GemSource back on-line

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

GemSource back on-line

Dale Henrichs
Early this morning we had more hardware troubles in our data center. we
lost two disk drives on the machine that hosts seaside.gemstone,con
along with a few other applications. Despite using RAID 5 we had quite a
few files corrupted on the server, including the GemStone db files (this
isn't supposed to happen!) .

GemSource hasn't been moved to new hardware in the VMWare data center
yet (although this incident should accelerate things a bit:).

Since we're concerned that we may experience more hardware problems
before being able to replace the hardware we've decided to start making
back ups a little more frequently (once every couple of hours instead of
once a day).

I suppose that there is a silverlining. It looks as though the secondary
repository code for bootstrapping Metacello worked pretty well, since I
didn't see anyone complaining about GemSource being down today ... I
know that I was testing ConfigurationOfSeaside30 loads this morning for
quite a while before noticing that GemSource was down (the IT guys had
noticed almost immediately:).

If any of you did notice problems with bootstrapping Metacello during
the day today I'd appreciate it if you dropped me a line with any
details you may have, since I'm interested in bullet proofing that code
as much as possible...

Dale


_______________________________________________
Pharo-project mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
Reply | Threaded
Open this post in threaded view
|

Re: GemSource back on-line

Schwab,Wilhelm K
"Despite using RAID" - I thought the job of a RAID controller was to ensure that the corrupted data made it onto all available platters :(

Thanks for the work on Metacello!

Bill




________________________________________
From: [hidden email] [[hidden email]] On Behalf Of Dale Henrichs [[hidden email]]
Sent: Wednesday, June 23, 2010 12:24 AM
To: GemStone Seaside beta discussion; [hidden email]
Subject: [Pharo-project] GemSource back on-line

Early this morning we had more hardware troubles in our data center. we
lost two disk drives on the machine that hosts seaside.gemstone,con
along with a few other applications. Despite using RAID 5 we had quite a
few files corrupted on the server, including the GemStone db files (this
isn't supposed to happen!) .

GemSource hasn't been moved to new hardware in the VMWare data center
yet (although this incident should accelerate things a bit:).

Since we're concerned that we may experience more hardware problems
before being able to replace the hardware we've decided to start making
back ups a little more frequently (once every couple of hours instead of
once a day).

I suppose that there is a silverlining. It looks as though the secondary
repository code for bootstrapping Metacello worked pretty well, since I
didn't see anyone complaining about GemSource being down today ... I
know that I was testing ConfigurationOfSeaside30 loads this morning for
quite a while before noticing that GemSource was down (the IT guys had
noticed almost immediately:).

If any of you did notice problems with bootstrapping Metacello during
the day today I'd appreciate it if you dropped me a line with any
details you may have, since I'm interested in bullet proofing that code
as much as possible...

Dale


_______________________________________________
Pharo-project mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

_______________________________________________
Pharo-project mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project