[Gluster-devel] Server-side AFR

Mon Jan 5 11:03:51 UTC 2009

Hi,

I have a setup using server-side afr (glusterfs-1.3.11). Is it possible
when one of the server went down, it will result in the other server
also clashing a few seconds later? Below are some of the log just before
the second server starting to crash after the first server went down.
Could someone also explain what " forced unwinding frame type(1) op(35)
reply=@0xb4091b58" and " frame : type(0) op(0)" means? Tks in advance.

2009-01-05 18:56:08 E [client-protocol.c:4834:client_protocol_cleanup]
gfs-storage1-ds: forced unwinding frame type(1) op(35) reply=@0xb4091b58

2009-01-05 18:56:08 E [client-protocol.c:4834:client_protocol_cleanup]
gfs-storage1-ds: forced unwinding frame type(1) op(35) reply=@0xb4091b58

2009-01-05 18:56:08 E [client-protocol.c:4834:client_protocol_cleanup]
gfs-storage1-ds: forced unwinding frame type(1) op(35) reply=@0xb4091b58

2009-01-05 18:56:08 E [client-protocol.c:4834:client_protocol_cleanup]
gfs-storage1-ds: forced unwinding frame type(1) op(35) reply=@0xb4091b58

2009-01-05 18:56:08 E [client-protocol.c:4834:client_protocol_cleanup]
gfs-storage1-ds: forced unwinding frame type(1) op(35) reply=@0xb4091b58

2009-01-05 18:56:08 E [client-protocol.c:4834:client_protocol_cleanup]
gfs-storage1-ds: forced unwinding frame type(1) op(35) reply=@0xb4091b58

.

.

.

2009-01-05 18:56:10 E [client-protocol.c:4430:client_lookup_cbk]
gfs-storage1-ds: no proper reply from server, returning ENOTCONN

2009-01-05 18:56:10 E [client-protocol.c:4834:client_protocol_cleanup]
gfs-storage1-ds: forced unwinding frame type(1) op(34) reply=@0xb4091b58

2009-01-05 18:56:10 E [client-protocol.c:4430:client_lookup_cbk]
gfs-storage1-ds: no proper reply from server, returning ENOTCONN

2009-01-05 18:56:10 D [tcp.c:87:tcp_disconnect] gfs-storage1-ds:
connection disconnected

2009-01-05 18:56:10 D [afr.c:6025:notify] gfs-ds-afr:
GF_EVENT_CHILD_DOWN from gfs-storage1-ds

2009-01-05 18:56:10 D [tcp-server.c:145:tcp_server_notify] server:
Registering socket (6) for new transport object of 172.18.229.31

2009-01-05 18:56:10 D [inode.c:321:__destroy_inode] gfs/inode: destroy
inode(123814120) [@0xb40afdb0]

2009-01-05 18:56:10 W [client-protocol.c:1296:client_close]
gfs-storage1-ds: no valid fd found, returning

2009-01-05 18:56:10 D [ip.c:120:gf_auth] gfs: allowed = "*", received ip
addr = "172.18.229.31"

2009-01-05 18:56:10 D [server-protocol.c:5674:mop_setvolume] server:
accepted client from 172.18.229.31:1020

2009-01-05 18:56:10 D [tcp-server.c:145:tcp_server_notify] server:
Registering socket (42) for new transport object of 172.18.229.41

2009-01-05 18:56:10 D [ip.c:120:gf_auth] gfs: allowed = "*", received ip
addr = "172.18.229.41"

2009-01-05 18:56:10 D [server-protocol.c:5674:mop_setvolume] server:
accepted client from 172.18.229.41:1019

2009-01-05 18:56:10 D [tcp-server.c:145:tcp_server_notify] server:
Registering socket (43) for new transport object of 172.18.229.33

2009-01-05 18:56:10 W [client-protocol.c:1296:client_close]
gfs-storage1-ds: no valid fd found, returning

2009-01-05 18:56:10 E [client-protocol.c:325:client_protocol_xfer]
gfs-storage1-ds: transport_submit failed

2009-01-05 18:56:10 W [client-protocol.c:1296:client_close]
gfs-storage1-ds: no valid fd found, returning

2009-01-05 18:56:10 D [tcp-client.c:77:tcp_connect] gfs-storage1-ds:
socket fd = 41

2009-01-05 18:56:10 D [inode.c:321:__destroy_inode] gfs/inode: destroy
inode(113328382) [@0xb3ac0348]

2009-01-05 18:56:10 E [server-protocol.c:186:generic_reply] server:
transport_writev failed

2009-01-05 18:56:10 D [tcp-client.c:107:tcp_connect] gfs-storage1-ds:
finalized on port `1023'

2009-01-05 18:56:10 D [tcp-client.c:128:tcp_connect] gfs-storage1-ds:
defaulting remote-port to 6996

2009-01-05 18:56:10 E [server-protocol.c:186:generic_reply] server:
transport_writev failed

2009-01-05 18:56:10 D [common-utils.c:179:gf_resolve_ip] resolver: DNS
cache not present, freshly probing hostname: cyvmvqb3.world.net

2009-01-05 18:56:10 D [server-protocol.c:6269:server_protocol_cleanup]
server: cleaned up transport state for client 172.18.229.33:1023

2009-01-05 18:56:10 E [server-protocol.c:186:generic_reply] server:
transport_writev failed

2009-01-05 18:56:10 E [server-protocol.c:186:generic_reply] server:
transport_writev failed

2009-01-05 18:56:10 E [server-protocol.c:186:generic_reply] server:
transport_writev failed

2009-01-05 18:56:10 D [server-protocol.c:6269:server_protocol_cleanup]
server: cleaned up transport state for client 172.18.229.33:1021

2009-01-05 18:56:10 E [server-protocol.c:186:generic_reply] server:
transport_writev failed

2009-01-05 18:56:10 D [server-protocol.c:6269:server_protocol_cleanup]
server: cleaned up transport state for client 172.18.229.31:1021

2009-01-05 18:56:10 D [common-utils.c:204:gf_resolve_ip] resolver:
returning IP:172.18.228.156[0] for hostname: cyvmvqb3.world.net

2009-01-05 18:56:10 D [common-utils.c:212:gf_resolve_ip] resolver:
flushing DNS cache

2009-01-05 18:56:10 E [server-protocol.c:186:generic_reply] server:
transport_writev failed

2009-01-05 18:56:10 D [tcp-client.c:161:tcp_connect] gfs-storage1-ds:
connect on 41 in progress (non-blocking)

2009-01-05 18:56:10 D [server-protocol.c:6269:server_protocol_cleanup]
server: cleaned up transport state for client 172.18.229.41:1020

2009-01-05 18:56:10 E [server-protocol.c:186:generic_reply] server:
transport_writev failed

2009-01-05 18:56:10 E [server-protocol.c:186:generic_reply] server:
transport_writev failed

2009-01-05 18:56:10 D [server-protocol.c:6269:server_protocol_cleanup]
server: cleaned up transport state for client 172.18.229.35:1023

2009-01-05 18:56:10 E [server-protocol.c:186:generic_reply] server:
transport_writev failed

2009-01-05 18:56:10 D [server-protocol.c:6269:server_protocol_cleanup]
server: cleaned up transport state for client 172.18.229.40:1021

TLA Repo Revision: glusterfs--mainline--2.5--patch-795

Time : 2009-01-05 18:56:10

Signal Number : 11

/usr/sbin/glusterfsd -f /etc/glusterfs/glusterfs-server.vol -l
/var/log/glusterfsd.log -L DEBUG

volume server

  type protocol/server

  option auth.ip.gfs.allow *

  option auth.ip.gfs-ds-locks.allow *,127.0.0.1

  option transport-type tcp/server

  subvolumes gfs

end-volume

volume gfs

  type performance/io-threads

  option cache-size 96MB

  option thread-count 4

  subvolumes gfs-ds-afr

end-volume

volume gfs-ds-afr

  type cluster/afr

  subvolumes gfs-ds-locks gfs-storage1-ds

end-volume

volume gfs-storage1-ds

  type protocol/client

  option transport-timeout 10

  option remote-subvolume gfs-ds-locks

  option remote-host cyvmvqb3.world.net

  option transport-type tcp/client

end-volume

volume gfs-ds-locks

  type features/posix-locks

  subvolumes gfs-ds

end-volume

volume gfs-ds

  type storage/posix

  option directory /media/gfs-ds

end-volume

frame : type(0) op(0)

frame : type(0) op(0)

frame : type(0) op(0)

frame : type(0) op(0)

frame : type(0) op(0)

frame : type(0) op(0)

regards,

melvin

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://supercolony.gluster.org/pipermail/gluster-devel/attachments/20090105/8ff6e5c6/attachment-0003.html>