August 11, 2008 at 7:46 am
I helped out on one of those once, except the tape backup wasn't working.
http://sqlinthewild.co.za/index.php/2008/03/06/on-recoverability/
They're still picking up the pieces of that one.
WOW just read that... did they get it sorted? Thank goodness I have never had this sort of issue. Working with a nice resource-rich infrastructure now though.
Multi-TB's, Clustered SAN, HP CA via fibre to offsite SCF clustered SAN, plus nightly LTO4 and shortly to come, online backups to an offsite nucleear bunker (facility).
So we cool? :unsure:
Adam Zacks-------------------------------------------Be Nice, Or Leave
August 11, 2008 at 7:47 am
Adam Zacks (8/11/2008)
when you change back to 'full' rec-mode from 'simple', should you then bakup the db and log (seperately) or just the db? I have only backed up the db when i have done my 'shrinkage trick' 😉And on the other line, if you want more smilies, then you shall have them :hehe:
Hows it feel to have a cyber-geek-fan? 🙂
You should take a full backup, so that when you restore your crashed database, you can start with the full backup, and apply the subsequent log backups. If you had no full backup after switching from simple, then SQL Server knows that the log files (apart from the active section) will be useless anyway, as you cannot apply them in case of a recovery.
Regards,
Andras
August 11, 2008 at 7:52 am
Ok, so what i have done in the past is ok then (to do a dull db backup after reset of rec-mode).
Started thinking t hat maybe a log backup was needed for some reason.
Thanks. 🙂
Adam Zacks-------------------------------------------Be Nice, Or Leave
August 11, 2008 at 9:21 am
Adam Zacks (8/11/2008)
WOW just read that... did they get it sorted? Thank goodness I have never had this sort of issue.
Nope. DB was completely unrecoverable, despite several hours of trying every trick in the book and some not. They're still rebuilding the 300 GB database. Fortunately, a lot of it was scanned in documents and so could be rescanned.
Working with a nice resource-rich infrastructure now though.
Multi-TB's, Clustered SAN, HP CA via fibre to offsite SCF clustered SAN, plus nightly LTO4 and shortly to come, online backups to an offsite nucleear bunker (facility).
So we cool? :unsure:
Maybe. 😀
The aforementioned disaster was in a resource-rich environment. It was just that several things happened at once that together caused the problem to be a disaster.
The mirrored server had been taken offline the previous day for a SAN upgrade (replacing the drives), and so there was no second hot copy of the database as there would normally be.
The off-site backup storage for this DB would have been ready the next week, as there had been a delay in obtaining new hardware there.
There was a shortage of rack space at DR, which was due to be rectified in a couple of days and hence there was no DR solution.
The primary server was due a SAN upgrade with more disk space once the mirror was back online.
If the glitch had happened a couple days earlier or a couple days later it would have been a non-issue.
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
August 11, 2008 at 10:50 am
Andras Belokosztolszki (8/11/2008)
Adam Zacks (8/11/2008)
Very business oriented Gail?
Umm - surely that's our job?
I'm a DBA.
I'm not paid to solve problems. I'm paid to prevent them.
August 11, 2008 at 10:57 am
Ever thought of expanding that slightly and submitting as an article on here? Seen soooo many posts on forums and usenet where people just fail to grasp the issues and you seem to have a way of expressing things with no bull**** which would be very helpful to those who don't understand the implications
I'm a DBA.
I'm not paid to solve problems. I'm paid to prevent them.
August 11, 2008 at 12:12 pm
andrew gothard (8/11/2008)
Ever thought of expanding that slightly and submitting as an article on here?
Who's that aimed at?
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
August 11, 2008 at 2:41 pm
Um - you Gail. Hope the suggestion's taken in the way it was meant ...... just thought it was nice and to the point
I'm a DBA.
I'm not paid to solve problems. I'm paid to prevent them.
August 11, 2008 at 2:54 pm
andrew gothard (8/11/2008)
Um - you Gail. Hope the suggestion's taken in the way it was meant ...... just thought it was nice and to the point
Wanted to be sure before responding. Are you referring the blog post or one of my other rants on this subject? 😀
If you think it's of value, I'll consider it.
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
August 11, 2008 at 4:40 pm
Was thinking of this one
http://sqlinthewild.co.za/index.php/2008/07/23/recovery-model-and-transaction-logs/
Think we all know how often people seem to get parachuted into 'looking after the database(s)' and over the years I've seen a pretty large number of questions regarding this topic which seems to be poorly understood gven hte critical importance
I'm a DBA.
I'm not paid to solve problems. I'm paid to prevent them.
August 12, 2008 at 1:13 am
andrew gothard (8/11/2008)
Was thinking of this onehttp://sqlinthewild.co.za/index.php/2008/07/23/recovery-model-and-transaction-logs/
Think we all know how often people seem to get parachuted into 'looking after the database(s)' and over the years I've seen a pretty large number of questions regarding this topic which seems to be poorly understood gven hte critical importance
Half the reason I wrote that was so that I wouldn't have to repeat myself every time the question comes up here.
I'll look at turning it into an article. Need to check with Steve to be sure that a blog post turned article is acceptable.
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
August 12, 2008 at 2:11 am
andrew gothard (8/11/2008)
Andras Belokosztolszki (8/11/2008)
Adam Zacks (8/11/2008)
Very business oriented Gail?Umm - surely that's our job?
Ummm well yes, I suppose... Depends what mood I'm in. Getting married next week with a 3.5 week holiday, so right now dont care about anything! 😉
Adam Zacks-------------------------------------------Be Nice, Or Leave
August 12, 2008 at 2:40 am
Adam Zacks (8/12/2008)
Ummm well yes, I suppose... Depends what mood I'm in. Getting married next week with a 3.5 week holiday, so right now dont care about anything! 😉
:hehe: Congratulations. Enjoy your honeymoon.
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
August 12, 2008 at 2:45 am
Thanks Gail. Goint Maritous and Dubai, was hoping for a new HP lappy in Dubai, heard they were cheaper but according to forumns NOT. :crying:
Hey-Ho at least I get to get burnt! :blush:
Adam Zacks-------------------------------------------Be Nice, Or Leave
August 12, 2008 at 7:30 am
Adam Zacks (8/12/2008)
andrew gothard (8/11/2008)
Andras Belokosztolszki (8/11/2008)
Adam Zacks (8/11/2008)
Very business oriented Gail?Umm - surely that's our job?
Ummm well yes, I suppose... Depends what mood I'm in. Getting married next week with a 3.5 week holiday, so right now dont care about anything! 😉
Congratulations mate - hope it goes like a dream
I'm a DBA.
I'm not paid to solve problems. I'm paid to prevent them.
Viewing 15 posts - 16 through 30 (of 39 total)
You must be logged in to reply to this topic. Login to reply