On Thu, 7 Jun 2001, Nelson Gomes wrote:
> On a cluster, is it possible to have several MySQL instances (one per node)
> each having several mysqld processes, working concurrently on NAS (network
> storage) device, that is mounted on every node, and accessed through a
> shared filesystem ??
>
> How does mysql controlles the access to database files? Through file system
> calls ?
>
> Does anyone used mysql on a cluster ? Suggestions ? Anyone ?
For some reason, the list goes very quiet when somebody mentions clusters.
Dunno what's going on in that respect, but it made me start thinking for
myself.
I am working on designing a multiple server cluster, using SQL Relay and
the mon/fake package. The idea is that the cluster will always have one
master server (which is the drawback; it can only have one master) and, if
possible, one or more slave servers. The SQL Relay service will be
monitored and duplicated, so will the master server. The 'other' server,
ie. the hot backup server, will run a slave db server on another service
ip address, so that you won't waste any hardware resources.
The design stage is almost complete; at the time I am playing around with
handling the hot backup and replication issues. If anybody is interested
(or if anybody has seen a proper solution, so that I don't have to invent
the wheel again), please don't hesitate to contact me. I will, of course,
post in entirety when this little project of mine is finished.
Sincerely,
Jan
--
Mr. Jan-Aage Bruvoll IT Project Manager
20 Min Holding, Thurgauerstrasse 40, CH-8050 Zurich
Office/fax: +44 2072408283 Mobile: +44 7740291600
---------------------------------------------------------------------
Before posting, please check:
http://www.mysql.com/manual.php (the manual)
http://lists.mysql.com/ (the list archive)
To request this thread, e-mail <[EMAIL PROTECTED]>
To unsubscribe, e-mail <[EMAIL PROTECTED]>
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
