List:General Discussion« Previous MessageNext Message »
From:DebugasRu Date:February 24 2005 5:48pm
Subject:Re[2]: wikipedia down, slashdot covering, mysql mentioned
View as plain text  
> At least this mailing list has progressed beyond "Why didn't they have
> a UPS?", I suppose. :-)
Yes that was my primary intend :)
I'd really like people to care more about their data safety and more
awareness about the problems involved.

>> That didn't get rid of the race condition, it just moved it. So now you
>> write the transaction, and you read it off, and then you write that it
>> was OK -- but what if that write gets interrupted?
I simply will not delete the old original data (data state of my DB before
transaction was started) that i keep just in case i will need to roll back.

>> 
>> So you write the transaction, and you read it off, and then you write
>> that it was OK, and then you read that off, and then you write that the
>> OK was OK. No, that could get interrupted to...

I don't see why i need to care to write that "Ok was Ok". I just go
and check if "it was Ok with original data". I don't see why i would
need to have additional log file on data log file.


>> Meanwhile, it turns out that you've been reading back from cache instead
>> of platters all along. Or perhaps it turns out that the battery-backed
>> cache in the disk controller has a dead battery, or simply doesn't work.
That's completely different problem - it is a problem of how the
interface between hard drive and OS software is defined. Is there a
function in this interface that will return physically written data
and not cashed one. At this point no DBMS will save you indeed.

>> I think you've confused an RDBMS with a system administrator. Next
>> thing you know you'll be complaining that MySQL isn't ACID-compliant
>> because it can't survive a fire.
if DB is resident across several servers there should be protection
from one server permanently going off line (kind of fire isn't it ?)

GW> Exactly.  No ACID database can ensure integerity in such a situation.
GW> Postgres, Oracle, or any other transactional DB would have suffered the 
GW> same fate in these two cases (LiveJournal, Wiki).
I can't talk about how Oracle deals with such problems because i have no access to its
source code
that's where mysql has a huge advantage! :)



Thread
wikipedia down, slashdot covering, mysql mentionedScott Haneda22 Feb
  • Re: wikipedia down, slashdot covering, mysql mentionedDebugasRu22 Feb
  • Re: wikipedia down, slashdot covering, mysql mentionedPeter Wilm22 Feb
  • Re: wikipedia down, slashdot covering, mysql mentionedDaniel Kasak22 Feb
    • Re: wikipedia down, slashdot covering, mysql mentionedGreg Whalin22 Feb
      • Re: wikipedia down, slashdot covering, mysql mentionedDaniel Kasak22 Feb
        • Re: wikipedia down, slashdot covering, mysql mentionedGreg Whalin23 Feb
        • Re: wikipedia down, slashdot covering, mysql mentionedDebugasRu24 Feb
          • Re: wikipedia down, slashdot covering, mysql mentionedRich Lafferty24 Feb
            • Re: wikipedia down, slashdot covering, mysql mentionedGreg Whalin24 Feb
              • Re: wikipedia down, slashdot covering, mysql mentionedRich Lafferty24 Feb
                • Re: wikipedia down, slashdot covering, mysql mentionedGreg Whalin24 Feb
              • Re[2]: wikipedia down, slashdot covering, mysql mentionedDebugasRu24 Feb
Re: wikipedia down, slashdot covering, mysql mentionedHeikki Tuuri22 Feb
  • Re: wikipedia down, slashdot covering, mysql mentionedGreg Whalin22 Feb
  • Re: wikipedia down, slashdot covering, mysql mentionedJochem van Dieten23 Feb
Re: wikipedia down, slashdot covering, mysql mentionedHeikki Tuuri23 Feb
Re: wikipedia down, slashdot covering, mysql mentionedHeikki Tuuri23 Feb
Re: wikipedia down, slashdot covering, mysql mentionedHeikki Tuuri23 Feb