<feed xmlns='http://www.w3.org/2005/Atom'>
<title>glusterfs.git/rpc/rpc-transport, branch v3.12dev</title>
<subtitle></subtitle>
<link rel='alternate' type='text/html' href='http://git.gluster.org/cgit/glusterfs.git/'/>
<entry>
<title>rpc: add options to manage socket keepalive lifespan</title>
<updated>2017-04-12T09:44:01+00:00</updated>
<author>
<name>Milind Changire</name>
<email>mchangir@redhat.com</email>
</author>
<published>2017-04-11T07:00:06+00:00</published>
<link rel='alternate' type='text/html' href='http://git.gluster.org/cgit/glusterfs.git/commit/?id=6b8df081b46ac4f485c86a5052fc30472e74bfbb'/>
<id>6b8df081b46ac4f485c86a5052fc30472e74bfbb</id>
<content type='text'>
Problem:
Default values for handling socket timeouts for brick responses are
insufficient for aggressive applications such as databases.

Solution:
Add 1:1 gluster options for keepalive, keepalive-idle,
keepalive-interval and keepalive-timeout as per the socket level options
available as per tcp(7) man page.

Default values for options are NOT agressive and continue to be values
which result in default timeout when only the keep alive option is
turned on.

These options are Linux specific and will not be applicable to the
*BSDs.

Change-Id: I2a08ecd949ca8ceb3e090d336ad634341e2dbf14
BUG: 1426059
Signed-off-by: Milind Changire &lt;mchangir@redhat.com&gt;
Reviewed-on: https://review.gluster.org/16731
Smoke: Gluster Build System &lt;jenkins@build.gluster.org&gt;
CentOS-regression: Gluster Build System &lt;jenkins@build.gluster.org&gt;
NetBSD-regression: NetBSD Build System &lt;jenkins@build.gluster.org&gt;
Reviewed-by: Raghavendra G &lt;rgowdapp@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Problem:
Default values for handling socket timeouts for brick responses are
insufficient for aggressive applications such as databases.

Solution:
Add 1:1 gluster options for keepalive, keepalive-idle,
keepalive-interval and keepalive-timeout as per the socket level options
available as per tcp(7) man page.

Default values for options are NOT agressive and continue to be values
which result in default timeout when only the keep alive option is
turned on.

These options are Linux specific and will not be applicable to the
*BSDs.

Change-Id: I2a08ecd949ca8ceb3e090d336ad634341e2dbf14
BUG: 1426059
Signed-off-by: Milind Changire &lt;mchangir@redhat.com&gt;
Reviewed-on: https://review.gluster.org/16731
Smoke: Gluster Build System &lt;jenkins@build.gluster.org&gt;
CentOS-regression: Gluster Build System &lt;jenkins@build.gluster.org&gt;
NetBSD-regression: NetBSD Build System &lt;jenkins@build.gluster.org&gt;
Reviewed-by: Raghavendra G &lt;rgowdapp@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>transport: allow OS to assign us a port</title>
<updated>2017-03-12T15:28:45+00:00</updated>
<author>
<name>Kevin Vigor</name>
<email>kvigor@fb.com</email>
</author>
<published>2016-12-15T21:41:14+00:00</published>
<link rel='alternate' type='text/html' href='http://git.gluster.org/cgit/glusterfs.git/commit/?id=eac6dfc314abe4dc50c54bf6c6cc004dfd73d5ac'/>
<id>eac6dfc314abe4dc50c54bf6c6cc004dfd73d5ac</id>
<content type='text'>
Replace complex and slow port selection code with bind(0) which
already respects privileged ports.

Change-Id: I408a8528e58e1aafcd32eba6a8f1a759e0bf274e
BUG: 1405628
Reviewed-on-release-3.8-fb: http://review.gluster.org/16150
Signed-off-by: Jeff Darcy &lt;jdarcy@redhat.com&gt;
Reviewed-on: https://review.gluster.org/16178
Smoke: Gluster Build System &lt;jenkins@build.gluster.org&gt;
Reviewed-by: Niels de Vos &lt;ndevos@redhat.com&gt;
NetBSD-regression: NetBSD Build System &lt;jenkins@build.gluster.org&gt;
CentOS-regression: Gluster Build System &lt;jenkins@build.gluster.org&gt;
Reviewed-by: Vijay Bellur &lt;vbellur@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Replace complex and slow port selection code with bind(0) which
already respects privileged ports.

Change-Id: I408a8528e58e1aafcd32eba6a8f1a759e0bf274e
BUG: 1405628
Reviewed-on-release-3.8-fb: http://review.gluster.org/16150
Signed-off-by: Jeff Darcy &lt;jdarcy@redhat.com&gt;
Reviewed-on: https://review.gluster.org/16178
Smoke: Gluster Build System &lt;jenkins@build.gluster.org&gt;
Reviewed-by: Niels de Vos &lt;ndevos@redhat.com&gt;
NetBSD-regression: NetBSD Build System &lt;jenkins@build.gluster.org&gt;
CentOS-regression: Gluster Build System &lt;jenkins@build.gluster.org&gt;
Reviewed-by: Vijay Bellur &lt;vbellur@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>rpc: avoid logging success on failure</title>
<updated>2017-03-07T12:05:38+00:00</updated>
<author>
<name>Milind Changire</name>
<email>mchangir@redhat.com</email>
</author>
<published>2017-03-05T16:09:20+00:00</published>
<link rel='alternate' type='text/html' href='http://git.gluster.org/cgit/glusterfs.git/commit/?id=89c6bedc1c2e978f67ca29f212a357984cd8a2dd'/>
<id>89c6bedc1c2e978f67ca29f212a357984cd8a2dd</id>
<content type='text'>
Avoid logging Success in the event of failure especially when errno has
no meaningful value w.r.t. the failure. In this case the errno is set to
zero when there's indeed a failure at the RPC level.

Change-Id: If2cc81aa1e590023ed22892dacbef7cac213e591
BUG: 1426032
Signed-off-by: Milind Changire &lt;mchangir@redhat.com&gt;
Reviewed-on: https://review.gluster.org/16730
Smoke: Gluster Build System &lt;jenkins@build.gluster.org&gt;
NetBSD-regression: NetBSD Build System &lt;jenkins@build.gluster.org&gt;
CentOS-regression: Gluster Build System &lt;jenkins@build.gluster.org&gt;
Reviewed-by: N Balachandran &lt;nbalacha@redhat.com&gt;
Reviewed-by: Jeff Darcy &lt;jdarcy@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Avoid logging Success in the event of failure especially when errno has
no meaningful value w.r.t. the failure. In this case the errno is set to
zero when there's indeed a failure at the RPC level.

Change-Id: If2cc81aa1e590023ed22892dacbef7cac213e591
BUG: 1426032
Signed-off-by: Milind Changire &lt;mchangir@redhat.com&gt;
Reviewed-on: https://review.gluster.org/16730
Smoke: Gluster Build System &lt;jenkins@build.gluster.org&gt;
NetBSD-regression: NetBSD Build System &lt;jenkins@build.gluster.org&gt;
CentOS-regression: Gluster Build System &lt;jenkins@build.gluster.org&gt;
Reviewed-by: N Balachandran &lt;nbalacha@redhat.com&gt;
Reviewed-by: Jeff Darcy &lt;jdarcy@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>Return ret as soon as `this` is freed</title>
<updated>2017-03-01T17:30:17+00:00</updated>
<author>
<name>Nigel Babu</name>
<email>nigelb@redhat.com</email>
</author>
<published>2017-03-01T01:46:19+00:00</published>
<link rel='alternate' type='text/html' href='http://git.gluster.org/cgit/glusterfs.git/commit/?id=9babd965ba885ca8bea4c58e708a3b9f0d6f076c'/>
<id>9babd965ba885ca8bea4c58e708a3b9f0d6f076c</id>
<content type='text'>
Omitting this return causes a confusing error message in the next block,
which is also a case of use after free. This bug was found by Coverity
scan.

BUG: 789278
Change-Id: Ifd4932de437e8ae875ff191033ea43cff81b701d
Signed-off-by: Nigel Babu &lt;nigelb@redhat.com&gt;
Reviewed-on: https://review.gluster.org/16790
Smoke: Gluster Build System &lt;jenkins@build.gluster.org&gt;
NetBSD-regression: NetBSD Build System &lt;jenkins@build.gluster.org&gt;
CentOS-regression: Gluster Build System &lt;jenkins@build.gluster.org&gt;
Reviewed-by: Jeff Darcy &lt;jdarcy@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Omitting this return causes a confusing error message in the next block,
which is also a case of use after free. This bug was found by Coverity
scan.

BUG: 789278
Change-Id: Ifd4932de437e8ae875ff191033ea43cff81b701d
Signed-off-by: Nigel Babu &lt;nigelb@redhat.com&gt;
Reviewed-on: https://review.gluster.org/16790
Smoke: Gluster Build System &lt;jenkins@build.gluster.org&gt;
NetBSD-regression: NetBSD Build System &lt;jenkins@build.gluster.org&gt;
CentOS-regression: Gluster Build System &lt;jenkins@build.gluster.org&gt;
Reviewed-by: Jeff Darcy &lt;jdarcy@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>rpc: log more about socket disconnects</title>
<updated>2017-03-01T17:27:54+00:00</updated>
<author>
<name>Milind Changire</name>
<email>mchangir@redhat.com</email>
</author>
<published>2017-02-23T12:28:46+00:00</published>
<link rel='alternate' type='text/html' href='http://git.gluster.org/cgit/glusterfs.git/commit/?id=67a35ac54bfd61a920c1919fbde588a04ac3358a'/>
<id>67a35ac54bfd61a920c1919fbde588a04ac3358a</id>
<content type='text'>
Log more about the different paths leading to socket disconnect for ease
of debugging.

Log via gf_log_callingfn() in __socket_disconnect() at loglevel TRACE if
socket connection is being torn down.

Change-Id: I1e551c2d685784b5ec747f481179f64d524c0461
BUG: 1426125
Signed-off-by: Milind Changire &lt;mchangir@redhat.com&gt;
Reviewed-on: https://review.gluster.org/16732
Smoke: Gluster Build System &lt;jenkins@build.gluster.org&gt;
NetBSD-regression: NetBSD Build System &lt;jenkins@build.gluster.org&gt;
CentOS-regression: Gluster Build System &lt;jenkins@build.gluster.org&gt;
Reviewed-by: Jeff Darcy &lt;jdarcy@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Log more about the different paths leading to socket disconnect for ease
of debugging.

Log via gf_log_callingfn() in __socket_disconnect() at loglevel TRACE if
socket connection is being torn down.

Change-Id: I1e551c2d685784b5ec747f481179f64d524c0461
BUG: 1426125
Signed-off-by: Milind Changire &lt;mchangir@redhat.com&gt;
Reviewed-on: https://review.gluster.org/16732
Smoke: Gluster Build System &lt;jenkins@build.gluster.org&gt;
NetBSD-regression: NetBSD Build System &lt;jenkins@build.gluster.org&gt;
CentOS-regression: Gluster Build System &lt;jenkins@build.gluster.org&gt;
Reviewed-by: Jeff Darcy &lt;jdarcy@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>socket: Avoid flooding of SSL messages in case of failure/success</title>
<updated>2017-02-28T04:27:28+00:00</updated>
<author>
<name>Mohit Agrawal</name>
<email>moagrawa@redhat.com</email>
</author>
<published>2017-02-27T05:50:14+00:00</published>
<link rel='alternate' type='text/html' href='http://git.gluster.org/cgit/glusterfs.git/commit/?id=753f6dec4c99fb29d4b386b96b364c5a142ca547'/>
<id>753f6dec4c99fb29d4b386b96b364c5a142ca547</id>
<content type='text'>
Problem: Avoid flodding of SSL messages in case of failure/success

Solution: 1) Ideally ssl_setup_connection should be call after getting success on
             connect so update the condition before call socket_spawn in socket_connect.
          2) Change the message type to debug in case of success.

BUG: 1427018
Change-Id: Icb6101e49304d5fe539609b4afacfb1b50b62f84
Signed-off-by: Mohit Agrawal &lt;moagrawa@redhat.com&gt;
Reviewed-on: https://review.gluster.org/16767
Smoke: Gluster Build System &lt;jenkins@build.gluster.org&gt;
NetBSD-regression: NetBSD Build System &lt;jenkins@build.gluster.org&gt;
CentOS-regression: Gluster Build System &lt;jenkins@build.gluster.org&gt;
Reviewed-by: Jeff Darcy &lt;jdarcy@redhat.com&gt;
Reviewed-by: Raghavendra G &lt;rgowdapp@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Problem: Avoid flodding of SSL messages in case of failure/success

Solution: 1) Ideally ssl_setup_connection should be call after getting success on
             connect so update the condition before call socket_spawn in socket_connect.
          2) Change the message type to debug in case of success.

BUG: 1427018
Change-Id: Icb6101e49304d5fe539609b4afacfb1b50b62f84
Signed-off-by: Mohit Agrawal &lt;moagrawa@redhat.com&gt;
Reviewed-on: https://review.gluster.org/16767
Smoke: Gluster Build System &lt;jenkins@build.gluster.org&gt;
NetBSD-regression: NetBSD Build System &lt;jenkins@build.gluster.org&gt;
CentOS-regression: Gluster Build System &lt;jenkins@build.gluster.org&gt;
Reviewed-by: Jeff Darcy &lt;jdarcy@redhat.com&gt;
Reviewed-by: Raghavendra G &lt;rgowdapp@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>Free iobuf after using it, not before</title>
<updated>2017-02-26T16:30:55+00:00</updated>
<author>
<name>Michael Scherer</name>
<email>misc@redhat.com</email>
</author>
<published>2017-02-22T16:58:24+00:00</published>
<link rel='alternate' type='text/html' href='http://git.gluster.org/cgit/glusterfs.git/commit/?id=730f520a1ee246a5f4a08a321b6f97bdf93536dc'/>
<id>730f520a1ee246a5f4a08a321b6f97bdf93536dc</id>
<content type='text'>
Coverity warn of use after free here. I assume that
under pressure, this might crash the whole process.

Change-Id: I15fb5cfc9b509705e96e4156b739988d816bbef5
BUG: 789278
Signed-off-by: Michael Scherer &lt;misc@redhat.com&gt;
Reviewed-on: https://review.gluster.org/16719
Smoke: Gluster Build System &lt;jenkins@build.gluster.org&gt;
Tested-by: Michael Scherer &lt;misc@fedoraproject.org&gt;
Reviewed-by: Shyamsundar Ranganathan &lt;srangana@redhat.com&gt;
NetBSD-regression: NetBSD Build System &lt;jenkins@build.gluster.org&gt;
CentOS-regression: Gluster Build System &lt;jenkins@build.gluster.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Coverity warn of use after free here. I assume that
under pressure, this might crash the whole process.

Change-Id: I15fb5cfc9b509705e96e4156b739988d816bbef5
BUG: 789278
Signed-off-by: Michael Scherer &lt;misc@redhat.com&gt;
Reviewed-on: https://review.gluster.org/16719
Smoke: Gluster Build System &lt;jenkins@build.gluster.org&gt;
Tested-by: Michael Scherer &lt;misc@fedoraproject.org&gt;
Reviewed-by: Shyamsundar Ranganathan &lt;srangana@redhat.com&gt;
NetBSD-regression: NetBSD Build System &lt;jenkins@build.gluster.org&gt;
CentOS-regression: Gluster Build System &lt;jenkins@build.gluster.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>socket: GF_REF_PUT should be called outside lock</title>
<updated>2017-02-06T11:13:46+00:00</updated>
<author>
<name>Rajesh Joseph</name>
<email>rjoseph@redhat.com</email>
</author>
<published>2017-01-05T18:28:21+00:00</published>
<link rel='alternate' type='text/html' href='http://git.gluster.org/cgit/glusterfs.git/commit/?id=b3188c61d248526a070b1b18df1ea1d181b349d6'/>
<id>b3188c61d248526a070b1b18df1ea1d181b349d6</id>
<content type='text'>
GF_REF_PUT was called inside lock which can call
socket_poller_mayday which inturn tries to take the
same lock. This can lead to deadlock scenario.

BUG: 1410701
Change-Id: Ib3b161bcfeac810bd3593dc04c10ef984f996b17
Signed-off-by: Rajesh Joseph &lt;rjoseph@redhat.com&gt;
Reviewed-on: https://review.gluster.org/16343
Reviewed-by: Raghavendra G &lt;rgowdapp@redhat.com&gt;
CentOS-regression: Gluster Build System &lt;jenkins@build.gluster.org&gt;
Smoke: Gluster Build System &lt;jenkins@build.gluster.org&gt;
NetBSD-regression: NetBSD Build System &lt;jenkins@build.gluster.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
GF_REF_PUT was called inside lock which can call
socket_poller_mayday which inturn tries to take the
same lock. This can lead to deadlock scenario.

BUG: 1410701
Change-Id: Ib3b161bcfeac810bd3593dc04c10ef984f996b17
Signed-off-by: Rajesh Joseph &lt;rjoseph@redhat.com&gt;
Reviewed-on: https://review.gluster.org/16343
Reviewed-by: Raghavendra G &lt;rgowdapp@redhat.com&gt;
CentOS-regression: Gluster Build System &lt;jenkins@build.gluster.org&gt;
Smoke: Gluster Build System &lt;jenkins@build.gluster.org&gt;
NetBSD-regression: NetBSD Build System &lt;jenkins@build.gluster.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>socket: retry connect immediately if it fails</title>
<updated>2017-02-02T20:21:36+00:00</updated>
<author>
<name>Jeff Darcy</name>
<email>jdarcy@redhat.com</email>
</author>
<published>2017-02-02T03:00:32+00:00</published>
<link rel='alternate' type='text/html' href='http://git.gluster.org/cgit/glusterfs.git/commit/?id=5a57c1592a34ee6632ca1fb38e076dde381d1ae2'/>
<id>5a57c1592a34ee6632ca1fb38e076dde381d1ae2</id>
<content type='text'>
Previously we relied on a complex dance of setting flags, shutting
down the socket, tearing stuff down, getting an event, tearing more
stuff down, and waiting for a higher-level retry.  What we really
need, in the case where we're just trying to connect prematurely e.g.
to a brick that hasn't fully come up yet, is a simple retry of the
connect(2) call.

This was discovered by observing failures in ec-new-entry.t with
multiplexing enabled, but probably fixes other random failures as
well.

Change-Id: Ibedb8942060bccc96b02272a333c3002c9b77d4c
BUG: 1385758
Signed-off-by: Jeff Darcy &lt;jdarcy@redhat.com&gt;
Reviewed-on: https://review.gluster.org/16510
Smoke: Gluster Build System &lt;jenkins@build.gluster.org&gt;
NetBSD-regression: NetBSD Build System &lt;jenkins@build.gluster.org&gt;
CentOS-regression: Gluster Build System &lt;jenkins@build.gluster.org&gt;
Reviewed-by: Shyamsundar Ranganathan &lt;srangana@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Previously we relied on a complex dance of setting flags, shutting
down the socket, tearing stuff down, getting an event, tearing more
stuff down, and waiting for a higher-level retry.  What we really
need, in the case where we're just trying to connect prematurely e.g.
to a brick that hasn't fully come up yet, is a simple retry of the
connect(2) call.

This was discovered by observing failures in ec-new-entry.t with
multiplexing enabled, but probably fixes other random failures as
well.

Change-Id: Ibedb8942060bccc96b02272a333c3002c9b77d4c
BUG: 1385758
Signed-off-by: Jeff Darcy &lt;jdarcy@redhat.com&gt;
Reviewed-on: https://review.gluster.org/16510
Smoke: Gluster Build System &lt;jenkins@build.gluster.org&gt;
NetBSD-regression: NetBSD Build System &lt;jenkins@build.gluster.org&gt;
CentOS-regression: Gluster Build System &lt;jenkins@build.gluster.org&gt;
Reviewed-by: Shyamsundar Ranganathan &lt;srangana@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>libglusterfs+transport+io-threads: fix 256KB stack abuse</title>
<updated>2017-02-02T00:59:25+00:00</updated>
<author>
<name>Jeff Darcy</name>
<email>jdarcy@redhat.com</email>
</author>
<published>2016-10-27T15:51:47+00:00</published>
<link rel='alternate' type='text/html' href='http://git.gluster.org/cgit/glusterfs.git/commit/?id=c8a23cc6cd289dd28deb136bf2550f28e2761ef3'/>
<id>c8a23cc6cd289dd28deb136bf2550f28e2761ef3</id>
<content type='text'>
Some functions were allocating 64K booleans, which are (crazily) mapped to
4-byte ints, for a total of 256KB per call.  Changed to use bitfields instead,
so usage is now only 8KB per call.  This was the impediment to changing the
io-threads stack size, so that has been adjusted too.

Change-Id: I8781c4f2c8f2b830f4535e366995fac8dd0a8653
BUG: 1418095
Signed-off-by: Jeff Darcy &lt;jdarcy@redhat.com&gt;
Reviewed-on: https://review.gluster.org/15745
Smoke: Gluster Build System &lt;jenkins@build.gluster.org&gt;
NetBSD-regression: NetBSD Build System &lt;jenkins@build.gluster.org&gt;
CentOS-regression: Gluster Build System &lt;jenkins@build.gluster.org&gt;
Reviewed-by: N Balachandran &lt;nbalacha@redhat.com&gt;
Reviewed-by: Shyamsundar Ranganathan &lt;srangana@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Some functions were allocating 64K booleans, which are (crazily) mapped to
4-byte ints, for a total of 256KB per call.  Changed to use bitfields instead,
so usage is now only 8KB per call.  This was the impediment to changing the
io-threads stack size, so that has been adjusted too.

Change-Id: I8781c4f2c8f2b830f4535e366995fac8dd0a8653
BUG: 1418095
Signed-off-by: Jeff Darcy &lt;jdarcy@redhat.com&gt;
Reviewed-on: https://review.gluster.org/15745
Smoke: Gluster Build System &lt;jenkins@build.gluster.org&gt;
NetBSD-regression: NetBSD Build System &lt;jenkins@build.gluster.org&gt;
CentOS-regression: Gluster Build System &lt;jenkins@build.gluster.org&gt;
Reviewed-by: N Balachandran &lt;nbalacha@redhat.com&gt;
Reviewed-by: Shyamsundar Ranganathan &lt;srangana@redhat.com&gt;
</pre>
</div>
</content>
</entry>
</feed>
