RepoMirrors/ceph
1
0
Fork 0
mirror of https://github.com/ceph/ceph synced 2025-01-01 08:32:24 +00:00
Code Issues Packages Projects Releases Wiki Activity
92,938 Commits 212 Branches 394 Tags 1.5 GiB
8cebc1c0ff
Commit Graph

92938 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Kefu Chai
df0fa55c35
Merge pull request #25424 from rzarzynski/wip-bl-c_str_tracker 
common: hint bufferlist's buffer_track_c_str accordingly.

Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Casey Bodley <cbodley@redhat.com>
 2018-12-10 15:17:43 +08:00
Kefu Chai
c147b2c163
Merge pull request #25439 from p-na/pna-wip-cleanup 
mgr/dashboard: Remove unused code

Reviewed-by: Stephan Müller <smueller@suse.com>
 2018-12-10 15:16:09 +08:00
Tang Junhui
a509b5719b mds: do not call Journaler::_trim twice 
Journaler::_trim is called in the routine:
PurgeQueue::_execute_item
==>Journaler::write_head
==>Journaler::_finish_write_head
==>Journaler::_trim
so do not register callback in PurgeQueue::_execute_item
to call Journaler::_trim again.

Signed-off-by: Tang Junhui <tangjunhui@sangfor.com>
Fixes: http://tracker.ceph.com/issues/37566
 2018-12-09 10:14:54 +08:00
Tang Junhui
ca61ce4485 mds: fix incorrect l_pq_executing_ops statistics when meet an invalid item in purge queue 
l_pq_executing_ops should subtract the ops we added previous
in the condition we meet an invalid item in the purge queue,
otherwise the statistics would always show some ops
in l_pq_executing_ops.

Signed-off-by: Junhui Tang <tangjunhui@sangfor.com.cn>
Fixes: http://tracker.ceph.com/issues/37567
 2018-12-09 10:12:32 +08:00
Tang Junhui
0fdfc0c5bc client: do not move f->pos untill success write 
writes maybe failed in Client::_write, so do not move f->pos when write
successfully.

Signed-off-by: Junhui Tang <tangjunhui@sangfor.com.cn>
Fixes: https://tracker.ceph.com/issues/37546
 2018-12-09 10:09:37 +08:00
Sage Weil
cc9968dacc Merge PR #25425 into master 
* refs/pull/25425/head:
	test/common/blkdev-udevadm-info-samples: add a few test cases
	ceph-volume: Micron SSDs don't include vendor name in ID_SERIAL
	common/blkdev: micron SSDs don't include vendor name in ID_SERIAL
	ceph-volume: update get_device_id to match in-tree implementation
	common/blkdev: improve device_id generation

Reviewed-by: Jan Fajerski <jfajerski@suse.com>
 2018-12-08 11:16:57 -06:00
Mehdi Abaakouk
2d7ed8ad15 doc: fix typo in mgr ssl setup 
Unlike `config set`, `config-key set` does not take the <who>
argument.

This change removes it from mgr ssl setup.

Signed-off-by: Mehdi Abaakouk <sileht@sileht.net>
 2018-12-08 15:11:35 +01:00
songweibin
d2c1541d02 cls/rbd: silence the log of get metadata error 
When getting the qos of an image without qos set,
there is a lot of printing in the OSD log file.
```
2018-12-07 15:23:51.122 7f01de695700  0 <cls> /home/swb/ceph/ceph-dev/src/cls/rbd/cls_rbd.cc:3728: error get metadata: (2) No such file or directory
```

Signed-off-by: songweibin <song.weibin@zte.com.cn>
 2018-12-08 15:08:22 +08:00
Jos Collin
6672ffef2b
Merge pull request #25312 from falcon78921/wip-doc-37451 
doc: broken link on troubleshooting-mon page

Reviewed-by: Jos Collin <jcollin@redhat.com>
 2018-12-08 11:12:50 +05:30
James McClune
ca05e45696 doc: broken link on troubleshooting-mon page 
Fixes: https://tracker.ceph.com/issues/37451

This fix involves adding :ref: labels to the add-or-rm-mons.rst
page. I also added :ref: labels for other headings within
add-or-rm-mons.rst (for future reference).

Signed-off-by: James McClune <jmcclune@mcclunetechnologies.net>
 2018-12-07 23:15:43 -05:00
Kefu Chai
c1660ad10c
Merge pull request #25410 from dragonylffly/rb 
tools/rados: fix memory leak in error path

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2018-12-08 11:57:20 +08:00
Kefu Chai
a7608b175f
Merge pull request #25393 from Aran85/fix-agent-state 
osd: avoid using null agent_state

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2018-12-08 11:54:54 +08:00
Kefu Chai
057060fc13
Merge pull request #25370 from majianpeng/sharedlru-addfunc 
common/shared_cache: bumps it to the front of the LRU if key existed.

Reviewed-by: Greg Farnum <gfarnum@redhat.com>
 2018-12-08 11:47:59 +08:00
Patrick Donnelly
115a2645f1
Merge PR #25275 into master 
* refs/pull/25275/head:
	osdc/Journaler: make try_read_entry store data in contiguous memory

Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
 2018-12-07 10:57:36 -08:00
Patrick Donnelly
4b9327bfca
Merge PR #25270 into master 
* refs/pull/25270/head:
	mds: use uniform write handler for PQ

Reviewed-by: Zheng Yan <zyan@redhat.com>
 2018-12-07 10:56:25 -08:00
Patrick Donnelly
7518b52c31
Merge PR #24991 into master 
* refs/pull/24991/head:
	mds: disallow dumping huge caches to formatter

Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Reviewed-by: Zheng Yan <zyan@redhat.com>
 2018-12-07 10:54:46 -08:00
Patrick Donnelly
9789f37fa5
Merge PR #24940 into master 
* refs/pull/24940/head:
	qa: add test for getfattr ceph.dir.pin
	client: support getfattr ceph.dir.pin extended attribute

Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
 2018-12-07 10:53:44 -08:00
Alfredo Deza
d3c0058911 ceph-volume tests.functional.batch symlink test_zap to all batch scenarios 
Signed-off-by: Alfredo Deza <adeza@redhat.com>
 2018-12-07 13:52:44 -05:00
Patrick Donnelly
4432aa5f26
Merge PR #24748 into master 
* refs/pull/24748/head:
	qa: use 6h timeout for pjd test

Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
 2018-12-07 10:50:57 -08:00
Alfredo Deza
25843b7629 ceph-volume tests.functional.batch add test_zap yaml to tox.ini 
Signed-off-by: Alfredo Deza <adeza@redhat.com>
 2018-12-07 12:30:27 -05:00
Alfredo Deza
6b8f0d483a ceph-volume tests.functional.batch create a separate test_zap playbook 
Signed-off-by: Alfredo Deza <adeza@redhat.com>
 2018-12-07 12:30:27 -05:00
Alfredo Deza
73bb17f79a ceph-volume lvm.zap update success message for OSD IDs 
Signed-off-by: Alfredo Deza <adeza@redhat.com>
 2018-12-07 12:30:27 -05:00
Alfredo Deza
acbf7c9b2a doc/man/ceph-volume add zapping by osd-id examples 
Signed-off-by: Alfredo Deza <adeza@redhat.com>
 2018-12-07 12:30:27 -05:00
Alfredo Deza
f82304f4b5 doc/ceph-volume add zapping by osd-id examples 
Signed-off-by: Alfredo Deza <adeza@redhat.com>
 2018-12-07 12:30:27 -05:00
Nathan Cutler
a58a29e0bf
Merge pull request #25347 from smithfarm/wip-37485 
build/ops: rpm: make ceph-grafana-dashboards own its directories

Reviewed-by: Tim Serong <tserong@suse.com>
 2018-12-07 18:22:29 +01:00
Casey Bodley
358fd01385
Merge pull request #25414 from ivancich/wip-rgw-user-stats-mismatch 
rgw: fix bad user stats on versioned bucket after reshard

Reviewed-by: Matt Benjamin <mbenjami@redhat.com>
 2018-12-07 11:53:30 -05:00
Sage Weil
eef391d380 test/common/blkdev-udevadm-info-samples: add a few test cases 
The erwan1 one unfortunately has no model information :(.  The other two
work with the current implementation, although erwan.v1.sdb has
"Shared_PERC8" for the model, and I suspect the Shared_ prefix is not
part of the real model?

Signed-off-by: Sage Weil <sage@redhat.com>
 2018-12-07 09:50:13 -06:00
Sage Weil
3e4a638c93 ceph-volume: Micron SSDs don't include vendor name in ID_SERIAL 
Signed-off-by: Sage Weil <sage@redhat.com>
 2018-12-07 09:50:13 -06:00
Sage Weil
3a4ebc60ed common/blkdev: micron SSDs don't include vendor name in ID_SERIAL 
According to Ryan Meredith <rmeredith@micron.com>, the NVMes begin with
MTFD, but their other devices use the standard Micron_$model_$serial.

Signed-off-by: Sage Weil <sage@redhat.com>
 2018-12-07 09:50:13 -06:00
Sage Weil
c967dbf81f ceph-volume: update get_device_id to match in-tree implementation 
Signed-off-by: Sage Weil <sage@redhat.com>
 2018-12-07 09:50:13 -06:00
Sage Weil
e9c3441b68 common/blkdev: improve device_id generation 
A few interesting cases:

1. autriche (my desktop)

$ sudo udevadm info /dev/nvme0n1 | grep ID_
E: ID_PART_TABLE_TYPE=gpt
E: ID_PART_TABLE_UUID=c83d5616-676b-4667-bcf3-c82fd4fc7e64
E: ID_SERIAL=Samsung SSD 960 EVO 250GB_S3ESNX0J958081E
E: ID_SERIAL_SHORT=S3ESNX0J958081E

- no ID_MODEL or ID_VENDOR
-> use ID_SERIAL

2. my dev box

gnit:~ (master) 09:09 AM $ udevadm info /dev/nvme0n1 | grep ID_
E: ID_FS_TYPE=xfs
E: ID_FS_USAGE=filesystem
E: ID_FS_UUID=860d4503-9c9d-4c24-af09-4266b7717a5c
E: ID_FS_UUID_ENC=860d4503-9c9d-4c24-af09-4266b7717a5c
E: ID_MODEL=INTEL SSDPEDMD400G4
E: ID_PATH=pci-0000:82:00.0-nvme-1
E: ID_PATH_TAG=pci-0000_82_00_0-nvme-1
E: ID_SERIAL=INTEL SSDPEDMD400G4_CVFT520200G7400BGN
E: ID_SERIAL_SHORT=CVFT520200G7400BGN
E: ID_WWN=nvme.8086-43564654353230323030473734303042474e-494e54454c205353445045444d443430304734-00000001

- no ID_VENDOR
-> ID_MODEL + ID_SERIAL_SHORT

gnit:~ (master) 09:12 AM $ udevadm info /dev/sda | grep ID_
E: ID_ATA=1
E: ID_ATA_DOWNLOAD_MICROCODE=1
E: ID_ATA_FEATURE_SET_HPA=1
E: ID_ATA_FEATURE_SET_HPA_ENABLED=1
E: ID_ATA_FEATURE_SET_PM=1
E: ID_ATA_FEATURE_SET_PM_ENABLED=1
E: ID_ATA_FEATURE_SET_SECURITY=1
E: ID_ATA_FEATURE_SET_SECURITY_ENABLED=0
E: ID_ATA_FEATURE_SET_SECURITY_ENHANCED_ERASE_UNIT_MIN=2
E: ID_ATA_FEATURE_SET_SECURITY_ERASE_UNIT_MIN=2
E: ID_ATA_FEATURE_SET_SECURITY_FROZEN=1
E: ID_ATA_FEATURE_SET_SMART=1
E: ID_ATA_FEATURE_SET_SMART_ENABLED=1
E: ID_ATA_ROTATION_RATE_RPM=0
E: ID_ATA_SATA=1
E: ID_ATA_SATA_SIGNAL_RATE_GEN1=1
E: ID_ATA_SATA_SIGNAL_RATE_GEN2=1
E: ID_ATA_WRITE_CACHE=1
E: ID_ATA_WRITE_CACHE_ENABLED=1
E: ID_BUS=ata
E: ID_MODEL=INTEL_SSDSC2BB240G4
E: ID_MODEL_ENC=INTEL\x20SSDSC2BB240G4\x20\x20\x20\x20\x20\x20\x20\x20\x20\x20\x20\x20\x20\x20\x20\x20\x20\x20\x20\x20\x20
E: ID_PART_TABLE_TYPE=dos
E: ID_PART_TABLE_UUID=bb35118c
E: ID_PATH=pci-0000:00:1f.2-ata-1
E: ID_PATH_TAG=pci-0000_00_1f_2-ata-1
E: ID_REVISION=D2010355
E: ID_SERIAL=INTEL_SSDSC2BB240G4_BTWL3414034J240NGN
E: ID_SERIAL_SHORT=BTWL3414034J240NGN
E: ID_TYPE=disk
E: ID_WWN=0x55cd2e404b4e47d8
E: ID_WWN_WITH_EXTENSION=0x55cd2e404b4e47d8

- no ID_VENDOR
-> ID_MODEL + ID_SERIAL_SHORT

3. mira lab machine (old areca controller in JBOD mode, I think)

root@mira055:~# udevadm info /dev/sdb | grep ID_
E: ID_BUS=scsi
E: ID_MODEL=HUS724040ALA640
E: ID_MODEL_ENC=HUS724040ALA640\x20
E: ID_PART_TABLE_TYPE=gpt
E: ID_PART_TABLE_UUID=957b2db6-de5c-46cb-a672-243fa12d55b2
E: ID_PATH=pci-0000:01:00.0-scsi-0:0:0:1
E: ID_PATH_TAG=pci-0000_01_00_0-scsi-0_0_0_1
E: ID_REVISION=R001
E: ID_SCSI=1
E: ID_SCSI_SERIAL=PN1334PBH5JMJS
E: ID_SERIAL=2001b4d2058da3a00
E: ID_SERIAL_SHORT=001b4d2058da3a00
E: ID_TYPE=disk
E: ID_VENDOR=HGST
E: ID_VENDOR_ENC=HGST\x20\x20\x20\x20

- ID_VENDOR and ID_MODEL
- ID_MODEL doesn't include vendor name + _
- ID_SERIAL and _SHORT are junk
- ID_SCSI_SERIAL has the serial number!  wth
-> ID_VENDOR + ID_MODEL + ID_SCSI_SHORT

Added a bunch of udevadm output samples.

So,

- if ID_VENDOR + ID_MODEL + ID_SCSI_SERIAL are present, use them.
- if ID_MODEL + ID_SERIAL_SHORT are present, use them
- if ID_SERIAL is present, use it.
- fail

Signed-off-by: Sage Weil <sage@redhat.com>
 2018-12-07 09:50:13 -06:00
Mykola Golub
bc28c3af4d pybind/rbd: optimize rbd_list2 
If rbd_list2 returns ERANGE, it sets num_images to the value required.
But it is better to retry with a bigger number, to avoid new retries
if images are added to the pool at that time.

Also the initial value for num_images looks too small.

Signed-off-by: Mykola Golub <mgolub@suse.com>
 2018-12-07 17:18:19 +02:00
Nathan Cutler
c1083b2f4f build/ops: rpm: make ceph-grafana-dashboards own its directories on SUSE 
This is only required to get the spec file to build in the OpenSUSE
Build Service (OBS). Also, this change could potentially make the
package impossible to install together with grafana (if the latter
owns the same directories with different ownership/permissions).
Therefore, make the change specific to SUSE.

Fixes: http://tracker.ceph.com/issues/37485
Signed-off-by: Nathan Cutler <ncutler@suse.com>
Signed-off-by: Tim Serong <tserong@suse.com>
 2018-12-07 16:16:42 +01:00
Jan Fajerski
c138daa794 pybind/mgr/status: fix ceph fs status in py3 environments. 
Fixes: http://tracker.ceph.com/issues/37573

Signed-off-by: Jan Fajerski <jfajerski@suse.com>
 2018-12-07 16:14:20 +01:00
Nathan Weinberg
731b3e4938 Added breadcrumb tests to Block menu items 
Fixes: https://tracker.ceph.com/issues/37294
Signed-off-by: Nathan Weinberg <nweinber@redhat.com>
 2018-12-07 09:17:51 -05:00
Patrick Nawracay
e704c1b4aa mgr/dashboard: Remove unused code 
Signed-off-by: Patrick Nawracay <pnawracay@suse.com>
 2018-12-07 09:34:51 +00:00
Mykola Golub
af96e16271
Merge pull request #25415 from dillaman/wip-rbd-cli-integration 
test/cli-integration/rbd: added new parent image attributes

Reviewed-by: Mykola Golub <mgolub@suse.com>
 2018-12-07 10:19:24 +02:00
Jan Fajerski
73caf0a816
Merge pull request #25234 from jan--f/prometheus-add-iface-objstore-metadata 
mgr/prometheus: add interface and objectstore to osd metadata
 2018-12-07 09:13:23 +01:00
Jianpeng Ma
9c73dcbc98 osd/OSD: preallocate for _get_pgs/_get_pgids to avoid reallocate. 
num_pgs is the most accurate value fore reserve when call _get_pgs &
_get_pgids.

Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
 2018-12-07 11:13:09 +08:00
Kefu Chai
b03c2b4a17
Merge pull request #19454 from ifed01/wip-ifed-pool-sizes 
osd: per-pool osd stats collection

Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>
 2018-12-07 10:35:23 +08:00
hsiang41
5381d6910b mgr/diskprediction_cloud: Remove needless library in the requirements file 
Remove needless library in the requirements file.
Fixes:http://tracker.ceph.com/issues/37533

Signed-off-by: Rick Chen <rick.chen@prophetstor.com>
 2018-12-07 10:07:37 +08:00
Shiyang Ruan
b3632a2357 TokenBucketThrottle: Add name for TokenBucketThrottle. 
In order to specific which one is failed when applying the limit.

Signed-off-by: Shiyang Ruan <ruansy.fnst@cn.fujitsu.com>
 2018-12-07 09:54:15 +08:00
Sage Weil
811887edb1 Merge PR #25243 into master 
* refs/pull/25243/head:
	osdc/Objecter: update op_target_t::paused in _calc_target

Reviewed-by: Sage Weil <sage@redhat.com>
 2018-12-06 17:22:57 -06:00
Sage Weil
c1d757a7c6 Merge PR #25399 into master 
* refs/pull/25399/head:
	osd: requeue waiting peering events from deleted slots in unprime_split_children

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
 2018-12-06 17:22:23 -06:00
Jason Dillaman
237a23eb29
Merge pull request #25428 from trociny/wip-open-flags 
librbd: corrected usage of ImageState::open flag parameter

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
 2018-12-06 17:27:51 -05:00
Matt Benjamin
0702314909
Merge pull request #25430 from cbodley/wip-pr-22539 
rgw: rgwgc:process coredump in some special case
 2018-12-06 16:41:57 -05:00
sdnets
575a790066 rgw: rgwgc:process coredump in some special case。 Gc processes obja, objb, objc in order and pool of objb is deleted (obja and objc is in the same pool and pool exits). RGW will coredump as ctx->io_ctx_impl is an empty point during delete objc. 
Fixes:http://tracker.ceph.com/issues/23199

Signed-off-by: zhaokun <develop@hikdata.com>
 2018-12-06 16:15:22 -05:00
Radoslaw Zarzynski
576a41c16a common: eradicate buffer_track_c_str in bufferlist. 
Signed-off-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
 2018-12-06 22:07:16 +01:00
Casey Bodley
a74f705336
Merge pull request #17789 from shashalu/fix-bucket-link-dm 
rgw: fix version bucket stats

Reviewed-by: Orit Wasserman <owasserm@redhat.com>
 2018-12-06 15:31:58 -05:00
Casey Bodley
d29b03dd11
Merge pull request #21622 from tianshan/misc_es_fix 
rgw: some fix for es sync

Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
 2018-12-06 15:31:18 -05:00