[GE users] Does a shadow master really improve reliability

chribo christian.bolliger at id.unizh.ch
Fri Mar 13 17:52:11 GMT 2009


Hi
Our experiences with the shadow master are very good. On the other hand the gridengine is rock stable. The last 4 years we fall back
  we only used the shadow master twice (unplanned, for upgrades it was handy as well). But both cases were caused bye us (sysadmins).

We currently run the BDB (and ldap) on a machine we call spof (single point of failure), the only failure occured when we did not
notice that the filesystem was full.

IF you want to have a setup without single point a possibility would be to have the BDB locally on the master and on the shadow
master and to keep the two db in sync. We plan to use this setting for our next cluster.

Best regards
Christian

ddavies wrote:
> Hi,
> I'm planning to setup BDB shadow master, but I'm not sure if I'm improving or reducing reliability based on the following statement on page 30 of this SGE whitepaper: http://www.sun.com/offers/details/Sun_Grid_Engine_62_install_and_config.html
> 
> "The goal of installing a shadow master is to eliminate a single point of failure in the cluster, but adding a remote Berkeley server for spooling adds an additional single point of failure."
> 
> This sounds like an overall reduction is reliability. Is that correct?
> 
> Regards,
> Dave Davies
> 
> ------------------------------------------------------
> http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=130074
> 
> To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].

------------------------------------------------------
http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=130094

To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].



More information about the gridengine-users mailing list