| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
If the glusterfs fuse client process is unable to
process the invalidate requests quickly enough, the
number of such requests quickly grows large enough
to use a significant amount of memory.
We are now introducing another option to set an upper
limit on these to prevent runaway memory usage.
Change-Id: Iddfff1ee2de1466223e6717f7abd4b28ed947788
Fixes: bz#1732717
Signed-off-by: N Balachandran <nbalacha@redhat.com>
|
| |
|
|
|
|
| |
Fixes: bz#1644322
Change-Id: I53e8fa362cd8c7d04fb1c4abb606a9abb642c592
Signed-off-by: Csaba Henk <csaba@redhat.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
The files which were created before ctime enabled would not
have "trusted.glusterfs.mdata"(stores time attributes) xattr.
Upon fops which modifies either ctime or mtime, the xattr
gets created with latest ctime, mtime and atime, which is
incorrect. It should update only the corresponding time
attribute and rest from backend
Solution:
Creating xattr with values from brick is not possible as
each brick of replica set would have different times.
So create the xattr upon successful lookup if the xattr
is not created
Note To Reviewers:
The time attributes used to set xattr is got from successful
lookup. Instead of sending the whole iatt over the wire via
setxattr, a structure called mdata_iatt is sent. The mdata_iatt
contains only time attributes.
Change-Id: I5e535631ddef04195361ae0364336410a2895dd4
fixes: bz#1593542
Signed-off-by: Kotresh HR <khiremat@redhat.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
For a normal volume, we are updating the pid from a the
process while we do a daemonization or at the end of the
init if it is no-daemon mode. Along with updating the pid
we also lock the file, to make sure that the process is
running fine.
With brick mux, we were updating the pidfile from gluterd
after an attach/detach request.
There are two problems with this approach.
1) We are not holding a pidlock for any file other than parent
process.
2) There is a chance for possible race conditions with attach/detach.
For example, shd start and a volume stop could race. Let's say
we are starting an shd and it is attached to a volume.
While we trying to link the pid file to the running process,
this would have deleted by the thread that doing a volume stop.
Change-Id: I29a00352102877ce09ea3f376ca52affceb5cf1a
Updates: bz#1722541
Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
glusterfs-fops.h was moved to rpc/xdr to support compound fops.
(ref: https://review.gluster.org/14032, 2f945b86d3)
This was fine as long as all these header files were in single
include directory after 'install'. With the move to separate out
glusterfs specific header files into another directory inside
/usr/include (ref: https://review.gluster.org/21746, 20ef211cfa),
glusterfs-fops.h file was not in the proper path when an external
.c file tried to include any of glusterfs specific .h file (like
xlator.h).
Now, we have removed compound-fops, with that, none of the enums
declared in glusterfs-fops.h are actually getting used on wire
anymore. Hence, it makes sense to get this to libglusterfs/src
as a single point of definition. With this change, the external
programs can use glusterfs header files.
also remove some enum definitions which are not used in code
anymore.
Updates: bz#1636297
Change-Id: I423c44d3dbe2efc777299c544ece3cb172fc7e44
Signed-off-by: Amar Tumballi <amarts@redhat.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The way delta_blocks is computed in shard is incorrect, when a file
is truncated to a lower size. The accounting only considers change
in size of the last of the truncated shards.
FIX:
Get the block-count of each shard just before an unlink at posix in
xdata. Their summation plus the change in size of last shard
(from an actual truncate) is used to compute delta_blocks which is
used in the xattrop for size update.
Change-Id: I9128a192e9bf8c3c3a959e96b7400879d03d7c53
fixes: bz#1705884
Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
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
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
It was written just before fill_void() call.
Note that there was a possible overflow if the hostname was too long
(unrelated to this patch), but it is now also fixed, as we use a smaller buffer
for the hostname. This, in turn, forces us to check if gethostname() failed
and add explicitly the terminating null to it.
Change-Id: I45fbc0a8e105f1247f3cbf61befac06fabbaea06
updates: bz#1193929
Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
|
| |
|
|
|
|
| |
Change-Id: Iec5ce7f17fbf899f881a58cd20c4c967e3b71668
fixes: bz#1642168
Signed-off-by: Anuradha Talur <atalur@commvault.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem:
Shd daemon is per node, which means they create a graph
with all volumes on it. While this is a great for utilizing
resources, it is so good in terms of performance and managebility.
Because self-heal daemons doesn't have capability to automatically
reconfigure their graphs. So each time when any configurations
changes happens to the volumes(replicate/disperse), we need to restart
shd to bring the changes into the graph.
Because of this all on going heal for all other volumes has to be
stopped in the middle, and need to restart all over again.
Solution:
This changes makes shd as a per volume daemon, so that the graph
will be generated for each volumes.
When we want to start/reconfigure shd for a volume, we first search
for an existing shd running on the node, if there is none, we will
start a new process. If already a daemon is running for shd, then
we will simply detach a graph for a volume and reatach the updated
graph for the volume. This won't touch any of the on going operations
for any other volumes on the shd daemon.
Example of an shd graph when it is per volume
graph
-----------------------
| debug-iostat |
-----------------------
/ | \
/ | \
--------- --------- ----------
| AFR-1 | | AFR-2 | | AFR-3 |
-------- --------- ----------
A running shd daemon with 3 volumes will be like-->
graph
-----------------------
| debug-iostat |
-----------------------
/ | \
/ | \
------------ ------------ ------------
| volume-1 | | volume-2 | | volume-3 |
------------ ------------ ------------
Change-Id: Idcb2698be3eeb95beaac47125565c93370afbd99
fixes: bz#1659708
Signed-off-by: Mohammed Rafi KC <rkavunga@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>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
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
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
design reference: https://review.gluster.org/#/c/glusterfs-specs/+/21925/
This patch adds the lock preempt support.
Note: The current model stores lock enforcement information as separate
xattr on disk. There is another effort going in parallel to store this
in stat(x) of the file. This patch is self sufficient to add fencing
support. Based on the availability of the stat(x) support either I will
rebase this patch or we can modify the necessary bits post merging this
patch.
Change-Id: If4a42f3e0afaee1f66cdb0360ad4e0c005b5b017
updates: #466
Signed-off-by: Susant Palai <spalai@redhat.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
With brick mux, the number of threads increases as the number of
bricks increases. As an initiative to reduce the number of
threads in brick mux scenario, replacing janitor thread to use
synctask infra.
Now close() and closedir() handle by separate janitor
thread which is linked with glusterfs_ctx.
Updates #475
Change-Id: I0c4aaf728125ab7264442fde59f3d08542785f73
Signed-off-by: Poornima G <pgurusid@redhat.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently mem-pool implementation provides api to get from the
mem pool based on the struct type. This is to retain api
compatibility with the old implementation of mem pool. Internally
in the mem pool structure there is a mapping from struct to size
based pools.
In this patch, we are adding new APIs to fetch memory from mem pool,
given a size.
Change-Id: Ib220ee45ebd134a7be8f6482db5a592dbb7b9211
Updates: #325
Signed-off-by: Poornima G <pgurusid@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 devel package headers are referenced in code using
include semantics for a program, this while it works can be better
especially when dealing with out of tree xlator builds or in
general out of tree devel package usage.
Towards this, the following changes are done,
- moved all devel headers under a glusterfs directory
- Included these headers using system header notation <> in all
code outside of libglusterfs
- Included these headers using own program notation "" within
libglusterfs
This change although big, is just moving around the headers and
making it correct when including these headers from other sources.
This helps us correctly include libglusterfs includes without
namespace conflicts.
Change-Id: Id2a98854e671a7ee5d73be44da5ba1a74252423b
Updates: bz#1193929
Signed-off-by: ShyamsundarR <srangana@redhat.com>
|
