| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This fix introduces infrastructure to support different
policies for promotion and demotion.
Currently the tier feature automatically promotes and demotes
files periodically based on access. This is good for testing
but too stringent for most real workloads. It makes it
difficult to fully utilize a hot tier- data will be demoted
before it is touched- its unlikely a 100GB hot SSD will have
all its data touched in a window of time.
A new parameter "mode" allows the user to pick promotion/demotion
polcies.
The "test mode" will be used for *.t and other general testing.
This is the current mechanism.
The "cache mode" introduces watermarks. The watermarks
represent levels of data residing on the hot tier.
"cache mode" policy:
The % the hot tier is full is called P.
Do not promote or demote more than D MB or F files.
A random number [0-100] is called R.
Rules for migration:
if (P < watermark_low) don't demote, always promote.
if (P >= watermark_low) && (P < watermark_hi) demote if R < P; promote if R > P.
if (P > watermark_hi) always demote, don't promote.
gluster volume set {vol} cluster.watermark-hi %
gluster volume set {vol} cluster.watermark-low %
gluster volume set {vol} cluster.tier-max-mb {D}
gluster volume set {vol} cluster.tier-max-files {F}
gluster volume set {vol} cluster.tier-mode {test|cache}
Change-Id: I157f19667ec95aa1d53406041c1e3b073be127c2
BUG: 1257911
Signed-off-by: Dan Lambright <dlambrig@redhat.com>
Reviewed-on: http://review.gluster.org/12039
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The script did not cleanly unmount/mount gluster and change the current
working directory when stopping and starting the volume. Most of the
time this problem would self-resolve before subsequent tests, but
very occasionally races would lead to the errors/failures.
Change-Id: I128b913a71e2745512ee81c3d71852311e3b4a1b
BUG: 1270328
Signed-off-by: Dan Lambright <dlambrig@redhat.com>
Reviewed-on: http://review.gluster.org/12327
Reviewed-by: Joseph Fernandes
Tested-by: Gluster Build System <jenkins@build.gluster.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Re-enable tier.t in automatic tests. Disable check
for BSD until recurring problem with SQLlite on it is understood.
Change-Id: Ib13b269ab841a59a0a41d8478c8627b180b16c61
BUG: 1231268
Signed-off-by: Dan Lambright <dlambrig@redhat.com>
Reviewed-on: http://review.gluster.org/12208
Tested-by: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: mohammed rafi kc <rkavunga@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
unlink fails with invalid argument for files that
are being present on cold tier, before attaching.
All of the fops will be hashed to hot_tier after
attach-tier (unless explicitly set the "rule"
option). Lookups sent to directory, will eventually
search the directory using readdirp, and will
populate inode_ctx for the inodes based on the output,
in respective dht_xlators. So the readdirp will
populate inodes_ctx for the files (that is already
present in volume before attaching) in cold-dht
only because it got the entries from the cold-tier.
So when an unlink comes on such an inode, the lookup
associated with the unlink will be send as a
re validate request to cold-tier only, since
already a lookup was performed on the inode,
and the new lookup will succeed. So from the
unlink of dht, it will hash to cold-tier but the
cached_subvol will be cold, since there is a
mismatch in hash and cach , it chose hashed
subvolume and will sent the fop to hot dht,
and the fops fail with EINVAL from the hot-dht
since it does not have inode_ctx stored for
that inode (because, no lookup was performed
from hot-dht).
Change-Id: Ib7c14a9297a22d615f7a890a060be4809b5a745a
BUG: 1236032
Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
Signed-off-by: Dan Lambright <dlambrig@redhat.com>
Reviewed-on: http://review.gluster.org/11675
Tested-by: NetBSD Build System <jenkins@build.gluster.org>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently the tier feature piggy backs off the rebalance command
syntax to obtain status and this is clumsy. Introduce a new
tier command that can do tier specific operations, starting
with volume status to display counters.
Old commands:
gluster volume attach-tier <vol> [replica count] {bricklist..}
gluster volume detach-tier <vol> {start|stop|commit}
New commands:
gluster volume tier <vol> attach [replica count] {bricklist} |
detach {start|stop|commit} |
status
Change-Id: Ic07b3c6260588162de7d34380f8cbd3d8a7f35d3
BUG: 1255693
Signed-off-by: Dan Lambright <dlambrig@redhat.com>
Reviewed-on: http://review.gluster.org/11984
Reviewed-by: Gaurav Kumar Garg <ggarg@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Posix acls on a linkto file change the file's permission
bits and cause DHT to treat it as a non-linkto file.This
happens on the migration failure of a file on which posix
acls were set.
The fix prevents posix acls from being set on a linkto
file and copies them across only after a file has
been successfully migrated.
Change-Id: Iccf7ff6fba49fe05d691d9b83bf76a240848b212
BUG: 1247563
Signed-off-by: Nithya Balachandran <nbalacha@redhat.com>
Signed-off-by: N Balachandran <nbalacha@redhat.com>
Reviewed-on: http://review.gluster.org/12025
Tested-by: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Tier daemon should always run with tier volume. If volume
is stopped and started again, we manually need to start
the tier-daemon, instead this patch will automatically trigger
tier process along with volume start.
A snapshot restored volume will not have node_state_info,
so we need to create and store it dynamically
Change-Id: I659387c914bec7a1b6929ee5cb61f7b406402075
BUG: 1238593
Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
Signed-off-by: Dan Lambright <dlambrig@redhat.com>
Reviewed-on: http://review.gluster.org/11525
Tested-by: NetBSD Build System <jenkins@build.gluster.org>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The check_counters function contained "grep -o '[0-9*]'", which
was in error. This patch corrects it to "grep -o '[0-9]*'".
This fix was necessary to accommodate for double-digit counters.
Change-Id: Idaa09de4403bf66e741176a7377eba264819ca3b
BUG: 1252121
Signed-off-by: Pamela Ousley <pousley@redhat.com>
Reviewed-on: http://review.gluster.org/11877
Reviewed-by: Dan Lambright <dlambrig@redhat.com>
Tested-by: NetBSD Build System <jenkins@build.gluster.org>
Tested-by: Dan Lambright <dlambrig@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The md5sum fingerprints were not correctly compared after moving
files between the hot and cold tiers.
This version of tier.t uses a new function, "check_counters", to
ensure that the number of promotions/demotions is as expected.
This is intended to avoid spurious timing-related errors that were
seen with the old script.
Change-Id: I4a0ae7315493bfd307a0f68f21fa3ea33c88b08f
BUG: 1231268
Signed-off-by: Pamela Ousley <pousley@redhat.com>
Reviewed-on: http://review.gluster.org/11285
Tested-by: NetBSD Build System <jenkins@build.gluster.org>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Dan Lambright <dlambrig@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
On a graph switch, a new xlator and private structures are
created. The tier migration daemon must stop using the
old xlator and private structures and begin using the
new ones. Otherwise, when RPCs arrive (such as counter
queries from glusterd), the new xlator will be consulted
but it will not have up to date information. The fix
detects a graph switch and exits the daemon in this
case. Typical graph switches for the tier case would
be turning off performance translators.
Change-Id: Ibfbd4720dc82ea179b77c81b8f534abced21e3c8
BUG: 1226005
Signed-off-by: Dan Lambright <dlambrig@redhat.com>
Reviewed-on: http://review.gluster.org/11372
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When a subvol goes down, tier daemon stopped immediately, and
the status shows as "Progressing".
With this change, with respect to tier xlator, when a subvol
goes offline it will update the status as failed.
Change-Id: I9f722ed0d35cda8c7fc1a7e75af52222e2d0fdb7
BUG: 1227803
Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
Reviewed-on: http://review.gluster.org/11068
Tested-by: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Dan Lambright <dlambrig@redhat.com>
Tested-by: Dan Lambright <dlambrig@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Volume set option related to tier volume can only be set
for tier volume, also currently all volume set i for tier
option accepts a non-negative integer. This patch validate
both condition.
Change-Id: I3611af048ff4ab193544058cace8db205ea92336
BUG: 1216960
Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
Signed-off-by: Dan Lambright <dlambrig@redhat.com>
Reviewed-on: http://review.gluster.org/10751
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Tested-by: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Joseph Fernandes
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
1) Avoid hangs on unmounting NFS on NetBSD
NetBSD umount(8) on a NFS mount whose server is gone will wait forever
because umount(8) calls realpath(3) and tries to access the mount before
it calls unmount(2). The non-portable, NetBSD-specific umount -R flag
prevent that behavior.
We therefore introduce UMOUNT_F, defined as "umount -f" on Linux and
"umount -f -R" on NetBSD to take care of forced unmounts, especially
in the NFS case.
2) Enforce usage of force_umount wrapper with timeout
Whenever umount is used it should be wrapped in force_umount with
tiemout handling. That saves us timing issues, and it handles the
NetBSD NFS case.
3) Cleanup kernel cache flush.
We used (cd $M0 && umount $M0 ) as a portable kernel cache flush
trick, but it does not flush everything we need on Linux. Introduce
a drop_cache() shell function that reverts to previously used
echo 3 > /proc/sys/vm/drop_caches on Linux, and keeps
(cd $M0 && umount $M0 ) on other systems.
BUG: 1129939
Change-Id: Iab1f5a023405f1f7270c42b595573702ca1eb6f3
Signed-off-by: Emmanuel Dreyfus <manu@netbsd.org>
Reviewed-on: http://review.gluster.org/11114
Tested-by: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Kaleb KEITHLEY <kkeithle@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When creating client volfiles, xlator tier-dht will
be loaded for each volume. So for services like nfs
have one or more volumes . So for each volume in the
graph a tier-dht xlator will be created. So the graph
parser will fail because of the redundant node in
graph.
By this change tier-dht will be renamed as volname-tier-dht
Change-Id: I3c9b9c23ddcb853773a8a02be7fd8a5d09a7f972
BUG: 1222840
Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
Reviewed-on: http://review.gluster.org/10820
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Tested-by: NetBSD Build System
Reviewed-by: Kaushal M <kaushal@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Fix for handling cli output for attach-tier and
detach-tier
Change-Id: I4d17f4b09612754fe1b8cec6c2e14927029b9678
BUG: 1211562
Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
Reviewed-on: http://review.gluster.org/10284
Reviewed-by: Dan Lambright <dlambrig@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Tested-by: NetBSD Build System
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
On Linux, kernel cache can be flushed using
echo 3 > /proc/sys/vm/drop_caches
This non-portable approach can be replaced by an on-purpose
failed attempt to unmount: if the mount point is the current
directory and umount is called, the kernel will flush inodes
until it realize it cannot complete the operation because
root of filesystem is busy:
( cd $M0 ; umount $M0 )
Unfortunately this does not flush everything. Entries may
still be present in the kenrel FUSE cache. Using $GFS to
mount the filesystem ensure --entry-timeout=0 and clears
this problem.
Some stall information may also remain in glusterfs caches,
and that may have to be adressed by appropriate volume option.
For instance tests/bugs/rpc/bug-954057.t needs to disable
performance.stat-prefetch. Qtherwise, root's new credentials
are not evaluated after root-quash is enabled. The test could
also be done with performance.stat-prefetch enabled using
various tricks: copying the file to read, creating a hard link
on it, or just waiting long enough for metadata cache to expire.
BUG: 1129939
Change-Id: I54929e899d55c04dcd9d947809133549f01fd0e1
Signed-off-by: Emmanuel Dreyfus <manu@netbsd.org>
Reviewed-on: http://review.gluster.org/10411
Reviewed-by: Dan Lambright <dlambrig@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
After attaching tier, we have to start tier rebalance process.
This patch is to trigger tier start along with attch-tier.
Change-Id: I39380f95123f0087a82213ef263f9f33adcc5adc
BUG: 1214222
Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
Reviewed-on: http://review.gluster.org/10363
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Dan Lambright <dlambrig@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Right now, attach-tier calls parsing function for add-brick.
Add-brick does not have any check for brick count and replca
count compatibility.
Change-Id: I44ec13eadffc003a9ebf8c4eb0193df559933a68
BUG: 1215122
Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
Reviewed-on: http://review.gluster.org/10428
Reviewed-by: Dan Lambright <dlambrig@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
These commands work in a manner analagous to rebalancing when removing a
brick. The existing migration daemon detects "detach start" and switches
to moving data off the hot tier. While in this state all lookups are
directed to the cold tier.
gluster v detach-tier <vol> start
gluster v detach-tier <vol> commit
The status and stop cli commands shall be submitted separately.
Change-Id: I24fda5cc3ba74f5fb8aa9a3234ad51f18b80a8a0
BUG: 1205540
Signed-off-by: Dan Lambright <dlambrig@redhat.com>
Signed-off-by: root <root@localhost.localdomain>
Signed-off-by: Dan Lambright <dlambrig@redhat.com>
Reviewed-on: http://review.gluster.org/10108
Reviewed-by: Kaleb KEITHLEY <kkeithle@redhat.com>
Tested-by: NetBSD Build System
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We did not set up the graph properly for hot tiers with replicated
subvolumes. Also add check that the file has not already been moved
by another replicated brick on the same node.
Change-Id: I9adef565ab60f6774810962d912168b77a6032fa
BUG: 1206517
Signed-off-by: Dan Lambright <dlambrig@redhat.com>
Reviewed-on: http://review.gluster.org/10054
Reviewed-by: Joseph Fernandes <josferna@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Kaleb KEITHLEY <kkeithle@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch resolves tiering translator issues taken from the list
in bug 1203776. These issues have been selected to be fixed
first. The rest will be fixed in a subsequent patch (or are not a
problem).
3. Replace hardcoded #defines of promote/demote file names
6. Use loc_wipe() in migrate_using_query_file()
9. Only promote/demote files on the same node on which they reside.
14. Replace calloc with GF_CALLOC in tier.c and ensure freeing done
properly.
15. Handle if parse_query_str fails
22. Only load gfdb library on server side, remove SQL references
from client.
Change-Id: I6563b11e58ab2e4c6b1ce44db755781ad6d930fb
BUG: 1203776
Signed-off-by: Dan Lambright <dlambrig@redhat.com>
Reviewed-on: http://review.gluster.org/9987
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: N Balachandran <nbalacha@redhat.com>
Reviewed-by: Niels de Vos <ndevos@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Need to wait for a few seconds for rebalancing to complete
before stopping volume.
Change-Id: Ib81c02645240e7d74ebfb3e31ccbc612fc77b119
BUG: 1194753
Signed-off-by: Dan Lambright <dlambrig@redhat.com>
Reviewed-on: http://review.gluster.org/9966
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Niels de Vos <ndevos@redhat.com>
|
|
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 <dlambrig@redhat.com>
Reviewed-on: http://review.gluster.org/9724
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
Tested-by: Vijay Bellur <vbellur@redhat.com>
|