summaryrefslogtreecommitdiffstats
path: root/src/mgr/MgrClient.cc (follow)
Commit message (Collapse)AuthorAgeFilesLines
...
* mgr/MgrClient: do not attempt to access a global variable for configJason Dillaman2017-07-251-5/+4
| | | | Signed-off-by: Jason Dillaman <dillaman@redhat.com>
* mgr: allow/track service registrationsSage Weil2017-07-091-5/+77
| | | | Signed-off-by: Sage Weil <sage@redhat.com>
* mgr/MgrClient: skip sending stats if no sessionSage Weil2017-06-151-1/+1
| | | | Signed-off-by: Sage Weil <sage@redhat.com>
* msg,common: do not include assert.h everywhere to clobber <assert.h>Kefu Chai2017-06-081-3/+0
| | | | | | | | * reverts c93daaf92b643a06a822558973d377733dbd9d85 * add "#include "include/assert.h" to where <cassert> or <assert.h> are included, or avoid including them if possible. Signed-off-by: Kefu Chai <kchai@redhat.com>
* Merge pull request #14866 from chardan/jfw-wip-halflife_atomic_t-varanKefu Chai2017-06-081-3/+6
|\ | | | | | | | | common,test: migrate atomic_t to std::atomic Reviewed-by: Kefu Chai <kchai@redhat.com>
| * compile fixups for doutJesse Williamson2017-06-021-0/+3
| | | | | | | | Signed-off-by: Jesse Williamson <jwilliamson@suse.de>
| * mgr: fixes compiler error because of atomic_t to <atomic>Jesse Williamson2017-06-021-3/+3
| | | | | | | | Signed-off-by: Jesse Williamson <jwilliamson@suse.de>
* | osd: stop sending pgstat if luminousKefu Chai2017-06-021-2/+6
|/ | | | | | | | we don't send pgstat to mon from osd once we set CEPH_OSDMAP_REQUIRE_LUMINOUS. and cleanup the pgstat queue when we do the switch. Signed-off-by: Kefu Chai <kchai@redhat.com>
* mgr: Mark session connections down on shutdownBrad Hubbard2017-05-241-1/+4
| | | | | | | | | Also assert in OSD::ms_get_authorizer since it is dangerous to proceed if we are shutting down. Fixes: http://tracker.ceph.com/issues/19900 Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
* mgr/MgrClient: fix reconnect event leakSage Weil2017-04-161-0/+5
| | | | Signed-off-by: Sage Weil <sage@redhat.com>
* mgr/MgrClient: assume missing MgrMap means no acces to mgr at allSage Weil2017-03-291-2/+4
| | | | | | | | If we get as far as authenticating and have no MgrMap that implies the mon didn't provide us one (despite our request) and we have no access to the mgr at all. Signed-off-by: Sage Weil <sage@redhat.com>
* mgr/MgrClient: cope with disappearing perf_countersSage Weil2017-03-291-33/+33
| | | | | | | | | Perfcounters can go away; deal with it. This collapses into a single loop, but it means that the mgr might stop getting certain counters without explicit notification. Signed-off-by: Sage Weil <sage@redhat.com>
* mgr/MgrClient: do not assert on control-cSage Weil2017-03-291-0/+4
| | | | Signed-off-by: Sage Weil <sage@redhat.com>
* mgr/MgrClient: move lock init into .hSage Weil2017-03-291-1/+0
| | | | Signed-off-by: Sage Weil <sage@redhat.com>
* mgr/MgrClient: throttle connection attemptsSage Weil2017-03-291-27/+50
| | | | | | Max of 1 per second. Signed-off-by: Sage Weil <sage@redhat.com>
* mgr/MgrClient: remove unused cond signal helpersSage Weil2017-03-291-16/+0
| | | | Signed-off-by: Sage Weil <sage@redhat.com>
* mgr/MgrClient: don't block command submission if no session yetSage Weil2017-03-291-1/+0
| | | | Signed-off-by: Sage Weil <sage@redhat.com>
* mgr/MgrClient: resend pending commands on reconnectSage Weil2017-03-291-23/+12
| | | | Signed-off-by: Sage Weil <sage@redhat.com>
* mgr/MgrClient: use unique_ptr for MgrClient::sessionKefu Chai2017-02-281-7/+6
| | | | | | | | | | * use unique_ptr<MgrSessionState> for MgrClient::session, so we don't leak it. * also reset it in MgrClient::shutdown(), so the connection held by session can be released properly, and hence the MonClient will not be asked for an AuthAuthorizer after being shut down. Fixes: http://tracker.ceph.com/issues/19097 Signed-off-by: Kefu Chai <kchai@redhat.com>
* mgr: remove unnecessary C_StdFunctionJohn Spray2016-11-091-1/+1
| | | | | | FunctionContext works fine and already existed. Signed-off-by: John Spray <john.spray@redhat.com>
* mgr/MgrClient: implement con reset handlingSage Weil2016-10-031-67/+60
| | | | Signed-off-by: Sage Weil <sage@redhat.com>
* mgr/MgrClient: fix ms_handle_resetSage Weil2016-10-031-1/+1
| | | | | | Return false because we don't handle the reset. Signed-off-by: Sage Weil <sage@redhat.com>
* mgr: update for Dispatcher::ms_handle_refusedJohn Spray2016-09-291-0/+5
| | | | Signed-off-by: John Spray <john.spray@redhat.com>
* common: refactor CommandTableJohn Spray2016-09-291-17/+17
| | | | | | | Avoid handling out raw pointers to the ops, which are in fact owned by the table. Signed-off-by: John Spray <john.spray@redhat.com>
* mgrc: less vebose logging of messagesJohn Spray2016-09-291-2/+7
| | | | | | ...no need to tell us about all the things you're *not* handling. Signed-off-by: John Spray <john.spray@redhat.com>
* mgrc: mark down connections with dead sessionsJohn Spray2016-09-291-1/+4
| | | | Signed-off-by: John Spray <john.spray@redhat.com>
* mgr: expose "df" to python modulesJohn Spray2016-09-291-0/+1
| | | | Signed-off-by: John Spray <john.spray@redhat.com>
* mgrc: enable sending pgstatsJohn Spray2016-09-291-0/+6
| | | | Signed-off-by: John Spray <john.spray@redhat.com>
* mgrc: fix report sending after failoverJohn Spray2016-09-291-0/+1
| | | | | | | | The send_report loop doesn't get kicked off on configure messages unless the period is set to its initial value (0). Signed-off-by: John Spray <john.spray@redhat.com>
* mgrc: reset send_report timer on session changeJohn Spray2016-09-291-3/+15
| | | | Signed-off-by: John Spray <john.spray@redhat.com>
* mgr: flesh out standby/HAJohn Spray2016-09-291-41/+42
| | | | Signed-off-by: John Spray <john.spray@redhat.com>
* mgrc: create MgrClientJohn Spray2016-09-291-0/+318
The ceph-mgr equivalent to monclient Signed-off-by: John Spray <john.spray@redhat.com>
generated by cgit v1.2.3 (git 2.39.1) at 2025-01-20 00:18:01 +0100