When "auto-invalidation" option was not specified for mount script, glusterfs cmdline ended with "--auto-invalidation=" option. This patch fixes that bug in mount script. Thanks to Amar for reporting it. Change-Id: Ie5cd4c6ffb3ac644d9d2b032035f914a935d05a8 Signed-off-by: Raghavendra Gowdappa <rgowdapp@redhat.com> updates: bz#1674364
Auto invalidation is necessary when same (meta)data is shared/access
across multiple mounts. 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, fuse-auto-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.
From glusterfs --help,
<snip>
--auto-invalidation[=BOOL] controls whether fuse-kernel can
auto-invalidate attribute, dentry and page-cache.
Disable this only if same files/directories are
not accessed across two different mounts
concurrently [default: "on"]
</snip>
Details on how disabling auto-invalidation helped to reduce pgbench
init times can be found at [1]. Time taken for pgbench init of scale
8000 was 8340s. That will be an improvement of 86% (59280s vs 8340s)
with auto-invalidations turned off along with other
optimizations. Just disabling auto-invalidation contributed 56%
improvement by reducing the total time taken by 33260s.
[1] https://www.spinics.net/lists/gluster-devel/msg25907.html
Change-Id: I0ed730dba9064bd9c576ad1800170a21e100e1ce
Signed-off-by: Raghavendra Gowdappa <rgowdapp@redhat.com>
updates: bz#1664934
fixes: bz#1622665 Change-Id: I777d67b1b62c284c62a02277238ad7538eef001e Signed-off-by: Iraj Jamali <ijamali@redhat.com>
This reverts commit b87c397091bac6a4a6dec4e45a7671fad4a11770. There seems to be some performance regression with the patch and hence recommended to have it reverted. Updates: #325 Change-Id: Id85d6203173a44fad6cf51d39b3e96f37afcec09
updates: bz#1622665 Change-Id: I9f3a75ed9be3d90f37843a140563c356830ef945 Signed-off-by: Amar Tumballi <amarts@redhat.com>
The current implementation of iobuf_pool has two problems: - prealloc of 12.5MB memory, this limits the scale factor of the gluster processes due to RAM requirements - lock contention, as the current implementation has one global iobuf_pool lock. Credits for debugging and addressing the same goes to Krutika Dhananjay <kdhananj@redhat.com>. Issue: #410 Hence changing the iobuf implementation to use per thread mem pool. This may theoritically appear to cause perf dip as there is no preallocation. But per thread mem pool will not have significant perf impact as the last allocated memory is kept alive for subsequent allocs, for some time. The worst case would be if iobufs requested are of random sizes each time. The best case is, if we get iobuf request of the same size. From the perf tests, this patch did not seem to cause any perf decrease. Note that, with this patch, the rdma performance is going to degrade drastically. In one of the previous patchsets we had fixes to not degrade rdma perf, but rdma is not supported and also not tested [1]. Hence the decision was to not have code in rdma that is not tested and not supported. [1] https://lists.gluster.org/pipermail/gluster-users.old/2018-July/034400.html Updates: #325 Change-Id: Ic2ef3bd498f9250dea25f25ba0c01fde19584b27 Signed-off-by: Poornima G <pgurusid@redhat.com>
Use the (f)getxattr based clearlocks interface to interrupt a pending lock request. updates: #465 Change-Id: I4e91a4d8791fc688fed400a02de4c53487e61be2 Signed-off-by: Csaba Henk <csaba@redhat.com>
The inode LRU mechanism is moot in fuse xlator (ie. there is no limit for the LRU list), as fuse inodes are referenced from kernel context, and thus they can only be dropped on request of the kernel. This might results in a high number of passive inodes which are useless for the glusterfs client, causing a significant memory overhead. This change tries to remedy this by extending the LRU semantics and allowing to set a finite limit on the fuse inode LRU. A brief history of problem: When gluster's inode table was designed, fuse didn't have any 'invalidate' method, which means, userspace application could never ask kernel to send a 'forget()' fop, instead had to wait for kernel to send it based on kernel's parameters. Inode table remembers the number of times kernel has cached the inode based on the 'nlookup' parameter. And 'nlookup' field is not used by no other entry points (like server-protocol, gfapi etc). Hence the inode_table of fuse module always has to have lru-limit as '0', which means no limit. GlusterFS always had to keep all inodes in memory as kernel would have had a reference to it. Again, the reason for this is, kernel's glusterfs inode reference was pointer of 'inode_t' structure in glusterfs. As it is a pointer, we could never free it (to prevent segfault, or memory corruption). Solution: In the inode table, handle the prune case of inodes with 'nlookup' differently, and call a 'invalidator' method, which in this case is fuse_invalidate(), and it sends the request to kernel for getting the forget request. When the kernel sends the forget, it means, it has dropped all the reference to the inode, and it will send the forget with the 'nlookup' parameter too. We just need to make sure to reduce the 'nlookup' value we have when we get forget. That automatically cause the relevant prune to happen. Credits: Csaba Henk, Xavier Hernandez, Raghavendra Gowdappa, Nithya B fixes: bz#1560969 Change-Id: Ifee0737b23b12b1426c224ec5b8f591f487d83a2 Signed-off-by: Amar Tumballi <amarts@redhat.com>
* libglusterfs changes to add new fop
* Fuse changes:
- Changes in fuse bridge xlator to receive and send responses
* posix changes to perform the op on the backend filesystem
* protocol and rpc changes for sending and receiving the fop
* gfapi changes for performing the fop
* tools: glfs-copy-file-range tool for testing copy_file_range fop
- Although, copy_file_range support has been added to the upstream
fuse kernel module, no release has been made yet of a kernel
which contains the support. It is expected to come in the
upcoming release of linux-4.20
So, as of now, executing copy_file_range fop on a fused based
filesystem results in fuse kernel module sending read on the
source fd and write on the destination fd.
Therefore a small gfapi based tool has been written to be able
test the copy_file_range fop. This tool is similar (in functionality)
to the example program given in copy_file_range man page.
So, running regular copy_file_range on a fuse mount point and
running gfapi based glfs-copy-file-range tool gives some idea about
how fast, the copy_file_range (or reflink) can be.
On the local machine this was the result obtained.
mount -t glusterfs workstation:new /mnt/glusterfs
[root@workstation ~]# cd /mnt/glusterfs/
[root@workstation glusterfs]# ls
file
[root@workstation glusterfs]# cd
[root@workstation ~]# time /tmp/a.out /mnt/glusterfs/file /mnt/glusterfs/new
real 0m6.495s
user 0m0.000s
sys 0m1.439s
[root@workstation ~]# time glfs-copy-file-range $(hostname) new /tmp/glfs.log /file /rrr
OPEN_SRC: opening /file is success
OPEN_DST: opening /rrr is success
FSTAT_SRC: fstat on /rrr is success
copy_file_range successful
real 0m0.309s
user 0m0.039s
sys 0m0.017s
This tool needs following arguments
1) hostname
2) volume name
3) log file path
4) source file path (relative to the gluster volume root)
5) destination file path (relative to the gluster volume root)
"glfs-copy-file-range <hostname> <volume> <log file path> <source> <destination>"
- Added a testcase as well to run glfs-copy-file-range tool
* io-stats changes to capture the fop for profiling
* NOTE:
- Added conditional check to see whether the copy_file_range syscall
is available or not. If not, then return ENOSYS.
- Added conditional check for kernel minor version in fuse_kernel.h
and fuse-bridge while referring to copy_file_range. And the kernel
minor version is kept as it is. i.e. 24. Increment it in future
when there is a kernel release which contains the support for
copy_file_range fop in fuse kernel module.
* The document which contains a writeup on this enhancement can be found at
https://docs.google.com/document/d/1BSILbXr_knynNwxSyyu503JoTz5QFM_4suNIh2WwrSc/edit
Change-Id: I280069c814dd21ce6ec3be00a884fc24ab692367
updates: #536
Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
Fixes: #164 Change-Id: I93ad6f0232a1dc534df099059f69951e1339086f Signed-off-by: Amar Tumballi <amarts@redhat.com>