Activity
From 05/11/2019 to 06/09/2019
06/08/2019
- 03:30 PM Bug #40114 (Pending Backport): msg: bad address encoding when CEPH_FEATURE_MSG_ADDR2 enabled
- 12:07 AM Bug #40209 (New): msg: connection thrash failure causes mon to be marked down
- I would expect this to be recoverable without a health warning. Is that correct?
Injected failure causing mon.1 (m...
06/07/2019
- 03:26 PM Bug #40115: handle_read_frame_preamble_main crc mismatch for main preamble rx_crc=2646747243 tx_c...
- PR: https://github.com/ceph/ceph/pull/28453
06/04/2019
- 01:24 PM Bug #40114: msg: bad address encoding when CEPH_FEATURE_MSG_ADDR2 enabled
- PR here: https://github.com/ceph/ceph/pull/28379
- 01:22 PM Bug #40114 (Fix Under Review): msg: bad address encoding when CEPH_FEATURE_MSG_ADDR2 enabled
- 07:13 AM Bug #40115: handle_read_frame_preamble_main crc mismatch for main preamble rx_crc=2646747243 tx_c...
- > shouldn't the stream handlers get reset unconditionally on reset?
Just making the reset unconditional will brake...
06/03/2019
- 08:59 PM Bug #40115: handle_read_frame_preamble_main crc mismatch for main preamble rx_crc=2646747243 tx_c...
- ...
- 08:41 PM Bug #40115: handle_read_frame_preamble_main crc mismatch for main preamble rx_crc=2646747243 tx_c...
- hmm, on the other end, the fault is waiting 15s (!)...
- 08:39 PM Bug #40115 (Resolved): handle_read_frame_preamble_main crc mismatch for main preamble rx_crc=2646...
- symptom is MON_DOWN in cluster log, due to the mon reconnects failing, and thus missing an election cycle.
seems t... - 05:07 PM Bug #40114: msg: bad address encoding when CEPH_FEATURE_MSG_ADDR2 enabled
- So the plan I think is to just have the clients assume that the field in question is LE. That means that LE hosts wit...
- 04:24 PM Bug #40114: msg: bad address encoding when CEPH_FEATURE_MSG_ADDR2 enabled
- Another bug too: Apparently BSD/OSX have a sa_len field in the sockaddr, but it does not reduce the length that it en...
- 04:06 PM Bug #40114 (Resolved): msg: bad address encoding when CEPH_FEATURE_MSG_ADDR2 enabled
- On Linux, entity_addr_t::encode just copies the sockaddr as-is into the bufferlist, without fixing up the endianness ...
05/27/2019
05/17/2019
- 01:32 PM Backport #39242 (Need More Info): mimic: msg/async: connection race + winner fault can leave conn...
- Non-trivial backport; requires special handling to avoid breaking RBD tests. Assigning to the developer.
05/15/2019
- 07:58 PM Backport #39242: mimic: msg/async: connection race + winner fault can leave connection stuck at r...
- Please also pull in PR #28050 as well to avoid breaking RBD tests [1]
[1] https://github.com/ceph/ceph/pull/28050 - 07:55 PM Bug #37499 (Resolved): AsyncConnection stuck
05/12/2019
- 01:59 PM Bug #39711 (Duplicate): "unable to find any IPv4 address in networks <ipv6-network>" after upgrad...
- After upgrading our OSDs to Nautilus they refused to start logging...
Also available in: Atom