still alive. After several years, the
batteries in a UPS will die, but you may not know it unless you test it.
We will also test your UPS software and make sure it is functioning
properly. This type of software shuts down the Unix system gracefully
during a period of prolonged power outage and reports any surges, sags,
or spikes in power.
Test your modem
connection, vital for support.
Check your system's security. We will advise you on security, and
address any of your concerns. We will change passwords if you desire. We
can also add a dialup password that pertains only to connections made
through the modem. Thus, if someone has the root password, he must
also know the dialup password to get in over the modem. We can
optionally create a logging function to record all users who log into
your system and what port they enter from.
Optionally register all unregistered software on your Unix system to
remove the annoying nag messages about unregistered software.
Give
you a report of our findings and make recommendations for upgrades and
enhancements that will be beneficial to your organization.
|
