Activity
From 10/18/2018 to 11/16/2018
11/16/2018
- 05:04 PM Bug #37292 (Resolved): RDMAStack - poll failed error when using IB and RDMACM
- The error message "polling - poll failed" occurs because interrupted system calls are not handled in method RDMADispa...
11/15/2018
- 09:06 AM Bug #36764: ceph-mgr: blocking requests sent to restful api server hangs sometimes
- * luminous: https://github.com/ceph/ceph/pull/25047
* mimic: https://github.com/ceph/ceph/pull/25113 - 06:17 AM Bug #36764 (Pending Backport): ceph-mgr: blocking requests sent to restful api server hangs somet...
11/13/2018
- 10:05 AM Bug #36764 (Fix Under Review): ceph-mgr: blocking requests sent to restful api server hangs somet...
- https://github.com/ceph/ceph/pull/25048
11/12/2018
- 05:06 AM Bug #36764: ceph-mgr: blocking requests sent to restful api server hangs sometimes
- https://github.com/ceph/ceph/pull/25047
- 04:34 AM Bug #36764 (Resolved): ceph-mgr: blocking requests sent to restful api server hangs sometimes
- We setup a ceph cluster (v12.2.2) with restful api plugin running, but sometimes requests got blocked forever without...
11/01/2018
- 01:28 PM Bug #36666: msg: rejoin message queued but not sent
- The events that trigger this issue is clear, but I don't have clear idea how to fix it.
- 10:38 AM Bug #36666: msg: rejoin message queued but not sent
- socket failure injection caused mds.0 to send connect message with connect_seq == 1...
10/31/2018
- 11:24 PM Bug #36666 (Duplicate): msg: rejoin message queued but not sent
- mds.0 stuck in rejoin because mds.1 does not receive the sent message. REsults in this failure:...
10/29/2018
- 02:19 AM Bug #36612 (Resolved): msg/async: connection stall
- ...
10/25/2018
- 07:47 PM Bug #36599: msg/async: should read data out of socket after shutdown(), before close()
- See e.g. https://stackoverflow.com/questions/8872988/properly-close-a-tcp-socket
- 07:46 PM Bug #36599 (New): msg/async: should read data out of socket after shutdown(), before close()
- From a user:
These are right steps to close a TCP session and avoid resets; RESETs are bad and generally should o... - 06:30 PM Bug #25027: mon: src/msg/async/AsyncConnection.cc: 1710: FAILED assert(can_write == WriteStatus::...
- Seen in mimic: /a/yuriw-2018-10-24_20:03:55-rados-wip-yuri2-testing-2018-10-23-1513-mimic-distro-basic-smithi/3180310/
10/24/2018
- 08:36 PM Bug #18664: ceph_test_msgr SIGABRT in decode_message, SimpleMessenger, can't decode unknown messa...
- Similar failure seen in luminous:...
- 11:12 AM Bug #36561 (Won't Fix): msg: big messenger performance regression
- 10:12 AM Bug #36561: msg: big messenger performance regression
- I checked ceph complied in Release mode. Its performance is about the same as mimic. Sorry for the noisy.
- 06:58 AM Bug #36561: msg: big messenger performance regression
- Debug mode. I will try ceph compiled in Release mode.
- 06:45 AM Bug #36561: msg: big messenger performance regression
- Zheng Yan wrote:
> my perf.data relies on my environment. all symbols will get lost in your environment. do you real... - 06:39 AM Bug #36561: msg: big messenger performance regression
- perf.data relies on my environment. all symbols will get lost in your environment. do you really want it?
- 06:26 AM Bug #36561: msg: big messenger performance regression
- @Zheng could you also provide the perf.data file for those runs?
10/23/2018
- 08:34 AM Bug #36561 (Won't Fix): msg: big messenger performance regression
- I use following code to check how fast messenger is. Request rate of current master code (commit 98fc7ebc99) is only ...
- 01:45 AM Bug #36495 (Resolved): ceph_test_msgr handle_connect_reply_2 connect got BADAUTHORIZER
10/22/2018
- 03:06 PM Bug #36495 (Fix Under Review): ceph_test_msgr handle_connect_reply_2 connect got BADAUTHORIZER
- https://github.com/ceph/ceph/pull/24697
- 02:30 PM Bug #36495: ceph_test_msgr handle_connect_reply_2 connect got BADAUTHORIZER
- /a/sage-2018-10-21_17:53:53-rados-wip-sage-testing-2018-10-21-1053-distro-basic-smithi/3170055
10/19/2018
- 10:45 PM Bug #36540 (New): msg: messages are queued but not sent
- > In CephFS testing, we've observed transient failures caused by what
> appears to messages being dropped [1,2]. The...
Also available in: Atom