summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Revert "tools: add ceph_monstore_tool with getosdmap"wip-rbd-close-imageJosh Durgin2013-04-292-139/+0
| | | | This reverts commit 1e6c390a67b2500d5a934b6537b8c7e9e1f096c6.
* ObjectCacher: wait for all reads when stopping flusherJosh Durgin2013-04-292-2/+23
| | | | | | | | | | | | | Stopping the flusher is essentially the shutdown step for the ObjectCacher - the next thing is actually destroying it. If we leave any reads outstanding, when they complete they will attempt to use the now-destroyed ObjectCacher. This is particularly a problem with rbd images, since an -ENOENT can instantly complete many readers, so the upper layers don't wait for the other rados-level reads of that object to finish before trying to shutdown the cache. Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
* Merge branch 'wip-mon-pg' into nextSage Weil2013-04-295-10/+53
|\ | | | | | | Reviewed-by: Samuel Just <sam.just@inktank.com>
| * mon: remap creating pgs on startupSage Weil2013-04-295-3/+25
| | | | | | | | | | | | | | | | | | After Monitor::init_paxos() has loaded all of the PaxosService state, we should then map creating pgs to osds. This ensures we do so after the osdmap has been loaded and the pgs actually map somewhere meaningful. Fixes: #4675 Signed-off-by: Sage Weil <sage@inktank.com>
| * mon: only map/send pg creations if osdmap is definedSage Weil2013-04-291-4/+8
| | | | | | | | | | | | | | | | | | | | | | | | This avoids calculating new pg creation mappings if the osdmap isn't loaded yet, which currently happens when during Monitor::paxos_init() on startup. Assuming osdmap epoch is nonzero, it should always be safe to do this (although possibly unnecessary). More cleanup here is certainly possible, but this is one step toward fixing the bad behavior for #4675. Signed-off-by: Sage Weil <sage@inktank.com>
| * mon: factor map_pg_creates() out of send_pg_creates()Sage Weil2013-04-292-5/+22
|/ | | | | | | Factor out the portion of the function that remaps creating pgs to osds from the part that sends those pending creates out. Signed-off-by: Sage Weil <sage@inktank.com>
* Merge branch 'wip_4836' into nextSamuel Just2013-04-295-23/+173
|\ | | | | | | | | Fixes: #4836 Reviewed-by: Sage Weil <sage@inktank.com>
| * OSDMonitor: when adding bucket, delay response if pending map has nameSamuel Just2013-04-262-15/+32
| | | | | | | | | | Fixes: #4836 Signed-off-by: Samuel Just <sam.just@inktank.com>
| * PaxosService: use get and put for version_tSamuel Just2013-04-261-8/+2
| | | | | | | | | | | | | | Otherwise, we just duplicate the logic for generating the version key names. Signed-off-by: Samuel Just <sam.just@inktank.com>
| * tools: add ceph_monstore_tool with getosdmapSamuel Just2013-04-262-0/+139
| | | | | | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* | mon: Fix leak of contextYan, Zheng2013-04-282-3/+5
| | | | | | | | | | | | | | Use Context::complete() to finish context, it frees the context after executing Context::finish(). Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
* | doc: Removed extra whitespace.John Wilkins2013-04-281-1/+1
| | | | | | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* | doc: Added rbd-fuse to TOC.John Wilkins2013-04-281-0/+1
| | | | | | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* | Added commentary and removed fourth column for now.John Wilkins2013-04-281-9/+20
| | | | | | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* | doc: Removed. Redunant information now.John Wilkins2013-04-281-23/+0
| | | | | | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* | doc: Added openssh-server mention, corrections, hyperlink fix.John Wilkins2013-04-281-2/+13
| | | | | | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* | doc: Added openssh-server mention.John Wilkins2013-04-281-0/+9
| | | | | | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* | doc: Added manpage link and hidden TOC.John Wilkins2013-04-281-2/+8
| | | | | | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* | doc: Removed installed Chef. This is now in the ceph wiki.John Wilkins2013-04-281-286/+0
| | | | | | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* | doc: Removed text for include directive. Wasn't behaving the way I'd hoped.John Wilkins2013-04-281-108/+0
| | | | | | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* | doc: Added ceph-mds to CephFS toc.John Wilkins2013-04-281-0/+1
| | | | | | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* | doc: Fix. ceph, not chef.John Wilkins2013-04-271-3/+3
| | | | | | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* | ceph-filestore-dump: fix warnings on i386 buildSage Weil2013-04-271-3/+3
|/ | | | | | | | | | | tools/ceph-filestore-dump.cc: In member function ‘int header::get_header()’: warning: tools/ceph-filestore-dump.cc:454:19: comparison between signed and unsigned integer expressions [-Wsign-compare] tools/ceph-filestore-dump.cc: In member function ‘int footer::get_footer()’: warning: tools/ceph-filestore-dump.cc:471:19: comparison between signed and unsigned integer expressions [-Wsign-compare] tools/ceph-filestore-dump.cc: In member function ‘int super_header::read_super()’: warning: tools/ceph-filestore-dump.cc:697:30: comparison between signed and unsigned integer expressions [-Wsign-compare] Signed-off-by: Sage Weil <sage@inktank.com>
* ceph.spec.in: remove conditional checks on tcmallocGary Lowell2013-04-261-14/+4
| | | | | | tcmalloc is available on all supported platforms now. Signed-off-by: Gary Lowell <gary.lowell@inktank.com>
* debian/rules: Fix tcmalloc breakageGary Lowell2013-04-261-8/+0
| | | | | | | | Since all currently supported platforms have tcmalloc available and it is now the default, remove broken check code that turns it off if the package is not listed in build-depends. Signed-off-by: Gary Lowell <gary.lowell@inktank.com>
* mon: cache osd epochsSage Weil2013-04-262-7/+44
| | | | | | | | | | | | | | The monitor may get a series of messages from the OSD that prompt it to send incremental maps (pg_temp updates, failures, probably more). Avoid sending the same incremental maps twice by keeping a cache of what epochs we think the OSDs have. This reduces monitor load, especially when the mon is a bit behind and is getting a stream of delayed messages, and the work associated with sending the inc maps prevents it from catching up. Signed-off-by: Sage Weil <sage@inktank.com> Reviewed-by: Greg Farnum <greg@inktank.com>
* monitor: assert out early if we get our own sync_start backGreg Farnum2013-04-261-0/+7
| | | | Signed-off-by: Greg Farnum <greg@inktank.com>
* mon: update assert for looser requirementsGreg Farnum2013-04-261-1/+1
| | | | | Signed-off-by: Greg Farnum <greg@inktank.com> Reviewed-by: Sage Weil <sage@inktank.com>
* doc: Deleted old index. Generates warnings otherwise.John Wilkins2013-04-261-9/+0
| | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* doc: General purpose pre-flight checklist.John Wilkins2013-04-261-0/+115
| | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* doc: Modified Ceph deployment landing page.John Wilkins2013-04-261-13/+32
| | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* doc: Added general pre-flight checklist for ceph-deploy.John Wilkins2013-04-261-0/+108
| | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* doc: Removed old ceph-deploy placeholder.John Wilkins2013-04-261-5/+0
| | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* doc: Removed Chef section. Now appears in new Ceph wiki.John Wilkins2013-04-261-262/+0
| | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* doc: Added Key Management for ceph-deploy.John Wilkins2013-04-261-0/+28
| | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* doc: Added "Add/Remove Monitors" section for ceph-deploy.John Wilkins2013-04-261-0/+53
| | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* doc: Added Create a Cluster section.John Wilkins2013-04-261-0/+65
| | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* doc: Added ceph-deploy package management (install | uninstall ) section.John Wilkins2013-04-261-0/+47
| | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* doc: Added new quick start preamble and index.John Wilkins2013-04-261-5/+64
| | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* doc: Added ceph-deploy preflight.John Wilkins2013-04-261-0/+137
| | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* doc: Added ceph-deploy quick start.John Wilkins2013-04-261-0/+179
| | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* Merge branch 'wip-4822' into nextDavid Zafman2013-04-261-6/+5
|\ | | | | | | Reviewed-by: Sam Just <sam.just@inktank.com>
| * osd: Fix logic in OSDMap::containing_subtree_is_down()David Zafman2013-04-261-6/+5
| | | | | | | | | | | | | | | | Check for up OSDs as we walk up the crushmap hierarchy fixes: #4822 Signed-off-by: David Zafman <david.zafman@inktank.com>
* | monitor: squash signed/unsigned comparison warningGreg Farnum2013-04-261-1/+1
| | | | | | | | | | | | | | This is a safe range to do comparisons against, and we compare against the signed rank inside the loop. Signed-off-by: Greg Farnum <greg@inktank.com>
* | mon: use brute force to find a sync provider if our first one failsGreg Farnum2013-04-261-0/+11
| | | | | | | | | | | | | | | | | | | | We try and select a random monitor first, but if that fails we should make sure that nobody's available before asserting. Fixes #4812 Signed-off-by: Greg Farnum <greg@inktank.com> Reviewed-by: Sage Weil <sage@inktank.com>
* | Merge branch 'wip-4760' into nextYehuda Sadeh2013-04-268-38/+43
|\ \
| * | rgw: fix compilation for certain architecturesYehuda Sadeh2013-04-261-1/+1
| | | | | | | | | | | | | | | | | | Casting. Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
| * | rgw: fix list buckets limitYehuda Sadeh2013-04-262-3/+11
| | | | | | | | | | | | | | | | | | | | | | | | There was an issue when limit was being set, we didn't break from the iterating loop if limit was reached. Also, S3 does not enforce any limit, so keep that behavior. Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
| * | rgw: fix bucket listing when reaching limitYehuda Sadeh2013-04-261-1/+1
| | | | | | | | | | | | | | | | | | Bucket listing was broken when limit was set. Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
| * | rgw: swift list containers can return 204Yehuda Sadeh2013-04-266-8/+12
| | | | | | | | | | | | | | | | | | | | | In order to keep compatibility with swift, if a plain formatter is being used, we should return 204 when there are no containers. Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
View Lorry Controller Status