On Thu, 3 Apr 2008 14:55:48 +0530 "Anand Avati" <avati@xxxxxxxxxxxxx> wrote: > Daniel, > maybe it is just taking long to detect connection failure. Can you > try with 'option transport-timeout 20' (sets response timeout to 20 > seconds) in all your protocol/client and see if you still face the > 'hang' ? I'll give that a go straight away. Is 20 seconds "too long" - i mean, in a production environment (i.e. serving web content) if a storage node is down, i don't want my I/O to hang for 20 seconds before Gluster tries another node. Can this be set lower ? What would be the potential drawbacks (if any) ? -- Daniel Maher <dma AT witbe.net>
