gluster-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gluster-devel] NUFA Scheduler


From: gordan
Subject: Re: [Gluster-devel] NUFA Scheduler
Date: Fri, 16 May 2008 11:12:34 +0100 (BST)
User-agent: Alpine 1.10 (LRH 962 2008-03-14)

OK, I changed my config more in line with the single-process server-side AFR with additional clients, and when there is only one server up, I can see the files with ls, but when I try to cat a file, I get this:

2008-05-16 11:01:28 E [tcp-client.c:190:tcp_connect] home2: non-blocking connect() returned: 111 (Connection refused) 2008-05-16 11:01:28 E [client-protocol.c:4423:client_lookup_cbk] home2: no proper reply from server, returning ENOTCONN 2008-05-16 11:01:28 E [afr.c:1985:afr_selfheal] home: none of the children are up for locking, returning EIO 2008-05-16 11:01:28 E [fuse-bridge.c:692:fuse_fd_cbk] glusterfs-fuse: 179: (12) /test => -1 (5)

However, with 2 protocol/server blocks, one for exporting storage/posix (for other server nodes) and one for exporting cluster/afr (for client-only machines), it all seems to work. And in the working configuration, I am connecting to the protocol/server volume as the remote subvolume, not one of the protocol/server's subvolumes. Attached is my working spec file (home.vol) and the one that produces the above error (home.vol.new).

Gordan

On Fri, 16 May 2008, address@hidden wrote:

On Thu, 15 May 2008, Amar S. Tumballi wrote:

And are you saying that it is possible to connect via a protocol/client volume to
      a remote volume of type other than protocol/server?


Always protocol client connects to protocol server. But the 'option remote-subvolume' will be one of the subvolumes of server protocol. I thought that was noticeable in all the spec file
examples given.

I see. So protocol/client's remote-subvolume should refer to a protocol/server's subvolume, rather than the protocol/server's volume name itself?

So, each server has to use a separate configuration file, rather than having one spec file with multiple protocol/server volumes? Or is it supposed to be done by having each protocol/server listen on a different port with the same spec file?

Gordan


_______________________________________________
Gluster-devel mailing list
address@hidden
http://lists.nongnu.org/mailman/listinfo/gluster-devel

Attachment: home.vol.new
Description: Text document

Attachment: home.vol
Description: Text document


reply via email to

[Prev in Thread] Current Thread [Next in Thread]