It looks like the gateway is experiencing a similar race condition to
what we reported before.
The rados object has a size of 0 bytes but the bucket index shows the
object listed and the object metadata shows a size of
7147520 bytes.
I have a lot of logs but I don't think any of them have the full data
from the upload of this object.
I thought this bug was fixed back in firefly/giant
https://www.mail-archive.com/[email protected]/msg19971.html
--------------------------
root@kg34-33:/srv/nfs/griffin_temp# rados -p .rgw.buckets stat
default.384153.1_2015/01/01/PAKC/NWS_NEXRAD_NXL2DP_PAKC_20150101110000_20150101115959.tar
..rgw.buckets/default.384153.1_2015/01/01/PAKC/NWS_NEXRAD_NXL2DP_PAKC_20150101110000_20150101115959.tar
mtime 1446672570, size 0
--------------------------
SError: [Errno 2] No such file or directory:
'/srv/nfs/griffin_tempnoaa-nexrad-l2/2015/01/01/PAKC/NWS_NEXRAD_NXL2DP_PAKC_20150101110000_20150101115959.tar'
In [13]: print(key.size)
7147520
We are currently using 94.5 and the file were uploaded to hammer as well
lacadmin@kh28-10:~$ ceph --version
ceph version 0.94.5 (9764da52395923e0b32908d83a9f7304401fee43)
lacadmin@kh28-10:~$ radosgw --version
ceph version 0.94.5 (9764da52395923e0b32908d83a9f7304401fee43)
The cluster is health_ok and was ok during the upload. I need to confirm
with the person who uploaded the data but I think they did it with
s3cmd. Has anyone seen this before? I think I need to file a bug :-(
_______________________________________________
ceph-users mailing list
[email protected]
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
