experimental xlators have been removed from the codebase. But we missed to remove the options related to experimental xlators from the codebase. This patch removes those options. fixes: bz#1683506 Change-Id: I3fa7e14c6cd8ebde5cebc8d2b0cb2409bf37c1ae Signed-off-by: Sanju Rakonde <srakonde@redhat.com> (cherry picked from commit 5cddd4d758014fe116d9c130632eada2ecded88c)
Explicit invalidation by calling inode_invalidate is necessary when same (meta)data is shared/access across multiple mounts. Without an explicit inode_invalidate call, caches in the mount which didn't witness writes wouldn't be aware of changes as writes wouldn't have passed through them. However, if (meta)data is not shared, all relevant I/O goes through the cache of single mount and hence is coherent with (meta)data on bricks always. So, explicit inode invalidation can be disabled for this case which gives a huge performance boost for workloads that write data and then immediately read the data they just wrote. Note that otherwise, local writes (which pass through the cache) will change ctime and cause unnecessary invalidations. The name of the option that controls this behavior is "performance.global-cache-invalidation". This option is global and it purges caches both in glusterfs and kernel stack for native FUSE mounts. For non-native FUSE mounts, it purges cache only from glusterfs stack. This option is effective only when performance.stat-prefetch is on. Note that there is a similar option "performance.cache-invalidation", but the scope of that option is limited to quick-read and md-cache. Change-Id: I462bb4b65ff9aae1f6ba76f50b1f2f94fb10323b Signed-off-by: Raghavendra Gowdappa <rgowdapp@redhat.com> updates: bz#1674364 (cherry picked from commit 2b5aa4489de2017a03bcb6ec8986286f0c76a670)
With the feature enabled, some of the performance testing results, specially those which create millions of small files, got approximately 4x regression compared to version before enabling this. On master without this patch: 765 creates/sec On master with this patch : 3380 creates/sec Also there seems to be regression caused by this in 'ls -l' workload. On master without this patch: 3030 files/sec On master with this patch : 16610 files/sec This is a feature added to handle multiple clients parallely operating (specially those which race for file creates with same name) on a single namespace/directory. Considering that is < 3% of Gluster's usecase right now, it makes sense to disable the feature by default, so we don't penalize the default users who doesn't bother about this usecase. Also note that the client side translators, specially, distribute, replicate and disperse already handle the issue upto 99.5% of the cases without SDFS, so it makes sense to keep the feature disabled by default. Credits: Shyamsunder <srangana@redhat.com> for running the tests and getting the numbers. Change-Id: Iec49ce1d82e621e9db25eb633fcb1d932e74f4fc Updates: bz#1670031 Signed-off-by: Amar Tumballi <amarts@redhat.com>
There is a low level security issue with fencing since one client can preempt another client's lock. This patch does not completely eliminate the issue of a client misbehaving, but certainly it adds a security layer for default use cases that does not need fencing. Change-Id: I55cd15f2ed1ae0f2556e3d27a2ef4bc10fdada1c updates: #466 Signed-off-by: Susant Palai <spalai@redhat.com>
With read-after-open being set to yes by default, if open-behind sees any reads, it'll do an open on backend (and hence flush/release later). This means with the current order of quick-read and open-behind, open-behind sees all reads and hence also does open bringing down performance for small file reads. Since for small files, reads are absorbed by quick-read, if quick-read is made a parent of open-behind, ob doesn't witness any reads. For read-only workloads, this means ob doen't do any opens (even with read-after-open yes and use-anonymous-fd no). Change-Id: I138a42b006d104cff43ee6f07829e39c36f6f234 Signed-off-by: Raghavendra Gowdappa <rgowdapp@redhat.com> Fixes: bz#1659327
Fixes: bz#1659868 Change-Id: I38675ba4d47c8ba7f94cfb4734692683ddb3dcfd Signed-off-by: Jiffin Tony Thottan <jthottan@redhat.com>
get_mux_limit_per_process () reads the global option dictionary and in case it doesn't find out a key, assumes that cluster.max-bricks-per-process option isn't configured however the default value should be picked up in such case. Change-Id: I35dd8da084adbf59793d58557e818d8e6c17f9f3 Fixes: bz#1656951 Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
We still use gnfs on our side, so do a little work to support server.all-squash. Just like server.root-squash, it's also a volume wide option. Also see bz#1285126 $ gluster volume set <VOLNAME> server.all-squash on Note: If you enable server.root-squash and server.all-squash at the same time, only server.all-squash works. Please refer to following table +---------------+-----------------+---------------------------+ | |all_squash | no_all_squash | +-------------------------------------------------------------+ | | |anonuid/anongid for root | |root_squash |anonuid/anongid |useruid/usergid for no-root| +-------------------------------------------------------------+ |no_root_squash |anonuid/anongid |useruid/usergid | +-------------------------------------------------------------+ Updates bz#1285126 Signed-off-by: Xie Changlong <xiechanglong@cmss.chinamobile.com> Signed-off-by: Xue Chuanyu <xuechuanyu@cmss.chinamobile.com> Change-Id: Iea043318fe6e9a75fa92b396737985062a26b47e
Change-Id: Ia2c6a10e2b76a4aa8bd4ea97e5ce33bdc813942e Fixes: bz#1652118 Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
A new constant named GF_NETWORK_TIMEOUT has been defined and all references to the hard-coded timeout of 42 seconds have been replaced with this constant. Change-Id: Id30f5ce4f1230f9288d9e300538624bcf1a6da27 fixes: bz#1652852 Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>