RepoMirrors/ceph
1
0
Fork 0
mirror of https://github.com/ceph/ceph synced 2024-12-28 06:23:08 +00:00
Code Issues Packages Projects Releases Wiki Activity
59,280 Commits 211 Branches 394 Tags 1.4 GiB
2bb9f9bcb3
Commit Graph

59280 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Sage Weil
20ee727792 Merge pull request #11213 from liewegas/wip-osd-fastinfo 
osd: add a pg _fastinfo attribute to reduce per-io metadata updates

Reviewed-by: Samuel Just <sjust@redhat.com>
 2016-10-18 21:39:58 -05:00
Sage Weil
0aeea8941b Merge pull request #11354 from liewegas/wip-rocksdb 
rocksdb: update to latest, and make it the default for the mons
 2016-10-18 21:38:53 -05:00
Sage Weil
d30ee64fc7 Merge pull request #11500 from liewegas/wip-lttng 
CMakeLists.txt: enable LTTNG by default

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2016-10-18 21:38:40 -05:00
Michal Jarzabek
b06b188e6b mds/SessionMap.h: remove unneeded function 
Signed-off-by: Michal Jarzabek <stiopa@gmail.com>
 2016-10-18 21:59:03 +01:00
Michal Jarzabek
be8bb2ad54 mds/SessionMap: add const to member functions 
Signed-off-by: Michal Jarzabek <stiopa@gmail.com>
 2016-10-18 21:55:14 +01:00
Sage Weil
ad2dde108b Merge pull request #11540 from liewegas/wip-bluestore-cow 
os/bluestore: default bluestore_clone_cow=true
 2016-10-18 15:53:37 -05:00
Sage Weil
dcbe9f05d4 Merge pull request #11539 from liewegas/wip-bluestore-gc 
os/bluestore: no garbage collection for uncompressed blobs

Reviewed-by: Sage Weil <sage@redhat.com>
 2016-10-18 15:53:25 -05:00
Sage Weil
369244314d os/bluestore: default bluestore_clone_cow=true 
This has been stable for a long time.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 16:53:09 -04:00
Sage Weil
108724b7eb Revert "ceph_test_objecstore: use 1TB block device" 
This reverts commit 5389133ff9.

This is too slow with the bitmap allocator.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 16:50:28 -04:00
Roushan Ali
ef8989ce74 os/bluestore: No garbage collection for uncompressed blobs 
Signed-off-by: Roushan Ali<roushan.ali@sandisk.com>
 2016-10-18 16:49:50 -04:00
Sage Weil
9cd59b27c5 cls_lock: encode client-facing addrs with client features 
This struct goes back to the client, not to disk.  This was done
incorrectly in e40dd920dc.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 16:40:12 -04:00
Sage Weil
ffee7155b5 objclass: expose client features to cls ops 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 16:40:12 -04:00
Sage Weil
f0cf281900 message/MMonProbe: reencode monmap for non-addr2 clients 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 16:40:12 -04:00
Sage Weil
1893315baa message/MMonMap: reencode MonMap for old (non-addr2) clients 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 16:40:12 -04:00
Sage Weil
5e51cea658 messages/MMDSMap: reencode MDSMap for old (non-addr2) clients 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 16:40:12 -04:00
Sage Weil
01824b7b11 ceph-dencoder: parse ull args 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 16:40:12 -04:00
Sage Weil
e6b78a2070 messages/MOSDMap: reencode if missing MSG_ADDR2 feature 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 16:40:11 -04:00
Sage Weil
c67d204635 msg: set addr type when creating from scratch 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 16:40:11 -04:00
Sage Weil
a9a8d05f81 cls/lock: set test entity_addr_t types 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 16:40:11 -04:00
Sage Weil
b7e1f76b0f test/librbd/exclusive_lock: use entity_addr_t::parse, not inet_pton 
Otherwise the entity_addr_t type doesn't get set.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 16:40:11 -04:00
Zhao Junwang
30b162c5e3 test_addrs: adding some addr unittest cases 
Some unittest need a new constructor, add a
entity_addr_t(__u32 _type, __u32 _nonce).

Signed-off-by: Zhao Junwang <zhjwpku@gmail.com>
 2016-10-18 16:40:11 -04:00
Zhao Junwang
27d736a531 msg: add a decode_legacy_addr_after_marker helper 
This can eliminates duplication between entity_addr_t::decode()
and entity_addrvec::decode().

Signed-off-by: Zhao Junwang <zhjwpku@gmail.com>
 2016-10-18 16:40:11 -04:00
Zhao Junwang
72096f6a34 msg: add entity_addrvec_t 
Signed-off-by: Zhao Junwang <zhjwpku@gmail.com>
 2016-10-18 16:40:11 -04:00
Sage Weil
fb5332a0a5 msg/msg_types: add TYPE_MSGR2 
Unused so far.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 16:40:11 -04:00
Sage Weil
0d02a188ea msg/msg_types: add type accessors 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 16:40:10 -04:00
Sage Weil
3561eb380a msg/msg_types: prefix addr with type; - for none 
Previously entityt_addr_t() rendered as ":/0".  Now it is
"-".

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 16:40:10 -04:00
Sage Weil
19f48b35f7 msg/async: fix bind 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 16:40:10 -04:00
Sage Weil
1a417a23ab msg/simple: fix bind 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 16:40:10 -04:00
Zhao Junwang
6a7fe5ae6c msg: change entity_addr_t encode/decode 
Signed-off-by: Zhao Junwang <zhjwpku@gmail.com>
 2016-10-18 16:40:10 -04:00
Zhao Junwang
36fb92d858 features: define the new ADDR2 feature 
Signed-off-by: Zhao Junwang <zhjwpku@gmail.com>
 2016-10-18 16:40:10 -04:00
Zhao Junwang
0b4813aebd msg: remove some useless 'break' 
Signed-off-by: Zhao Junwang <zhjwpku@gmail.com>
 2016-10-18 16:40:10 -04:00
Sage Weil
2eea741e04 msg/async: fix learned_addr 
Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 16:40:10 -04:00
Sage Weil
9ebd56502f msg/simple: fix learned_addr 
There is a type in entity_addr_t that needs to get set
too.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 16:40:10 -04:00
Sage Weil
ae0d6eb1c0 mon/OSDMonitor: health warn if require_{jewel,kraken} flags aren't set 
We want to prompt users to set these flags as soon as their
upgrades complete.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 16:40:09 -04:00
Sage Weil
d9baaa1d6b Merge pull request #11536 from liewegas/wip-fullbit-dump 
mds: fix EMetaBlob::fullbit xattr dump

Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
 2016-10-18 13:43:17 -05:00
Sage Weil
15acca0bc5 Merge pull request #11533 from ifed01/wip-kstore-rmcoll-fix 
os/kstore: rmcoll fix to satisfy store_test

Reviewed-by: Sage Weil <sage@redhat.com>
 2016-10-18 13:22:11 -05:00
Sage Weil
325aff551c mds: fix EMetaBlob::fullbit xattr dump 
- show xattr names!
- bound the values

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 14:03:17 -04:00
Loic Dachary
469a53a4ad tests: ceph-disk: force debug monc = 0 
The sh function will collect both stderr and stdout and debug
will mess the json parsing.

Fixes: http://tracker.ceph.com/issues/17607

Signed-off-by: Loic Dachary <ldachary@redhat.com>
 2016-10-18 17:33:23 +02:00
Sage Weil
e234d8bec3 Merge pull request #11525 from wjwithagen/wip-wjw-freebsd-getopt 
tools/ceph-monstore-update-crush.sh: FreeBSD getopt is not compatible…

Reviewed-by: Sage Weil <sage@redhat.com>
 2016-10-18 09:45:24 -05:00
Sage Weil
c65f8d8d83 Merge pull request #11524 from wjin/filestore_perfcounter 
os/filestore: clean filestore perfcounters

Reviewed-by: Sage Weil <sage@redhat.com>
 2016-10-18 09:41:54 -05:00
Sage Weil
c4d774003f Merge pull request #11518 from umullan/uday 
doc: fix broken links
 2016-10-18 09:31:54 -05:00
Sage Weil
2a8ee42023 Merge pull request #11522 from wjwithagen/wip-wjw-freebsd-readable.sh 
FreeBSD:test/encoding/readable.sh": fix nproc and ls -v calls

Reviewed-by: Sage Weil <sage@redhat.com>
 2016-10-18 09:30:06 -05:00
Sage Weil
c82f30d823 Merge pull request #11528 from jcsp/wip-vstart-1osd 
vstart: decrease pool size if <3 OSDs

Reviewed-by: Sage Weil <sage@redhat.com>
 2016-10-18 09:28:26 -05:00
Sage Weil
bd407ce81f osd/PG: add perfcounters for pg info updates 
Note info updates (of any kind), fastinfo updates,
and biginfo updates.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 10:16:13 -04:00
Igor Fedotov
ebf81e3039 os/bluestore: proper set resulting pnext to avoid returning local var 
Signed-off-by: Igor Fedotov <ifedotov@mirantis.com>
 2016-10-18 13:50:17 +00:00
Igor Fedotov
5b36a93fd7 os/kstore: fix KStore rm_coll handler to behave similar to bluestore one 
Signed-off-by: Igor Fedotov <ifedotov@mirantis.com>
 2016-10-18 13:47:00 +00:00
Sage Weil
d698ab3e34 ceph-objectstore-tool: relax cur_struct_v checks 
All compat pg formats can handle past interval removal
and mark-complete.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 09:26:31 -04:00
Sage Weil
8e86de6078 osd: make fastinfo apply conditionally 
This way we don't have to bother deleting the fastinfo
attr, which will reduce the amount of work the
underlying store has to do.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 09:26:31 -04:00
Sage Weil
1d5dc29a13 osd: add _fastinfo PG attr for common pg_info_t updates 
For most IO operations we only update a handful of fields
in the pg_info_t structure.  However, the full struct,
when encoded, is on the order of 800 bytes.

This adds a new attribute, _fastinfo, which contains only
the most commonly updated fields.  When present, the
fastinfo fields should be overlayed on top of the full
info struct contained in the existing info attr.  If
a field outside of the "fast" set is updated, we clear
the fastinfo attribute and update the full info attr.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 09:26:31 -04:00
Sage Weil
203a721298 osd/PG: pass last_writen_info to _prepare_write_info 
We'll use this shortly.

Signed-off-by: Sage Weil <sage@redhat.com>
 2016-10-18 09:26:30 -04:00