Re: [Gluster-devel] Problem with little write/read files in Gluster 3.0.0

[nicolas prochazka]

Hi,

client replicate ( see my precedent email )

server 1 : glusterfsd and glusterfs  (replicate s1 s2 )

server 2 : glusterfsd and glustefs ( replicate s1 s2 ) 

cut the cable between server1 and server2

waiting for Five minutes

two servers reboots.

I think is fuse impact, because only fuse can panic the kernel .

2010/1/28 Vijay Bellur <address@hidden>

nicolas prochazka wrote:

Hi, before to release a 3.0.2 can you just test this scenario : 2 servers / client in replicate mode TCP/IP
down the link between servers, and ... see the server reboot .
I does not see any patch for this between 3.0.1 and 3.0.2 in git.

Hi Nicolas,

Are you talking of server side or client side replication?
Are you bringing down the link between two servers or between server and client?

Have you figured out the reason for the server reboot when the link is brought down?

Thanks,
Vijay