Author: Bobby Ang
Date: 19:19:56 06/23/01
Go up one level in this thread
Here is the official reply from Darooha:
Dear Chuck:
What we wrote in our announcements about this crisis are all
absolutely accurate. Here are a few more details.
We were making regular backps, but not of everything. For example the
huge games database is not being regularly backed up. Instead we were
depending on the reliability of our expensive RAID system. (This
will, of course, be corrected.) The rough chronology is this:
* Our expensive RAID system died.
* We immediately restarted the system using the backups.
Unfortunately, this did not include everything. For example, we
were missing all the games, including adjourned games and
correspondence games.
* We managed (by using fail safe mode) to copy the data from the
RAID system onto a Seagate disk.
* We then mounted the Seagate disk in our main server, and could not
get it to work.
* We discovered that the disk itself was not functioning. We
eventually resorted to a data recovery lab, which successfully
grabbed an image of the disk. After more work we were able to
construct the entire file system from the image.
* We've started to restore all the things that were not being
backed up. It will take a few more days before everything is
back in order.
I know it's almost unbelievable, but during a period of about one day
when we were depending on the Seagate drive, it died. If you still
want more details, I'll be glad to discuss this with you. Give me a
call. My number is 412-422-5377.
Daniel Sleator, PhD, President of the Internet Chess Club
Phone: 412-521-5553 (office) 412-422-5377 (home)
Fax: 412-521-5575 (office) 412-422-1164 (home)
Email: sleator@chessclub.com
On June 22, 2001 at 20:11:30, Chuck wrote:
>If you log onto ICC presently you may find the following statement concerning
>disk problems they've had:
>
>***** DISK-PROBLEMS *****
>
>As you may be aware, on June 2 part of ICC's disk system crashed.
>We restored the player data and other operational files that same
>day, but had less luck with the old games, game libraries and
>some other data such as TriviaBot's database. Our in-house
>efforts (working with consultants) to read the bad disk have
>failed, so now we are sending it to a data recovery lab. We are
>hopeful that they will be able to recover all of the data, but
>we probably won't know for about a week.
>
>We are very sorry, of course; this was quite a disaster for us!
>It was very bad luck that not only our RAID array died but that
>while we were repairing it, the other disk that had a copy of
>that data crashed. Now we are setting up a new disk system and
>better backup procedures.
>
>****
>
>Now I wish to make an issue of this statement. I don't have anything against
>ICC, but IMO, this statement is a bunch of hogwash. My problem with it is that
>ICC is not being honest. They should just admit they didn't have a good backup
>in place, take the heat and move on. Don't try to tell us that while you were
>trying to run your backup the backup crashed. This is like telling your teacher
>your dog ate your homework. Even if it were true, no one would believe it.
>
>And if it were true [that they had good regular backups], they could just roll a
>backup from the day before and hardly miss a beat. In such a case the above
>statement would not be necessary.
>
>I don't wish to start a big thread about whether they did or not, I just expect
>they are not being honest and everyone should consider they are probably doing
>things with the money you pay them other than what they say they're doing.
>Personally, I wouldn't pay that annual fee to watch them blow smoke.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.