glusterfs.git/xlators/cluster/dht/src/dht-shared.c, branch v3.7.1

dht: make lookup-unhashed=auto do something actually useful

2015-05-10T04:55:09+00:00

The key concept here is to determine whether a directory is "clean" by
comparing its last-known-good topology to the current one for the
volume.  These are stored as "commit hashes" on the directory and the
volume root respectively.  The volume's commit hash changes whenever a
brick is added or removed, and a fix-layout is done.  A directory's
commit hash changes only when a full rebalance (not just fix-layout)
is done on it.  If all bricks are present and have a directory
commit hash that matches the volume commit hash, then we can assume
that every file is in its "proper" place. Therefore, if we look for
a file in that proper place and don't find it, we can assume it's not
on any other subvolume and *safely* skip the global (broadcast to all)
lookup.

Change-Id: Id6ce4593ba1f7daffa74cfab591cb45960629ae3
BUG: 1220064
Reviewed-on-master: http://review.gluster.org/#/c/7702/
Signed-off-by: Jeff Darcy 
Signed-off-by: Shyam 
Reviewed-on: http://review.gluster.org/10729
Tested-by: Gluster Build System 
Reviewed-by: Krishnan Parthasarathi 
Reviewed-by: Vijay Bellur

Restore build on non Linux systems

2015-05-07T18:52:10+00:00

This change broke the build on NetBSD, FreeBSD, and MacOS X:
http://review.gluster.org/10526/

We restore the build with two fixes:
- Use POSIX-compliant sysconf(_SC_NPROCESSORS_ONLN) to get the
  number of processors, instead of Linux specific get_nprocs().
  That let us remove Linux-specific #include 
- Only define MAX() if it is not already defined. NetBSD defines
  it in  which is already included

Backport of: I62341c670598670e47ea2f69ab94864f96588b18
BUG: 1212676
Signed-off-by: Emmanuel Dreyfus 

Change-Id: I0f098153e76954bb85b5dca3f054a069e31dd94c
Reviewed-on: http://review.gluster.org/10653
Reviewed-by: Shyamsundar Ranganathan 
Reviewed-by: Kaleb KEITHLEY 
Tested-by: Vijay Bellur

dht/rebalance: Throttle rebalance

2015-05-07T17:26:54+00:00

Throttle value will be "normal" by default. For throttling down,
a thread will be put in to sleep. And for throttling up,
gf_defrag_process_dir will wake up the sleeping threads.

Change-Id: I4892ab14982a1ff305aeb2d8bbd33c79d6877b69
BUG: 1219579
Signed-off-by: Susant Palai 
Reviewed-on: http://review.gluster.org/10526
Reviewed-by: Raghavendra G 
Tested-by: Raghavendra G 
Reviewed-on: http://review.gluster.org/10629
Reviewed-by: Vijay Bellur 
Tested-by: Vijay Bellur

rebalance: Introducing local crawl and parallel migration

2015-05-07T09:37:02+00:00

The current patch address two part of the design proposed.
1. Rebalance multiple files in parallel
2. Crawl only bricks that belong to the current node

Brief design explanation for the above two points.

1. Rebalance multiple files in parallel:
   -------------------------------------
The existing rebalance engine is single threaded. Hence, introduced
multiple threads which will be running parallel to the crawler. The
current rebalance migration is converted to a "Producer-Consumer"
frame work.

Where Producer is : Crawler
      Consumer is : Migrating Threads

Crawler: Crawler is the main thread. The job of the crawler is now
limited to fix-layout of each directory and add the files which are
eligible for the migration to a global queue in a round robin manner
so that we will use all the disk resources efficiently. Hence, the
crawler will not be "blocked" by migration process.

Producer: Producer will monitor the global queue. If any file is
added to this queue, it will dqueue that entry and migrate the file.
Currently 20 migration threads are spawned at the beginning of the
rebalance process. Hence, multiple file migration happens in parallel.

2. Crawl only bricks that belong to the current node:
   --------------------------------------------------
As rebalance process is spawned per node, it migrates only the files
that belongs to it's own node for the sake of load balancing. But it
also reads entries from the whole cluster, which is not necessary as
readdir hits other nodes.

New Design:
        As part of the new design the rebalancer decides the subvols
that are local to the rebalancer node by checking the node-uuid of
root directory prior to the crawler starts. Hence, readdir won't hit
the whole cluster  as it has already the context of local subvols and
also node-uuid request for each file can be avoided. This makes the
rebalance process "more scalable".

Change-Id: I6f1b44086a09df8ca23935fd213509c70cc0c050
BUG: 1217381
Signed-off-by: Susant Palai 
Reviewed-on: http://review.gluster.org/10466
Tested-by: Gluster Build System 
Tested-by: NetBSD Build System
Reviewed-by: N Balachandran

dht : coverity fixes

2015-04-07T10:52:05+00:00

CID : 1124352,1124365 (unchecked return value),
      1124377 ( logically dead code),
      1124511 (null dereference)

Change-Id: I61e029a078559cfe15d36bf0aa53418f6214e5cb
BUG: 789278
Signed-off-by: Manikandan Selvaganesh 
Reviewed-on: http://review.gluster.org/9622
Tested-by: Gluster Build System 
Reviewed-by: Shyamsundar Ranganathan 
Reviewed-by: Kaleb KEITHLEY 
Reviewed-by: N Balachandran 
Reviewed-by: Vijay Bellur

Avoid conflict between contrib/uuid and system uuid

2015-04-04T17:48:35+00:00

glusterfs relies on Linux uuid implementation, which
API is incompatible with most other systems's uuid. As
a result, libglusterfs has to embed contrib/uuid,
which is the Linux implementation, on non Linux systems.
This implementation is incompatible with systtem's
built in, but the symbols have the same names.

Usually this is not a problem because when we link
with -lglusterfs, libc's symbols are trumped. However
there is a problem when a program not linked with
-lglusterfs will dlopen() glusterfs component. In
such a case, libc's uuid implementation is already
loaded in the calling program, and it will be used
instead of libglusterfs's implementation, causing
crashes.

A possible workaround is to use pre-load libglusterfs
in the calling program (using LD_PRELOAD on NetBSD for
instance), but such a mechanism is not portable, nor
is it flexible. A much better approach is to rename
libglusterfs's uuid_* functions to gf_uuid_* to avoid
any possible conflict. This is what this change attempts.

BUG: 1206587
Change-Id: I9ccd3e13afed1c7fc18508e92c7beb0f5d49f31a
Signed-off-by: Emmanuel Dreyfus 
Reviewed-on: http://review.gluster.org/10017
Tested-by: Gluster Build System 
Reviewed-by: Niels de Vos

DHT: move regular expression log message to DEBUG

2015-04-02T10:24:01+00:00

Change-Id: I8652208aa2c3a600816c911a9e8af557c67d37c4
BUG: 1197585
Signed-off-by: vmallika 
Reviewed-on: http://review.gluster.org/9777
Tested-by: Gluster Build System 
Reviewed-by: Shyamsundar Ranganathan 
Reviewed-by: Raghavendra G 
Tested-by: Raghavendra G

Gfdb Query Fix and Volume option fix

2015-03-30T20:39:20+00:00

1) Query fix in find_changed_with_freq()
2) Volume option typo fix for write_freq_threshold
   and read_freq_threshold

Change-Id: I38e154818178aab412b2d7b2914cd29acef66ffb
BUG: 1207343
Signed-off-by: Joseph Fernandes 
Reviewed-on: http://review.gluster.org/10050
Reviewed-by: Dan Lambright 
Tested-by: Gluster Build System 
Reviewed-by: Niels de Vos

cluster/dht: Add tier translator.

2015-03-21T16:50:29+00:00

The tier translator shares most of DHT's code. It differs in how
subvolumes are chosen for I/Os, and how file migration (cache promotion
and demotion) is managed. That different functionality is split to either
DHT or tier logic according to the "tier_methods" structure.

A cache promotion and demotion thread is created in a manner
similar to the rebalance daemon. The thread operates a timing
wheel which periodically checks for promotion and demotion candidates
(files). Candidates are queued and then migrated. Candidates must exist on
the same node as the daemon and meet other critera per caching policies.

This patch has two authors (Dan Lambright and Joseph Fernandes). Dan
did the DHT changes and Joe wrote the cache policies. The fix depends on
DHT readidr changes and the database library which have been submitted
separately.  Header files in libglusterfs/src/gfdb should be reviewed in
patch 9683.

For more background and design see the feature page [1].

[1]
http://www.gluster.org/community/documentation/index.php/Features/data-classification

Change-Id: Icc26c517ccecf5c42aef039f5b9c6f7afe83e46c
BUG: 1194753
Signed-off-by: Dan Lambright 
Reviewed-on: http://review.gluster.org/9724
Reviewed-by: Vijay Bellur 
Tested-by: Vijay Bellur

cluster/dht: Change the subvolume encoding in d_off to be a "global"

2015-03-18T11:47:41+00:00

position in the graph rather than relative (local) to a particular
translator.

Encoding the volume in this way allows a single translator to manage
which brick is currently being scanned for directory entries. Using a
single translator minimizes allocated bits in the d_off. It also allows
multiple DHT translators in the same graph to have a common frame of
reference (the graph position) for which brick is being read. Multiple
DHT translators are needed for the Tiering feature.

The fix builds off a previous change (9332) which removed subvolume
encoding from AFR. The fix makes an equivalent change to the EC
translator.

More background can be found in fix 9332 and gluster-dev discussions [1].

DHT and AFR/EC are responsibile (as before) for choosing which brick to
enumerate directory entries in over the readdir lifecycle.

The client translator receiving the readdir fop encodes the dht_t. It
is referred to as the "leaf node" in the graph and corresponds to the
brick being scanned.

When DHT decodes the d_off, it translates the leaf node to a local
subvolume, which represents the next node in the graph leading to
the brick.

Tracking of leaf nodes is done in common utility functions. Leaf nodes
counts and positional information are updated on a graph switch.

[1] www.gluster.org/pipermail/gluster-devel/2015-January/043592.html

Change-Id: Iaf0ea86d7046b1ceadbad69d88707b243077ebc8
BUG: 1190734
Signed-off-by: Dan Lambright 
Reviewed-on: http://review.gluster.org/9688
Reviewed-by: Xavier Hernandez 
Reviewed-by: Krishnan Parthasarathi 
Reviewed-by: Vijay Bellur 
Tested-by: Vijay Bellur