Activity
From 02/03/2019 to 03/04/2019
03/04/2019
- 11:29 PM Bug #38577 (Resolved): Messenger/MessengerTest.MissingServerIdenTest2/0 msg/async/AsyncMessenger....
- ...
- 10:10 PM Bug #38493: msg/async: connection race + winner fault can leave connection stuck at replacing for...
- Hmm, I thought Sage just fixed this bug, what's the exact sha1?
- 03:04 PM Backport #38571 (Need More Info): mimic: msg/simple: rados bench segv in ceph::buffer::list::iter...
- Sage wrote: "We need to think about how to backport this in the most non-disruptive way."
- 03:03 PM Backport #38571 (Rejected): mimic: msg/simple: rados bench segv in ceph::buffer::list::iterator_i...
- 03:04 PM Backport #38570 (Need More Info): luminous: msg/simple: rados bench segv in ceph::buffer::list::i...
- Sage wrote: "We need to think about how to backport this in the most non-disruptive way."
- 03:03 PM Backport #38570 (Rejected): luminous: msg/simple: rados bench segv in ceph::buffer::list::iterato...
- 02:32 PM Bug #22480 (Pending Backport): msg/simple: rados bench segv in ceph::buffer::list::iterator_impl:...
- We need to think about how to backport this in the most non-disruptive way.
- 02:30 PM Bug #38569 (Resolved): msg/async/AsyncConnection.cc: 319: FAILED ceph_assert(center->in_thread())
- ...
03/02/2019
- 05:01 AM Bug #38524: AsyncConnection: segmentation fault
- Same job failed in another run, so it looks reproducible: /ceph/teuthology-archive/pdonnell-2019-03-01_18:13:24-multi...
03/01/2019
- 03:26 PM Bug #37778 (Resolved): msg/async: mark_down vs accept race leaves connection registered
- 03:26 PM Backport #37896 (Resolved): mimic: msg/async: mark_down vs accept race leaves connection registered
- 03:25 PM Backport #37897 (Resolved): luminous: msg/async: mark_down vs accept race leaves connection regis...
- 11:50 AM Bug #38524: AsyncConnection: segmentation fault
- There's a strange behavior in the log just before the segfault.
The peer that is trying to connect to this MDS is ... - 04:31 AM Bug #38524 (Resolved): AsyncConnection: segmentation fault
- ...
- 10:00 AM Bug #38457: common/msg: sockaddr on FreeBSD differs from Linux, has sa_len
- Willem Jan Withagen wrote:
> Needs to be backported to
> Mimic
> Luminous
@Willem: There was no "Backport" fiel...
02/28/2019
- 09:04 PM Bug #22480 (Fix Under Review): msg/simple: rados bench segv in ceph::buffer::list::iterator_impl:...
- 02:30 PM Bug #22480: msg/simple: rados bench segv in ceph::buffer::list::iterator_impl::advance(), Pipe::r...
- yep, this made the failures go away:...
- 01:57 AM Bug #22480: msg/simple: rados bench segv in ceph::buffer::list::iterator_impl::advance(), Pipe::r...
- /a/sage-22480-b/3642573
looks like there was some rx_buffers activity on the connection right before it crashed.... - 07:29 PM Backport #37897: luminous: msg/async: mark_down vs accept race leaves connection registered
- Nathan Cutler wrote:
> https://github.com/ceph/ceph/pull/25956
merged
02/27/2019
- 06:49 AM Bug #38493 (Resolved): msg/async: connection race + winner fault can leave connection stuck at re...
- 2019-02-02 09:31:03.402291 7f5f4935e700 20 -- 100.100.7.118:6789/0 >> 100.100.7.122:6789/0 conn(0x55ec02821000 :6789 ...
02/26/2019
- 11:44 PM Bug #22480 (In Progress): msg/simple: rados bench segv in ceph::buffer::list::iterator_impl::adva...
- 11:44 PM Bug #22480: msg/simple: rados bench segv in ceph::buffer::list::iterator_impl::advance(), Pipe::r...
- it's always a standalone test, either thrash-eio or the newer thrash-backfill.
reproduces very easily, see http://...
02/25/2019
- 11:06 PM Bug #38355 (Resolved): msg/async: sendmsg points to unaddressable bytes in AsyncConnection::_try_...
- Pretty sure this is fixed by 774bd9d99ed9d9f0630dba9658d756812b5e3253
- 02:43 PM Bug #22480: msg/simple: rados bench segv in ceph::buffer::list::iterator_impl::advance(), Pipe::r...
- /a/sage-2019-02-24_19:27:53-rados-wip-sage-testing-2019-02-24-1127-distro-basic-smithi/3634199
- 02:42 PM Bug #22480: msg/simple: rados bench segv in ceph::buffer::list::iterator_impl::advance(), Pipe::r...
- /a/sage-2019-02-24_19:27:53-rados-wip-sage-testing-2019-02-24-1127-distro-basic-smithi/3634191
02/23/2019
- 11:54 AM Bug #38457: common/msg: sockaddr on FreeBSD differs from Linux, has sa_len
- Needs to be backported to
Mimic
Luminous - 11:39 AM Bug #38457 (New): common/msg: sockaddr on FreeBSD differs from Linux, has sa_len
On Linux the definition of a sockaddr is:
```
struct sockaddr {
sa_family_t sa_family;
char ...
02/22/2019
- 12:57 PM Bug #22480: msg/simple: rados bench segv in ceph::buffer::list::iterator_impl::advance(), Pipe::r...
- ...
02/21/2019
02/20/2019
- 11:35 PM Bug #37799 (Can't reproduce): msg/async: RESETSESSION due to connection reset during initial conn...
- Closing this one. Haven't seen it since we rewrote the v2 implementation. It's possible something liek this still e...
- 08:23 AM Bug #38393 (New): All the metrics exposed by ceph-metrics should contain cluster_id as label
- The cluster_id field is required in metrics labels, because in OCP world there are possibilities of having multiple c...
- 05:08 AM Bug #37292 (Resolved): RDMAStack - poll failed error when using IB and RDMACM
- 04:14 AM Bug #37292 (Fix Under Review): RDMAStack - poll failed error when using IB and RDMACM
- 04:03 AM Bug #38392 (Duplicate): msg/async: the poll needs to retry when it is interrupted by signal
- 03:11 AM Bug #38392 (Duplicate): msg/async: the poll needs to retry when it is interrupted by signal
- src/msg/async/rdma/RDMAStack.cc
From the implementation today, if the poll in the function RDMADispatcher::pollin... - 02:53 AM Bug #38391: msg async rdma: fix rdma exchange port, parse string bug
- https://github.com/ceph/ceph/pull/26525/
- 02:34 AM Bug #38391 (Resolved): msg async rdma: fix rdma exchange port, parse string bug
- In src/msg/async/rdma/Infiniband.cc
function *send_msg* encode struct IBSYNMsg to a string, and *recv_msg* parse t...
02/19/2019
- 02:25 PM Bug #24119: osd: leaked PipeConnection
- /a/sage-2019-02-18_23:58:27-rados-wip-sage-testing-2019-02-18-1341-distro-basic-smithi/3609170
osd.7, as usual
02/16/2019
- 03:38 PM Bug #24119: osd: leaked PipeConnection
- /a/sage-2019-02-15_22:48:00-rados:verify-wip-v2-leaks-distro-basic-smithi/3595938
- 03:34 PM Bug #38355 (Resolved): msg/async: sendmsg points to unaddressable bytes in AsyncConnection::_try_...
- ...
02/15/2019
02/13/2019
- 04:46 PM Bug #38216: "HEALTH_WARN 3 monitors have not enabled msgr2" in rados
- https://github.com/ceph/ceph/pull/26389
02/12/2019
02/11/2019
- 05:03 PM Bug #38247 (Fix Under Review): msg/async: v2 reconnect_seq assert due to interfering connections
- https://github.com/ceph/ceph/pull/26365
- 01:02 PM Bug #38216 (In Progress): "HEALTH_WARN 3 monitors have not enabled msgr2" in rados
- I have a big wip-v2-upgrade branch in progress that resolves all of the (many) upgrade problems.
02/09/2019
- 02:28 PM Bug #38247: msg/async: v2 reconnect_seq assert due to interfering connections
- easy fix:
- change client_ident to include who we *want* to connect to, so receiving end can ignore it early.
h... - 02:25 PM Bug #38247: msg/async: v2 reconnect_seq assert due to interfering connections
- The problem: We send a client_ident identifying *ourselves*, but not who we are trying to connect to. And on the ser...
- 01:14 PM Bug #38249 (Won't Fix): SimpleMessenger.cc: 260: FAILED ceph_assert(!cleared)
- ...
02/08/2019
- 10:34 PM Bug #38247 (Resolved): msg/async: v2 reconnect_seq assert due to interfering connections
- ...
- 02:47 PM Backport #38242 (Resolved): luminous: msg/async: connection race + winner fault can leave connect...
- https://github.com/ceph/ceph/pull/27583
- 02:47 PM Backport #38241 (Rejected): mimic: msg/async: connection race + winner fault can leave connection...
- 12:09 AM Bug #38068 (Resolved): ceph_test_msgr: connect couldn't create socket (97) Address family not sup...
- 12:08 AM Bug #37779 (Pending Backport): msg/async: connection race + winner fault can leave connection in ...
- see https://github.com/ceph/ceph/pull/25754
02/07/2019
- 04:40 PM Bug #38216: "HEALTH_WARN 3 monitors have not enabled msgr2" in rados
- Apparently we'll need to add a command enabling msgr2 on the MONs, to be run after the upgrade completes. But I don't...
- 12:29 AM Bug #38216: "HEALTH_WARN 3 monitors have not enabled msgr2" in rados
- @Sage what do you suggest ?
- 12:29 AM Bug #38216: "HEALTH_WARN 3 monitors have not enabled msgr2" in rados
- looks like also cause for failed tests https://github.com/ceph/ceph/pull/26302
02/06/2019
- 09:04 PM Bug #38216 (Resolved): "HEALTH_WARN 3 monitors have not enabled msgr2" in rados
- Run: http://pulpito.ceph.com/yuriw-2019-02-04_20:42:30-upgrade:mimic-x-master-distro-basic-smithi/
Jobs: all
Logs: ...
02/04/2019
- 06:00 AM Bug #38118 (Resolved): msgr2: race connection results in hang
- https://github.com/ceph/ceph/pull/26256
Also available in: Atom