Sandy Xu
- Login: Sandy4999
- Email: sandyxu4999@gmail.com
- Registered on: 08/29/2015
- Last sign in: 06/28/2020
Issues
open | closed | Total | |
---|---|---|---|
Assigned issues | 0 | 0 | 0 |
Reported issues | 0 | 6 | 6 |
Activity
06/09/2017
- 09:54 AM CephFS Bug #19955: Too many stat ops when MDS trying to probe a large file
- Later I found two more related problems that may need further discussion:
# Some tools(like fio) will set the size o...
05/18/2017
- 01:14 AM CephFS Bug #19955: Too many stat ops when MDS trying to probe a large file
- John Spray wrote:
> File size recovery should only happen when a client has failed to reconnect properly during the ...
05/17/2017
- 03:29 AM CephFS Bug #19955 (Resolved): Too many stat ops when MDS trying to probe a large file
- When MDS recovers, it may emit tons of `stat` ops to OSDs trying to probe a large file, which as a result prevents cl...
04/25/2017
- 03:39 AM CephFS Bug #19755 (Resolved): MDS became unresponsive when truncating a very large file
- We were trying to copy a very large file (7TB exactly) between two directories through Samba/CephFS, and cancelled it...
03/07/2017
- 09:20 AM CephFS Bug #19204 (Resolved): MDS assert failed when shutting down
- We encountered a failed assertion when trying to shutdown an MDS. Here is a snippet of the log:
> -14> 2017-01-22 ...
08/12/2016
- 09:34 PM Ceph Revision f9637743 (ceph): rgw: add a method to purge all associate keys when removing a subuser
- Fixes: #12890
When removing a subuser, make sure all of its keys, including the swift key and possible s3 keys, are ...
02/17/2016
- 02:45 AM rgw Bug #14764: RGW: swift API 'swift stat' double counts objects
- Probably a known issue, you may want to take a look at http://tracker.ceph.com/issues/13140
01/15/2016
- 11:12 AM Ceph Revision 0a4d1a41 (ceph): mon: add an independent option for max election time
- add a new configure option mon_election_timeout, which is not dependent on mon_lease
Signed-off-by: Sangdi Xu <xu.sa...
12/17/2015
- 09:30 AM Ceph Revision 42692630 (ceph): crush: fix the default type 0 name if it's not specified when decompiled
- Change the type 0 name from "device" to "osd", just to make it consistent
Signed-off-by: Sangdi Xu <xu.sangdi@h3c.com> - 09:28 AM Ceph Revision 895bf8b8 (ceph): crushtool: set type 0 name "osd" for --build option
- The crushmap built by crushtool cannot be used directly as the name of type 0 is not specified.
We may decompile the ...
Also available in: Atom