Currently, the list of xattrs that md-cache can cache is hard coded in the md-cache.c file, this necessiates code change and rebuild everytime a new xattr needs to be added to md-cache xattr cache list. With this patch, the user will be able to configure a comma seperated list of xattrs to be cached by md-cache Updates #297 Change-Id: Ie35ed607d17182d53f6bb6e6c6563ac52bc3132e Signed-off-by: Poornima G <pgurusid@redhat.com>
Problem: In replica 3 volumes there is a possibilities of ending up in split brain scenario, when multiple clients writing data on the same file at non overlapping regions in parallel. Scenario: - Initially all the copies are good and all the clients gets the value of data readables as all good. - Client C0 performs write W1 which fails on brick B0 and succeeds on other two bricks. - C1 performs write W2 which fails on B1 and succeeds on other two bricks. - C2 performs write W3 which fails on B2 and succeeds on other two bricks. - All the 3 writes above happen in parallel and fall on different ranges so afr takes granular locks and all the writes are performed in parallel. Since each client had data-readables as good, it does not see file going into split-brain in the in_flight_split_brain check, hence performs the post-op marking the pending xattrs. Now all the bricks are being blamed by each other, ending up in split-brain. Fix: Have an option to take either full lock or range lock on files while doing data transactions, to prevent the possibility of ending up in split brains. With this change, by default the files will take full lock while doing IO. If you want to make use of the old range lock change the value of "cluster.full-lock" to "no". Change-Id: I7893fa33005328ed63daa2f7c35eeed7c5218962 BUG: 1535438 Signed-off-by: karthik-us <ksubrahm@redhat.com>
The locks xlator now is able to send a contention notification to the current owner of the lock. This is only a notification that can be used to improve performance of some client side operations that might benefit from extended duration of lock ownership. Nothing is done if the lock owner decides to ignore the message and to not release the lock. For forced release of acquired resources, leases must be used. Change-Id: I7f1ad32a0b4b445505b09908a050080ad848f8e0 Signed-off-by: Xavier Hernandez <xhernandez@datalab.es>
Change-Id: Ic4fcf2087f295d3dade944efb8fd08f7e2d7d516 BUG: 1531149 Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
In glusterd volfile generation code options should be ornamented with the VOLOPT_FLAG_* flags. However, some are ornamented with OPT_FLAG_* flags (which are to be used in xlator context). The impact is: the OPT_FLAG_* that occurs is OPT_FLAG_CLIENT_OPT, which has the same value as VOLOPT_FLAG_XLATOR_OPT, so what was meant is "option affects clients" and what was there means "option enables/disables xlators". Because of this semantic shift, op version might be incorrectly calculated for volumes and clients. (At this point it's a theoretical possibility. Actual occurrence might depend on connecting client & server versions; it's also possible that there exists a proof of concept scenario but it's irrealistic.) This commit eliminates the OPT_FLAG_* occurrences from glusterd code, and replaces them with the appropriate VOLOPT_FLAG_* flags. Change-Id: Ia4e6fbac738d5a8d889c0f5561c4dea6783250b1 Signed-off-by: Csaba Henk <csaba@redhat.com>
First of all, this patch reverts commit 635c1c3 as the same is causing a regression with bricks not coming up on time when a node is rebooted. This patch tries to fix the problem in a different way by just trying to connect to an existing running brick when quorum status is not applicable. Change-Id: I0efb5901832824b1c15dcac529bffac85173e097 BUG: 1509845 Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
This patch re-enables the geo-rep test cases. Along with it does following optimizations. 1. Use EXPECT_WITHIN instead of sleep 2. Clean up geo-rep ssh key after test 3. Changes to gverify.sh and S56glusterd-geo-rep-create-post.sh to use the given ssh identity file for geo-rep create 4. Make gluster-command-dir configurable and introduce slave-gluster-command-dir which points the parent directory of gluster binaries in master and slave respectively. Change-Id: Ia7696278d9dd3ba04224dcd7c3564088ca970b04 BUG: 1480491 Signed-off-by: Kotresh HR <khiremat@redhat.com>
Commit 30e0b86 tried to address all the stale port issues glusterd had in case of a brick is abruptly killed. For brick multiplexing case because of a bug the portmap entry was not getting removed. This patch addresses the same. Change-Id: Ib020b967a9b92f1abae9cab9492f0cacec59aaa1 BUG: 1530281 Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
In a replicated volume it was allowing to set the quorum-count value between the range [1 - 2147483647]. This patch adds validation for allowing only maximum of replica_count number of quorum-count value to be set on a volume. Change-Id: I13952f3c6cf498c9f2b91161503fc0fba9d94898 BUG: 1529515 Signed-off-by: karthik-us <ksubrahm@redhat.com>
Problem : after brick reset/replace snapshot creation fails
Solution : During brick reset/replace when we validate and aggrigate
dictionary data from another node it was rewriting
'mount_dir' value to NULL which is critical for snapshot
creation.
Change-Id: Iabefbfcef7d8ac4cbd2a241e821c0e51492c093e
BUG: 1512451
Signed-off-by: Sunny Kumar <sunkumar@redhat.com>