[OpenAFS] Memory cache question

Martha.Cofran@cexp.com Martha.Cofran@cexp.com
Fri, 8 Nov 2002 10:12:30 -0700


Thanks for the rsync suggestion, I hadn't thought of that.

And yes, I know this is a stupid question based on a stupid requirement.  In
fact, I already have 2 DB and fileservers to handle fail-over.  But, if I
can automate it and it keeps management happy and off my back, I'm all for
it :)

Martha

> -----Original Message-----
> From:	Paul Blackburn [SMTP:mpb@est.ibm.com]
> Sent:	Friday, November 08, 2002 9:44 AM
> To:	Martha.Cofran@cexp.com
> Cc:	openafs-info@openafs.org
> Subject:	Re: [OpenAFS] Memory cache question
> 
> Martha.Cofran@cexp.com wrote:
> 
> >First I want to thank everyone on this list that's helped me in the past.
> >My site went to production last week (finally) and everything is working
> >great.
> >
> >My question:  Is there a way to prevent a file pulled from the AFS server
> to
> >the client from being cached?  I want most stuff to be cached, I just
> want
> >to prevent certain stuff.
> >
> >AFS is serving catalog images to a mission critical on-line ordering
> >web-site.  Speed is essential, so I've got a 500 Mg memory cache on each
> of
> >my web-servers to cache the most frequently used images (done
> automatically
> >at boot so users aren't affected) of the 1Gig that we have.  Management
> >wants to tar up all of the images on AFS every night and copy them to the
> >web servers, so that if AFS disappears, recovery would only consist of
> >untarring the current data and reconfiguring the directory the web server
> >pulls from.  I figured the best way to do this is to run a cron routine
> to
> >tar it to a volume.  Then each client can cp the tar from the AFS
> directory
> >to a local directory.  But, if I do this, will the 1Gig tar file
> completely
> >flush all of the real data out of my memory cache?  Is there a better way
> to
> >accomplish this?
> >
> >Thanks for your thoughts.
> >
> >Martha Cofran
> >_______________________________________________
> >OpenAFS-info mailing list
> >OpenAFS-info@openafs.org
> >https://lists.openafs.org/mailman/listinfo/openafs-info
> >  
> >
> The obvious way to not access the cache, is to access data outside of
> /afs.
> I would look at using the rsync command to maintain a local disk replica 
> copy of
> the whole 1gb of images files. Don't mess around with 500 mb. Disk is
> cheap.
> The rysnc command is "clever" because it only copies what it has to copy.
> If there are no changes, rsync will copy nothing.
> 
> You could run a hourly crontab'ed rsync to keep the local copies
> up-to-date.
> 
> The other comment I would like to make is about "if AFS disappears".
> Really, you can build a robust AFS cell with multiple DB servers and
> multiple fileservers.
> --
> cheers
> paul                            http://acm.org/~mpb
>