Commit Graph

204 Commits

Author SHA1 Message Date
Patrick Donnelly
06c94de584
mds: support limiting cache by memory
This introduces two config parameters:

    mds_cache_memory_limit: Sets the soft maximum of the cache to the given
    byte count. (Like mds_cache_size, this doesn't actually limit the maximum
    size of the cache. It just dictates the steady-state size.)

    mds_cache_reservation: This replaces mds_health_cache_threshold everywhere
    except the Beacon heartbeat sent to the mons. The idea here is to specify a
    reservation of memory (5% by default) for operations and the MDS tries to
    always maintain that reservation. So, the MDS will recall caps from clients
    when it begins dipping into its reservation of memory.

mds_cache_size still limits the cache by Inode count but is now by-default 0
(i.e. unlimited). The new preferred way of specifying cache limits is by memory
size. The default is 1GB.

Fixes: http://tracker.ceph.com/issues/20594
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1464976

Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
2017-09-12 20:02:41 -07:00
Abhishek Lekshmanan
74801fda1c doc: 12.2.0 major release announcements
Reordered the RC releases sections back to their respective components,
added a ceph-mon section, added links to documentation wherever
possible, and a few forgotten RGW announcements. Also cleared up the
PendingReleaseNotes upto this point

Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>
2017-08-29 14:45:01 +02:00
Abhishek L
5a00b3905d Merge pull request #16782 from wjwithagen/wip-wjw-FreeBSD-Release
PendingReleaseNotes:Announce FreeBSD availability

Reviewed-By: Abhishek Lekshmanan <abhishek@suse.com>
Reviewed-By: Jason Dillaman <jdillaman@redhat.com>
2017-08-08 18:46:16 +02:00
Patrick Donnelly
41849fdcd1
doc: remove accidental additions to release notes
Presumably this was caused by a bad rebase.

Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
2017-08-08 09:28:57 -07:00
Willem Jan Withagen
fcc2e134b9 PendingReleaseNotes:Announce FreeBSD availability
Signed-off-by: Willem Jan Withagen <wjw@digiware.nl>
2017-08-08 16:26:21 +02:00
Douglas Fuller
8f33e3bc2b PendingReleaseNotes: add note about df filtering
Signed-off-by: Douglas Fuller <dfuller@redhat.com>
2017-08-03 14:11:46 -04:00
Piotr Dałek
b57fb1eef1 PendingReleaseNotes: mention forced recovery
Mention forced recovery in release notes.

Signed-off-by: Piotr Dałek <piotr.dalek@corp.ovh.com>
2017-08-03 09:45:28 +02:00
Abhishek Lekshmanan
a80b99fd41 doc: 12.1.1 release notes add major changes
Also cleanup PendingReleasenotes to an empty file so that only newer
changes are tracked, adding the relevant section back to
RC1 where relevant. Moving all the RC1 announcements back to RC2, when
we go to 12.2.0 we'll collapse all of these back to the release
announcments

Signed-off-by: Abhishek Lekshmanan <alekshmanan@suse.com>
2017-08-02 17:41:13 +02:00
Sage Weil
e0bd667320 PendingReleaseNotes: note about put -> set
Signed-off-by: Sage Weil <sage@redhat.com>
2017-07-26 14:13:22 -04:00
Jason Dillaman
6b44cb6034 PendingReleaseNotes: added blurb for new RBD cap profiles
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2017-07-21 14:30:38 -04:00
Sage Weil
7832c53375 mon/PGMap: remove skewed utilizatoin warning
This has a few problems:

1- It does not do it's analysis over CRUSH rule roots/classes, which
means that an innocent user of classes will see skewed usage (bc hdds are
more full than ssds, say)

2- It does not take degraded clusters into account, which means the warning
will appear when a fresh OSD is added.

See http://tracker.ceph.com/issues/20730

Signed-off-by: Sage Weil <sage@redhat.com>
2017-07-21 11:58:08 -04:00
Wido den Hollander
dacd41e043
release notes: Add Zabbix ceph-mgr module to PendingReleaseNotes
Signed-off-by: Wido den Hollander <wido@42on.com>
2017-07-20 14:08:45 +02:00
Jason Dillaman
e5b9fd8142 PendingReleaseNotes: included details for the pool tags
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2017-07-19 13:13:17 -04:00
Yuri Weinstein
c95ab13384 Merge pull request #16368 from theanalyst/fix/rgw-reshard-pool-ns
rgw: use a namespace for rgw reshard pool for upgrades as well

Reviewed-by: Casey Bodley <cbodley@redhat.com>
Reviewed-by: Orit Wasserman <owasserm@redhat.com>
2017-07-19 08:42:36 -07:00
Nathan Cutler
060084f708 doc: PendingReleaseNotes: "ceph -w" behavior has changed drastically
Signed-off-by: Joao Eduardo Luis <joao@suse.de>
Signed-off-by: Nathan Cutler <ncutler@suse.com>
2017-07-19 16:13:17 +02:00
Abhishek Lekshmanan
5c8622499c PendingReleaseNotes: add a note on RGW's pool change for resharding
Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>
2017-07-17 15:52:48 +02:00
Sage Weil
dd61a7f737 Merge pull request #16189 from bassam/pr-msgr-bind-addr
mon: add support public_bind_addr option

Reviewed-by: Sage Weil <sage@redhat.com>
2017-07-16 21:26:23 -05:00
Bassam Tabbara
c0dc20533b docs: add docs for public_bind_addr
Signed-off-by: Bassam Tabbara <bassam.tabbara@quantum.com>
2017-07-14 10:41:49 -07:00
John Spray
1486397f31 ceph.in: filter out audit from ceph -w
...and add an optional --watch-channel argument.

Signed-off-by: John Spray <john.spray@redhat.com>
2017-07-14 11:35:18 -04:00
Sage Weil
e170405fd8 mon: remove Formatter arg to QuorumService::get_health()
This is used to dump extra weirdness to the health detail structured
output, but we are about to remove all of that in luminous.

Signed-off-by: Sage Weil <sage@redhat.com>
2017-07-12 12:51:30 -04:00
Sage Weil
6068e1856d mon/Monitor: move time sync status from 'ceph health' to 'time-sync-status' command
Signed-off-by: Sage Weil <sage@redhat.com>
2017-07-12 12:51:30 -04:00
Ilya Dryomov
891798d6e1 PendingReleaseNotes: add a note on choose_args encoding
It's unlikely that it's been used yet, but worth a note.

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
2017-06-30 14:39:57 +02:00
John Spray
ec09fc2f52 common: enable status module by default
It's still sort of awkward to prefix these commands
with "mgr tell" but this makes them at least
somewhat accessible to the average user.

Signed-off-by: John Spray <john.spray@redhat.com>
2017-06-27 06:53:33 -04:00
Sage Weil
1e7cd3550c Merge pull request #13683 from liewegas/wip-crush-rulesets
crush: detect and (usually) fix ruleset != rule id

Reviewed-by: Loic Dachary <ldachary@redhat.com>
2017-06-21 08:45:42 -05:00
Sage Weil
80e3b73803 crush: simplify osd_pool_default_crush_rule config
Make an incompat change here with a release note since
this only affects pool creation, a rare event, and folks
who have customized their configs (also rare).

Keep it simple: a config sets the default rule, or else we pick
the first TYPE_REPLICATED pool in the crush map.

Signed-off-by: Sage Weil <sage@redhat.com>
2017-06-19 19:09:06 -04:00
Sage Weil
96afb19378 mon: remove 'crush_ruleset' property
The 'crush_rule' name has been present since before hammer; users should
use that instead

Signed-off-by: Sage Weil <sage@redhat.com>
2017-06-19 19:09:06 -04:00
John Spray
62a466dc50 tools/cephfs: remove apply mode of cephfs-journal-tool
This is undocumented and untested -- it was something
written before and superceded by the "recover_dentries"
subcommand.  While we're at it, also
s/scavenge_dentries/recover_dentries/
internally.

Signed-off-by: John Spray <john.spray@redhat.com>
2017-06-15 18:06:17 -04:00
Sage Weil
e9973adf1f PendingReleaseNotes: notes on whiteouts vs pgnls
Signed-off-by: Sage Weil <sage@redhat.com>
2017-06-08 09:41:25 -04:00
Sage Weil
ca3d9e6a70 osd: reduce rados_max_object_size from 100 GB -> 128 MB
I'm not sure why we had this so huge!  128 MB is a pretty reasonable size.

Signed-off-by: Sage Weil <sage@redhat.com>
2017-06-07 18:07:23 -04:00
Sage Weil
7cacb701ad mon/PGMap: call requests blocked for 128x as long ERR not WARN
- rename the option (max -> warn)
- add an err_..._ratio multiplier
- switch to HEALTH_ERR once requests are blocked long enough
- make the error ratio high (default is 32*128s -> about an hour) so that
we don't trigger on a heavily loaded cluster.

Signed-off-by: Sage Weil <sage@redhat.com>
2017-06-07 12:16:58 -04:00
Abhishek Lekshmanan
f498aa5954 doc: PendingReleaseNotes move the 12.0.2 section of release notes
So that we can track whatever diff goes in after

Signed-off-by: Abhishek Lekshmanan <abhishek@suse.com>
2017-05-19 10:58:50 +02:00
Sage Weil
3c6755e036 PendingReleaseNotes: warning about 'osd rm ...' and #19119
See http://tracker.ceph.com/issues/19119

Signed-off-by: Sage Weil <sage@redhat.com>
2017-05-17 11:42:55 -04:00
John Spray
1eb9438bc9 Merge pull request #14636 from jcsp/wip-19691
osdc: remove journaler_allow_split_entries option

Reviewed-by: Yan, Zheng <zyan@redhat.com>
2017-04-24 22:12:35 +01:00
Sage Weil
630e778951 rados: out json 'df' values as numbers, not strings
Fixes: http://tracker.ceph.com/issues/15546
Signed-off-by: Sage Weil <sage@redhat.com
2017-04-20 16:54:33 -04:00
Josh Durgin
25e8e1a08e PendingReleaseNotes: note about ec overwrites
Signed-off-by: Josh Durgin <jdurgin@redhat.com>
2017-04-19 17:45:43 -07:00
Josh Durgin
930eae214c OSDMonitor: get stripe_width via stripe_unit in ec profile
With bluestore, making the smallest write match min_alloc_size avoids
write amplification. With EC pools this is the stripe unit, or
stripe_width / num_data_chunks. Rather than requiring people to divide
by k to get the smallest ec write, allow it to be specified directly
via stripe_unit. Store it in the ec profile so changing a monitor
config option isn't necessary to set it.

This is particularly important for ec overwrites since they allow random i/o
which should match bluestore's checksum granularity (aka min_alloc_size).

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
2017-04-19 17:45:43 -07:00
John Spray
b4c32b1b58 osdc: remove journaler_allow_split_entries option
This had been broken for some time, as since the new
JournalStream stuff, zero padding was no longer a valid
encoding.

Fixes: http://tracker.ceph.com/issues/19691
Signed-off-by: John Spray <john.spray@redhat.com>
2017-04-19 11:41:28 +01:00
John Spray
e05493fe54 PendingReleaseNotes: recent cephfs changes
Signed-off-by: John Spray <john.spray@redhat.com>
2017-04-13 12:27:07 -04:00
John Spray
c743ea7d00 mds: enable dirfrags by default in new filesystems
Signed-off-by: john Spray <john.spray@redhat.com>
2017-03-28 23:26:51 +01:00
John Spray
4b0654f26a Merge pull request #13628 from jcsp/wip-19075
mds: extend 'p' auth cap to cover all vxattr stuff

Reviewed-by: Yan, Zheng <zyan@redhat.com>
2017-03-15 21:29:48 +00:00
Casey Bodley
5877780a47 Merge pull request #13760 from zhangsw/fix-rgw-torrent-uri
rgw: get torrent request's parameter is not the same as amazon s3.

Reviewed-by: Casey Bodley <cbodley@redhat.com>
2017-03-09 09:53:27 -05:00
Zhang Shaowen
a0aa859e62 rgw: get torrent request's parameter is not the same as amazon s3.
Fixes: http://tracker.ceph.com/issues/19136

Signed-off-by: Zhang Shaowen <zhangshaowen@cmss.chinamobile.com>
2017-03-09 11:37:02 +08:00
John Spray
f788686d5c mds: extend 'p' auth cap to cover all vxattr stuff
Fixes: http://tracker.ceph.com/issues/19075
Signed-off-by: John Spray <john.spray@redhat.com>
2017-03-08 14:02:57 +00:00
Sage Weil
3f5269d8b5 Merge pull request #13323 from yehudasa/wip-18079-2
librados: use cursor for nobjects listing

Reviewed-by: Sage Weil <sage@redhat.com>
2017-03-07 08:41:08 -06:00
Sage Weil
9076e695fb osd: remove copy-get-classic
This hasn't been used since well before hammer.

Signed-off-by: Sage Weil <sage@redhat.com>
2017-02-23 11:15:43 -05:00
Sage Weil
ccc05005a1 PendingReleaseNotes: notes about object listing API removal
Signed-off-by: Sage Weil <sage@redhat.com>
2017-02-15 15:28:54 -08:00
Sage Weil
c1ff854f92 PendingReleaseNotes: notes about object listing API removal
Signed-off-by: Sage Weil <sage@redhat.com>
2017-02-08 21:51:22 -05:00
John Spray
1d9668cd0e PendingReleaseNotes: note the fuse fstab format change
Signed-off-by: John Spray <john.spray@redhat.com>
2017-02-04 20:19:44 +00:00
Sage Weil
923e7f5ce5 osd: do not send ENXIO on misdirected op by default
In practice this tends to get bubbled up the stack as an error on
the caller, and they usually do not handle it properly.  For example,
with librbd, this turns into EIO and break the VM.

Instead, this will manifest as a hung op on the client.  That is
also not ideal, but given that the root cause here is generally a
bug, it's not clear what else would be better.

We already log an error in the cluster log, so teuthology runs will
continue to fail.

Signed-off-by: Sage Weil <sage@redhat.com>
2017-01-31 15:14:59 -05:00
Sage Weil
d0740a3514 Merge branch 'master' into wip-wjw-empty-cluster_network 2017-01-23 21:16:09 -06:00