June 18, 2010 at 8:57 pm
WayneS (6/18/2010)
Just my luck.....Wednesday, I updated a test box to SQL 2008. Pretty old (> 9 years)...
Last night, the drive controller failed spectacularly (sparks, smoke, everything), completely corrupting the HDDs connected to it. Totally unsalvageable.
Sure hope it's not related...
Hopefully it's not - we just put up our new servers on W2008 R2 and SQL 2008.
I did have my dev machine die a couple months ago - both internal drives and the USB drive.
I suspected a power surge, but our maintenance people said no way.
I'll just keep pretending I don't see my monitor flicker numerous time throughout the day.
Greg E
June 18, 2010 at 9:11 pm
Greg Edwards-268690 (6/18/2010)
WayneS (6/18/2010)
Just my luck.....Wednesday, I updated a test box to SQL 2008. Pretty old (> 9 years)...
Last night, the drive controller failed spectacularly (sparks, smoke, everything), completely corrupting the HDDs connected to it. Totally unsalvageable.
Sure hope it's not related...
Hopefully it's not - we just put up our new servers on W2008 R2 and SQL 2008.
I did have my dev machine die a couple months ago - both internal drives and the USB drive.
I suspected a power surge, but our maintenance people said no way.
I'll just keep pretending I don't see my monitor flicker numerous time throughout the day.
Greg E
Heh... ohhhhh... so that's what's been happening to my desktop system at work. 😀
--Jeff Moden
Change is inevitable... Change for the better is not.
June 18, 2010 at 11:06 pm
WayneS (6/18/2010)
Just my luck.....Wednesday, I updated a test box to SQL 2008. Pretty old (> 9 years)...
Last night, the drive controller failed spectacularly (sparks, smoke, everything), completely corrupting the HDDs connected to it. Totally unsalvageable.
Sure hope it's not related...
Wouldn't that be irony and cool:-D
I hate it when that happens. Had one do that to me during an upgrade. Spent the next 24 hours salvaging the wreck. The entire business was down during that time.
Jason...AKA CirqueDeSQLeil
_______________________________________________
I have given a name to my pain...MCM SQL Server, MVP
SQL RNNR
Posting Performance Based Questions - Gail Shaw[/url]
Learn Extended Events
June 18, 2010 at 11:09 pm
Jeff Moden (6/18/2010)
Greg Edwards-268690 (6/18/2010)
WayneS (6/18/2010)
Just my luck.....Wednesday, I updated a test box to SQL 2008. Pretty old (> 9 years)...
Last night, the drive controller failed spectacularly (sparks, smoke, everything), completely corrupting the HDDs connected to it. Totally unsalvageable.
Sure hope it's not related...
Hopefully it's not - we just put up our new servers on W2008 R2 and SQL 2008.
I did have my dev machine die a couple months ago - both internal drives and the USB drive.
I suspected a power surge, but our maintenance people said no way.
I'll just keep pretending I don't see my monitor flicker numerous time throughout the day.
Greg E
Heh... ohhhhh... so that's what's been happening to my desktop system at work. 😀
And just about anything I plug into a wall down here in LV. Dirty dirty electricity. When you have to pay in gold bullion, you would expect pristine electricity.
Jason...AKA CirqueDeSQLeil
_______________________________________________
I have given a name to my pain...MCM SQL Server, MVP
SQL RNNR
Posting Performance Based Questions - Gail Shaw[/url]
Learn Extended Events
June 19, 2010 at 3:38 am
CirquedeSQLeil (6/18/2010)
And just about anything I plug into a wall down here in LV. Dirty dirty electricity. When you have to pay in gold bullion, you would expect pristine electricity.
I used to lose a hard drive or 2 every winter from dirty power, until I invested in some heavy-duty UPSs for all electronics. You should hear these things complaining during the winter months. Either it's spiking, or the frequency has shifted slightly or the overall voltage is too low.
Since buying them I've only lost one hard drive and I can work through most power failures without interruption.
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
June 19, 2010 at 11:29 am
Amazing how some people think that it is pessimistic to think that most developers understand the issues of using the NOLOCK query hint. After many of the discussions here I simply find it pragmatic.
June 19, 2010 at 12:19 pm
Talking about any thread in particular?
edit: Nevermind. Found it and getting on my soapbox.
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
June 19, 2010 at 12:22 pm
Jeff/Lynn, I think this one's for you. http://www.sqlservercentral.com/Forums/Topic939955-169-1.aspx
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
June 20, 2010 at 6:47 am
Steve Jones - Editor (6/10/2010)
A nice blog, in the spirit of Jeff: Killer I/O
I rate it pretty low. It's exaggerated, and risks misleading people into assuming that IO volume is the only thing that matters and ignoring overall design issues. In most cases what drives performance is overall system design, and it's easy to screw that up without doing excess IO (it must be very easy, or else it would be done less often).
Not every database is large, and I've known applications that were designed to ensure that the data working set could be ram-resident, so disc IO occured only for log writes, lazy writes, and extremely rare reads outside the working set (except when things had been ballsed up so that it ended up thrashing because intermediate data exceeded available RAM, but that's not the sort of IO that the blog is referring to, as is clearly demonstrated by the analogy referenced which doesn't cover that at all). Even with large databases good system design can often achieve the sort of small working set that renders IO volume a non-problem. But these databases can still have enormous performance problems caused by bad application and query design (even when they don't suffer memory thrashing) and those problems are not going to be solved by worrying about database IO because that is already negligible.
So maybe the message needs qualifying a bit, so that it says in large databases the main performance killer is often that too much unneeded IO is done.
Tom
June 20, 2010 at 7:15 am
Lynn Pettis (6/11/2010)
Anybody out there know much about this individual, scott.pletcher?I know we have several MVP's here on SSC, but most of you don't advertise it, and perhaps its just me but the tone in his responses seems to lean to the arrogant side a tad.
I know nothing about him except what I've seen here, but on what I've seen for example in http://www.sqlservercentral.com/Forums/Topic936326-338-1.aspx I reckon "a tad" is a gross understatement. I can understand people getting sharp and hard when someone is asserting something that is disputable (although even then one can dispute politely), but I don't think anyone had in that thread.
Tom
June 20, 2010 at 8:46 am
Lynn Pettis (6/19/2010)
Amazing how some people think that it is pessimistic to think that most developers understand the issues of using the NOLOCK query hint. After many of the discussions here I simply find it pragmatic.
Well, I think I'm optimistic and I think that most developers who use NOLOCK understand the issues. I don't understand what can be pssimistic about thinking that?
But I'm pessimistic about people's understanding of SET TRANSACTION ISOLATION LEVEL. Most people don';t understand it at all. Some people use SERIALIZABLE because it's safe and they haven't a clue about the potential response time and concurrency impacts and have no idea whether they need it or not, others will use the default READ COMMITTED when a quick examination of the system reveals either that REPEATABLE READ is needed or that only SERIALIZABLE is acceptable, yet others will try to use the READPAST table hint with an isolation level that doesn't permit it, some people use lots of NOLOCK hints when what they really need is READ UNCOMMITED isolation level, and so on.
Tom
June 20, 2010 at 10:08 am
CirquedeSQLeil (6/15/2010)
Gianluca Sartori (6/15/2010)
Paul White NZ (6/15/2010)
Jan Van der Eecken (6/15/2010)
Paul, Good Luck to New Zealand in their game against Slovakia in less than 4 hours from now.Thanks Jan, I think we're just glad to be part of it. I'll be happy if we score a goal at some stage - I think winning any games might be a bit beyond us, but you never know :-):-)
Don't worry, you'll be playing against Italy soon 😛
Sounds like another 1-1 tie in the making.
good prediction Jason
-------------------------------------------------------------
"It takes 15 minutes to learn the game and a lifetime to master"
"Share your knowledge. It's a way to achieve immortality."
June 20, 2010 at 10:21 am
D.Oc (6/20/2010)
CirquedeSQLeil (6/15/2010)
Gianluca Sartori (6/15/2010)
Paul White NZ (6/15/2010)
Jan Van der Eecken (6/15/2010)
Paul, Good Luck to New Zealand in their game against Slovakia in less than 4 hours from now.Thanks Jan, I think we're just glad to be part of it. I'll be happy if we score a goal at some stage - I think winning any games might be a bit beyond us, but you never know :-):-)
Don't worry, you'll be playing against Italy soon 😛
Sounds like another 1-1 tie in the making.
good prediction Jason
Haha - maybe I should have made a few bets this last week.:-D
Jason...AKA CirqueDeSQLeil
_______________________________________________
I have given a name to my pain...MCM SQL Server, MVP
SQL RNNR
Posting Performance Based Questions - Gail Shaw[/url]
Learn Extended Events
June 21, 2010 at 1:24 am
CirquedeSQLeil (6/20/2010)
.....
Haha - maybe I should have made a few bets this last week.:-D
Horrible match. Looks like the "world champions" have disappeared.
Just a few notes:
1. Killen should be named "Killer". His elbow hit more than one face yesterday.
2. NZ scored in offside
3. Italy's penalty was inexistant
4. Please, somebody stop those vuvuzelas!!!!!!
Congrats to all Kiwis!
-- Gianluca Sartori
June 21, 2010 at 2:45 am
Gianluca Sartori (6/21/2010)
CirquedeSQLeil (6/20/2010)
.....
Haha - maybe I should have made a few bets this last week.:-D
Horrible match. Looks like the "world champions" have disappeared.
Just a few notes:
1. Killen should be named "Killer". His elbow hit more than one face yesterday.
2. NZ scored in offside
3. Italy's penalty was inexistant
4. Please, somebody stop those vuvuzelas!!!!!!
Congrats to all Kiwis!
I didn't see the whole NZ / Italy match, just the highlights, but it was totally riveting, even to a footy pleb. A joy to watch, especially the spectacular performance from NZ.
Apparently some idiot fan found his way into the England changing rooms after their match, claiming he was looking for the toilet. Doesn't seem unreasonable to me - after all, he was simply following 11 other people who'd spent the last 90 minutes looking like they were, too. Can we borrow Ricki Herbert, please?
For fast, accurate and documented assistance in answering your questions, please read this article.
Understanding and using APPLY, (I) and (II) Paul White
Hidden RBAR: Triangular Joins / The "Numbers" or "Tally" Table: What it is and how it replaces a loop Jeff Moden
Viewing 15 posts - 15,871 through 15,885 (of 66,688 total)
You must be logged in to reply to this topic. Login to reply