glusterfs.git/transport, branch v3.0.5rc4

socket: make tcp keepalive work on OS X

2010-05-27T09:00:56+00:00

Signed-off-by: Csaba Henk 
Signed-off-by: Anand V. Avati 

BUG: 754 (enable tcp keepalive)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=754

BUG: 754 (enable tcp keepalive)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=754

socket: Support TCP-KEEPALIVE

2010-05-21T07:32:27+00:00

Introduces two new socket options:

1. transport.socket.keepalive: bool
Sets keepalive on a transport. On by default.

2. transport.socket.keepalive-interval: integer
Number of seconds between each keepalive message on the socket.
Default in linux is 2 hours. To change that value for a given
transport, use this option. Integer value is interval in seconds.

Signed-off-by: Shehjar Tikoo 
Signed-off-by: Anand V. Avati 

BUG: 754 (enable tcp keepalive)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=754

Adding GF_LOG_OCCASIONALLY to prevent repeated log messages when server goes down

2010-05-13T19:12:42+00:00

Signed-off-by: Anush Shetty 
Signed-off-by: Anand V. Avati 

BUG: 906 (when server is down, client log gets filled up with error msgs)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=906

Adding GF_LOG_OCCASIONALLY to prevent repeated log messages

2010-05-11T17:10:11+00:00

Signed-off-by: Anush Shetty 
Signed-off-by: Anand V. Avati 

BUG: 10 ([ glusterfs 2.0.0 ] - Repeated Log messages with invalid remote-host in protocol/client)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=10

transport/socket: guess an appropriate address family when it is not specified in configuration.

2010-03-31T08:41:01+00:00

Signed-off-by: Raghavendra G 
Signed-off-by: Anand V. Avati 

BUG: 561 (with address family set to 'inet-sdp', transport/socket doesn't work correctly)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=561

socket: Add NULL check for failed ioq entry allocation

2010-03-02T16:22:12+00:00

This is seen in OOM situations.

Signed-off-by: Shehjar Tikoo 
Signed-off-by: Anand V. Avati 

BUG: 679 (Crash in socket_ioq_new)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=679

transport/socket: use appropriate address family macro when address family is inet-sdp.

2010-01-22T13:34:35+00:00

- sa_family should be set to AF_INET_SDP only during call to socket system
    call. For all other calls, sa_family should be AF_INET.

Signed-off-by: Raghavendra G 
Signed-off-by: Anand V. Avati 

BUG: 561 (with address family set to 'inet-sdp', transport/socket doesn't work correctly)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=561

transport/socket: Enable NODELAY even if no option is set.

2010-01-12T10:12:49+00:00

Signed-off-by: Vijay Bellur 
Signed-off-by: Anand V. Avati 

BUG: 524 (TCP_NODELAY not set when expected on client)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=524

transport/ib-verbs: assign to qpreg before accessing it in __ib_verbs_lookup_peer.

2009-11-19T10:57:45+00:00

Signed-off-by: Raghavendra G 
Signed-off-by: Anand V. Avati 

BUG: 381 (glusterfs crash in ib-verbs)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=381

transport/ib-verbs: fix race-condition resulting in freeing of transport while it was still being used.

2009-11-16T08:41:36+00:00

- while handling a failed work completion status, the transport is
    disconnected. But further down in the code, the transport is still
    used. There can be a possibility that transport might've been freed
    by the time control reaches this point. More detailed description of
    events leading to this situation can be found at
    http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=381

Signed-off-by: Raghavendra G 
Signed-off-by: Anand V. Avati 

BUG: 381 (glusterfs crash in ib-verbs)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=381