Project

General

Profile

Activity

From 06/22/2014 to 07/21/2014

07/21/2014

01:23 PM Fix #8892 (New): msgr: clean up local Session invocation functions
We are fairly inconsistent about delivering connect() setups for the loopback Connection in the SimpleMessenger. We s... Greg Farnum

07/19/2014

09:24 AM Bug #8880: msg/Pipe.cc: 1538: FAILED assert(0 == "old msgs despite reconnect_seq feature")
Which daemon was this?
Looks like that commit does include the fix for #8504... :(
Greg Farnum
09:10 AM Bug #8880 (Resolved): msg/Pipe.cc: 1538: FAILED assert(0 == "old msgs despite reconnect_seq featu...
ubuntu@teuthology:/a/teuthology-2014-07-18_02:32:01-rados-master-testing-basic-plana/368391... Sage Weil

07/01/2014

11:06 AM Bug #8504 (Resolved): msgr: FAILED assert(0 == "old msgs despite reconnect_seq feature")
This doesn't actually require any backports. :)
Prior to the fast dispatch work, the pre_publish_lock was held by ...
Greg Farnum

06/30/2014

09:50 PM Bug #8504 (Pending Backport): msgr: FAILED assert(0 == "old msgs despite reconnect_seq feature")
Merged to next in commit:8697d6a10ee822e8a032a559c78e211e2182799a
Unfortunately back porting it will require a lit...
Greg Farnum

06/27/2014

03:14 PM Bug #8504: msgr: FAILED assert(0 == "old msgs despite reconnect_seq feature")
Okay, and after more discussion (thanks Sage!), the "Messenger bug" is the same as the OSD bug — the surviving OSD ot... Greg Farnum
02:50 PM Bug #8504: msgr: FAILED assert(0 == "old msgs despite reconnect_seq feature")
Of course the lookup is supposed to be protected by the map reservation system, but since we're doing mark_down prior... Greg Farnum
02:41 PM Bug #8504: msgr: FAILED assert(0 == "old msgs despite reconnect_seq feature")
get_con_osd_cluster might be okay, since a marked_down Connection won't send any outgoing messages, which is what we ... Greg Farnum
02:20 PM Bug #8504: msgr: FAILED assert(0 == "old msgs despite reconnect_seq feature")
OSD bug:
1) In handle_osd_map(), we need to wait until nobody is using a (previously-)reserved map prior to doing an...
Greg Farnum
01:49 PM Bug #8504: msgr: FAILED assert(0 == "old msgs despite reconnect_seq feature")
Greg Farnum wrote:
> So it looks like osd.4 called mark_down on the addr-based interface and nuked the Pipe from han...
Greg Farnum

06/26/2014

04:46 PM Bug #8504: msgr: FAILED assert(0 == "old msgs despite reconnect_seq feature")
The one I originally reported is much simpler:
osd.0 got marked down
osd.1 called mark_down and then forgot about i...
Greg Farnum
03:21 PM Bug #8504: msgr: FAILED assert(0 == "old msgs despite reconnect_seq feature")
Okay, so the crashed OSD (osd.5) was marked down by the other endpoint (osd.4). Then osd.5 reconnected to osd.4, whic... Greg Farnum
10:54 AM Bug #8504 (In Progress): msgr: FAILED assert(0 == "old msgs despite reconnect_seq feature")
Not a bug we introduced, but the existing patch only fixes the bug on the accept side. These crashes are on the conne... Greg Farnum
 

Also available in: Atom