sqlcrossjoin's profile picture. The Eternal Gospel cross join'd with a 20+ year old product from Microsoft.

Aaron Morelli

@sqlcrossjoin

The Eternal Gospel cross join'd with a 20+ year old product from Microsoft.

Aaron Morelli reposted

the korean government lost all of its data, because it put all of its servers in one physical location with no backups

airkatakana's tweet image. the korean government lost all of its data, because it put all of its servers in one physical location with no backups

the south korean government placed all of the servers for hundreds of critical government services in one physical location a single battery setting on fire brought all of them down. it has been a week and only 15% of services have been restored

airkatakana's tweet image. the south korean government placed all of the servers for hundreds of critical government services in one physical location 

a single battery setting on fire brought all of them down. it has been a week and only 15% of services have been restored


Lol (from Pg docs)

sqlcrossjoin's tweet image. Lol (from Pg docs)

Aaron Morelli reposted

Abortion is murdering a real person. Human life begins at conception— even the "day after pill" murders a real person. I know the Lord offers tender grace for all who have participated in an abortion and repent of it. We must be clear on how serious abortion is. #Pro-Life


High-end DBMS's are like Medieval Indulgences. Open-source DBMS's can scale, but more attention is needed to design & code quality. E.g. Oracle and SQL Server's fancy perf features allow sloppier design, but you pay for that "forgiveness" up front.


e_waitPortOpen seems more complex; I've seen a number of cases where PortOpen waits are either producer-side or something else entirely. It'd be nice to know I could trust which branch a task is in if it is in PortClose or SyncCons. Thanks! 3/3


1 key difference between the 2, though: PortClose waits always seem to be blocked behind a task on the other side of the exchange. SyncCons always seem to be blocked behind a task on the SAME side of the exchange. I'm primarily looking to confirm b/c 2/3


Today's task: grok AWS IAM. Been excited to dive into this lake for a long time.


I just typed a recursive query...without googling for syntax reminders. :-/


Stepping off of Twitter for a few weeks. (New baby on the way). See y'all on the other side.


@AdamMachanic I think the right logic is "WHEN page = 3 OR (page - 1) % 511232 = 0". Similar for BCM/DCM. 3/3


@AdamMachanic for a larger DB, running DBCC PAGE on a multiple of 511233 often gives a page type (m_type) of 0. 2/


@AdamMachanic Adam, sp_WhoIsactive uses "WHEN x.page_no = 3 OR _no % 511233 = 0 THEN 'SGAM'", but 1/


[blog] (updated) V6 of my trace flag repository: sqlcrossjoin.wordpress.com/2013/10/28/a-t…


Aaron Morelli reposted

With how much love Jesus looks at us! With how much love He heals our sinful heart! Our sins never scare Him.


Turing Test variation: Can you tell when a co-worker is "talking" to an automated customer service "rep"?


"This website's certificate has known Mafia ties. Are you sure you want to continue?"


Anybody remember Kane from Command & Conquer? I'm pretty sure he took on the pseudonym "Joe Celko"


United States Trends

Loading...

Something went wrong.


Something went wrong.