glusterfs.git/glusterfsd/src/glusterfsd.h, branch v7.1 core: Log level changes do not effect on running client process 2019-04-15T04:30:43+00:00 Mohit Agrawal moagrawal@redhat.com 2019-04-04T04:26:11+00:00 798aadbe51a9a02dd98a0f861cc239ecf7c8ed57 Problem: commit c34e4161f3cb6539ec83a9020f3d27eb4759a975 set log-level per xlator during reconfigure only for a brick process not for the client process. Solution: 1) Change per xlator log-level only if brick_mux is enabled.To make sure about brick multiplex introudce a flag brick_mux at ctx->cmd_args. Note: There are two other changes done with this patch 1) Ignore client-log-level option to attach a brick with already running brick if brick_mux is enabled 2) Add a log to print pid of the running process to make easier debugging Change-Id: I39e85de778e150d0685cd9a79425ce8b4783f9c9 Signed-off-by: Mohit Agrawal <moagrawal@redhat.com> Fixes: bz#1696046 
Problem: commit c34e4161f3cb6539ec83a9020f3d27eb4759a975 set log-level
         per xlator during reconfigure only for a brick process not for
         the client process.

Solution: 1) Change per xlator log-level only if brick_mux is enabled.To make sure
             about brick multiplex introudce a flag brick_mux at ctx->cmd_args.

Note: There are two other changes done with this patch
      1) Ignore client-log-level option to attach a brick with
         already running brick if brick_mux is enabled
      2) Add a log to print pid of the running process to make easier
         debugging

Change-Id: I39e85de778e150d0685cd9a79425ce8b4783f9c9
Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
Fixes: bz#1696046
core: implement a global thread pool 2019-02-18T02:58:24+00:00 Xavi Hernandez xhernandez@redhat.com 2019-01-24T17:44:06+00:00 dddcf52020004d98f688ebef968de51d76cbf9a6 This patch implements a thread pool that is wait-free for adding jobs to the queue and uses a very small locked region to get jobs. This makes it possible to decrease contention drastically. It's based on wfcqueue structure provided by urcu library. It automatically enables more threads when load demands it, and stops them when not needed. There's a maximum number of threads that can be used. This value can be configured. Depending on the workload, the maximum number of threads plays an important role. So it needs to be configured for optimal performance. Currently the thread pool doesn't self adjust the maximum for the workload, so this configuration needs to be changed manually. For this reason, the global thread pool has been made optional, so that volumes can still use the thread pool provided by io-threads. To enable it for bricks, the following option needs to be set: config.global-threading = on This option has no effect if bricks are already running. A restart is required to activate it. It's recommended to also enable the following option when running bricks with the global thread pool: performance.iot-pass-through = on To enable it for a FUSE mount point, the option '--global-threading' must be added to the mount command. To change it, an umount and remount is needed. It's recommended to disable the following option when using global threading on a mount point: performance.client-io-threads = off To enable it for services managed by glusterd, glusterd needs to be started with option '--global-threading'. In this case all daemons, like self-heal, will be using the global thread pool. Currently it can only be enabled for bricks, FUSE mounts and glusterd services. The maximum number of threads for clients and bricks can be configured using the following options: config.client-threads config.brick-threads These options can be applied online and its effect is immediate most of the times. If one of them is set to 0, the maximum number of threads will be calcutated as #cores * 2. Some distributions use a very old userspace-rcu library (version 0.7) for this reason, some header files from version 0.10 have been copied into contrib/userspace-rcu and are used if the detected version is 0.7 or older. An additional change has been made to io-threads to prevent that threads are started when iot-pass-through is set. Change-Id: I09d19e246b9e6d53c6247b29dfca6af6ee00a24b updates: #532 Signed-off-by: Xavi Hernandez <xhernandez@redhat.com> 
This patch implements a thread pool that is wait-free for adding jobs to
the queue and uses a very small locked region to get jobs. This makes it
possible to decrease contention drastically. It's based on wfcqueue
structure provided by urcu library.

It automatically enables more threads when load demands it, and stops
them when not needed. There's a maximum number of threads that can be
used. This value can be configured.

Depending on the workload, the maximum number of threads plays an
important role. So it needs to be configured for optimal performance.
Currently the thread pool doesn't self adjust the maximum for the
workload, so this configuration needs to be changed manually.

For this reason, the global thread pool has been made optional, so that
volumes can still use the thread pool provided by io-threads.

To enable it for bricks, the following option needs to be set:

   config.global-threading = on

This option has no effect if bricks are already running. A restart is
required to activate it. It's recommended to also enable the following
option when running bricks with the global thread pool:

   performance.iot-pass-through = on

To enable it for a FUSE mount point, the option '--global-threading'
must be added to the mount command. To change it, an umount and remount
is needed. It's recommended to disable the following option when using
global threading on a mount point:

   performance.client-io-threads = off

To enable it for services managed by glusterd, glusterd needs to be
started with option '--global-threading'. In this case all daemons, like
self-heal, will be using the global thread pool.

Currently it can only be enabled for bricks, FUSE mounts and glusterd
services.

The maximum number of threads for clients and bricks can be configured
using the following options:

   config.client-threads
   config.brick-threads

These options can be applied online and its effect is immediate most of
the times. If one of them is set to 0, the maximum number of threads
will be calcutated as #cores * 2.

Some distributions use a very old userspace-rcu library (version 0.7)
for this reason, some header files from version 0.10 have been copied
into contrib/userspace-rcu and are used if the detected version is 0.7
or older.

An additional change has been made to io-threads to prevent that threads
are started when iot-pass-through is set.

Change-Id: I09d19e246b9e6d53c6247b29dfca6af6ee00a24b
updates: #532
Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
mount/fuse: expose auto-invalidation as a mount option 2019-02-02T03:07:35+00:00 Raghavendra Gowdappa rgowdapp@redhat.com 2019-01-29T02:35:07+00:00 a229ee1c8cdf8e0ac1abaeb60cabe6ab08f60546 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 
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
fuse: add --lru-limit option 2018-12-14T17:34:28+00:00 Amar Tumballi amarts@redhat.com 2018-10-16T11:01:49+00:00 d49b41e817d592c1904b6f01716df6546dad3ebe 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> 
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>
fuse: diagnostic FLUSH interrupt 2018-11-06T04:21:57+00:00 Csaba Henk csaba@redhat.com 2018-08-21T10:44:54+00:00 4c6b063463ae48b3509ff8e66cd391f8637a86af We add dummy interrupt handling for the FLUSH fuse message. It can be enabled by the "--fuse-flush-handle-interrupt" hidden command line option, or "-ofuse-flush-handle-interrupt=yes" mount option. It serves no other than diagnostic & demonstational purposes -- to exercise the interrupt handling framework a bit and to give an usage example. Documentation is also provided that showcases interrupt handling via FLUSH. Change-Id: I522f1e798501d06b74ac3592a5f73c1ab0590c60 updates: #465 Signed-off-by: Csaba Henk <csaba@redhat.com> 
We add dummy interrupt handling for the FLUSH
fuse message. It can be enabled by the
"--fuse-flush-handle-interrupt" hidden command line
option, or "-ofuse-flush-handle-interrupt=yes"
mount option.

It serves no other than diagnostic & demonstational
purposes -- to exercise the interrupt handling framework
a bit and to give an usage example.

Documentation is also provided that showcases interrupt
handling via FLUSH.

Change-Id: I522f1e798501d06b74ac3592a5f73c1ab0590c60
updates: #465
Signed-off-by: Csaba Henk <csaba@redhat.com>
Land clang-format changes 2018-09-12T11:52:48+00:00 Gluster Ant bugzilla-bot@gluster.org 2018-09-12T11:52:48+00:00 45a71c0548b6fd2c757aa2e7b7671a1411948894 Change-Id: I6f5d8140a06f3c1b2d196849299f8d483028d33b 
Change-Id: I6f5d8140a06f3c1b2d196849299f8d483028d33b
New flag to glusterfsd binary to print libexec dir 2018-09-05T06:44:59+00:00 Aravinda VK avishwan@redhat.com 2018-08-17T04:08:39+00:00 560bd8671f2f9ae17530ad1d23a209dc95d41f82 New CLI option for `glusterfsd` binary to get the path of libexec directory. This helps glusterd2 to detect the installed path of `gsyncd` and other binaries. Usage: `glusterfsd --print-libexecdir` Updates: bz#1193929 Change-Id: I8c1a74afd9acec7ee7bd3deabed9d9f20fe3fb5f Signed-off-by: Aravinda VK <avishwan@redhat.com> 
New CLI option for `glusterfsd` binary to get the path of
libexec directory. This helps glusterd2 to detect the
installed path of `gsyncd` and other binaries.

Usage: `glusterfsd --print-libexecdir`

Updates: bz#1193929
Change-Id: I8c1a74afd9acec7ee7bd3deabed9d9f20fe3fb5f
Signed-off-by: Aravinda VK <avishwan@redhat.com>
fuse: add support for kernel writeback cache 2018-05-04T17:42:12+00:00 Csaba Henk csaba@redhat.com 2018-05-03T08:22:18+00:00 2ac79ed8048753dfd2494d3a4d3b0e9411673e3a - Added kernel-writeback-cache command line and xlator option for requesting utilisation of the writeback cache of the kernel in FUSE_INIT (see [1]). - Added attr-times-granularity command line and xlator option via which granularity of the {a,m,c}time in stat (attr) data that we support can be indicated to kernel. This is a means to avoid divergence of the attr times between kernel and userspace that could occur with writeback-cache, while still maintaining maximum time precision the FUSE server is capable of (see [2]). - Handling FATTR_CTIME flag in FUSE_SETATTR that indicates presence of ctime in setattr payload. Currently we cannot associate arbitrary ctimes to files on backend, so we just touch them to update their ctimes to current time. Having ctimes in setattr payload is also a side effect of writeback cache (see [3] and [4]). [1]: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=4d99ff8, "fuse: Turn writeback cache on" [2]: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=e27c9d3, "fuse: fuse: add time_gran to INIT_OUT" [3]: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=1e18bda, "fuse: add .write_inode" [4]: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=ab9e13f, "fuse: allow ctime flushing to userspace" Updates: #435 Change-Id: Id174c8e0c815c4456c35f8c53e41a6a507d91855 Signed-off-by: Csaba Henk <csaba@redhat.com> 
- Added kernel-writeback-cache command line and xlator
  option for requesting utilisation of the writeback
  cache of the kernel in FUSE_INIT (see [1]).
- Added attr-times-granularity command line and xlator
  option via which granularity of the {a,m,c}time in
  stat (attr) data that we support can be indicated to
  kernel. This is a means to avoid divergence of the
  attr times between kernel and userspace that could
  occur with writeback-cache, while still maintaining
  maximum time precision the FUSE server is capable of
  (see [2]).
- Handling FATTR_CTIME flag in FUSE_SETATTR that
  indicates presence of ctime in setattr payload.
  Currently we cannot associate arbitrary ctimes to
  files on backend, so we just touch them to update
  their ctimes to current time. Having ctimes in setattr
  payload is also a side effect of writeback cache
  (see [3] and [4]).

[1]: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=4d99ff8,
     "fuse: Turn writeback cache on"
[2]: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=e27c9d3,
     "fuse: fuse: add time_gran to INIT_OUT"
[3]: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=1e18bda,
     "fuse: add .write_inode"
[4]: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=ab9e13f,
     "fuse: allow ctime flushing to userspace"

Updates: #435
Change-Id: Id174c8e0c815c4456c35f8c53e41a6a507d91855
Signed-off-by: Csaba Henk <csaba@redhat.com>
Make glusterfsd binary print statedump & xlator dir 2018-04-19T09:48:40+00:00 Prashanth Pai ppai@redhat.com 2018-04-19T07:20:20+00:00 9e7082b756058fe04d79c046fbd6dcbfa75a1716 The glusterd2 needs following options, some of which are provided by gluster CLI today: --print-xlatordir --print-statedumpdir --print-logdir However, the CLI package need not be present on the machine running glusterd2. This change adds the above CLI options to glusterfsd binary which glusterd2 depends on. Reverts 9a1ae47c8d60836ae0628a04a153f28c1085c0e8 Related changes: https://review.gluster.org/#/c/19882/ https://github.com/gluster/glusterd2/pull/663 Updates: bz#1193929 Change-Id: I18c123b0d3350d2bd4f2400783e3b94e402a4e29 Signed-off-by: Prashanth Pai <ppai@redhat.com> 
The glusterd2 needs following options, some of which are provided by
gluster CLI today:

--print-xlatordir
--print-statedumpdir
--print-logdir

However, the CLI package need not be present on the machine running
glusterd2. This change adds the above CLI options to glusterfsd binary
which glusterd2 depends on.

Reverts 9a1ae47c8d60836ae0628a04a153f28c1085c0e8

Related changes:
https://review.gluster.org/#/c/19882/
https://github.com/gluster/glusterd2/pull/663

Updates: bz#1193929
Change-Id: I18c123b0d3350d2bd4f2400783e3b94e402a4e29
Signed-off-by: Prashanth Pai <ppai@redhat.com>
mount/fuse: Add support for multi-threaded fuse readers 2018-04-02T06:10:30+00:00 Krutika Dhananjay kdhananj@redhat.com 2018-01-09T09:41:00+00:00 08fadcc2a706342e4eee79dc7d9b48ba01fcb312 Usage: Use 'reader-thread-count=<NUM>' as command line option to set the thread count at the time of mounting the volume. Next task is to make these threads auto-scale based on the load, instead of having the user remount the volume everytime to change the thread count. Updates #412 Change-Id: I94aa1505e5ae6a133683d473e0e4e0edd139b76b Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com> 
Usage: Use 'reader-thread-count=<NUM>' as command line option to
set the thread count at the time of mounting the volume.

Next task is to make these threads auto-scale based on the load,
instead of having the user remount the volume everytime to change
the thread count.

Updates #412

Change-Id: I94aa1505e5ae6a133683d473e0e4e0edd139b76b
Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>