On 1/15/16 4:36 PM, Yehuda Sadeh-Weinraub wrote:
The head object of a multipart object has 0 size, so it's expected. What's missing is the tail of the object. I don't assume you have any logs from when the object was uploaded? Yehuda On Fri, Jan 15, 2016 at 2:12 PM, seapasulli@xxxxxxxxxxxx <seapasulli@xxxxxxxxxxxx> wrote:Sorry for the confusion:: When I grepped for the prefix of the missing object:: "2015\/01\/01\/PAKC\/NWS_NEXRAD_NXL2DP_PAKC_20150101110000_20150101115959.tar.2~pcu5Hz6foFXjlSxBat22D8YMcHlQOBD" I am not able to find any chunks of the object:: lacadmin@kh28-10:~$ rados -p .rgw.buckets ls | grep 'pcu5Hz6' lacadmin@kh28-10:~$ The only piece of the object that I can seem to find is the original one I posted:: lacadmin@kh28-10:~$ rados -p .rgw.buckets ls | grep 'NWS_NEXRAD_NXL2DP_PAKC_20150101110000_20150101115959' default.384153.1_2015/01/01/PAKC/NWS_NEXRAD_NXL2DP_PAKC_20150101110000_20150101115959.tar And when we stat this object is is 0 bytes as shown earlier:: lacadmin@kh28-10:~$ 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 2015-11-04 15:29:30.000000, size 0 Sorry again for the confusion. On 1/15/16 3:58 PM, Yehuda Sadeh-Weinraub wrote:Ah, I see. Misread that and the object names were very similar. No, don't copy it. You can try to grep for the specific object name and see if there are pieces of it lying around under a different upload id. Yehuda On Fri, Jan 15, 2016 at 1:44 PM, seapasulli@xxxxxxxxxxxx <seapasulli@xxxxxxxxxxxx> wrote:Sorry I am a bit confused. The successful list that I provided is from a different object of the same size to show that I could indeed get a list. Are you saying to copy the working object to the missing object? Sorry for the confusion. On 1/15/16 3:20 PM, Yehuda Sadeh-Weinraub wrote:That's interesting, and might point at the underlying issue that caused it. Could be a racing upload that somehow ended up with the wrong object head. The 'multipart' object should be 4M in size, and the 'shadow' one should have the remainder of the data. You can run 'rados stat -p .rgw.buckets <oid>' to validate that. If that's the case, you can copy these to the expected object names: $ src_uploadid=wksHvto9gRgHUJbhm_TZPXJTZUPXLT2 $ dest_uploadid=pcu5Hz6foFXjlSxBat22D8YMcHlQOBD $ rados -p .rgw.buckets cp default.384153.1__multipart_2015/01/01/KABR/NWS_NEXRAD_NXL2DP_KABR_20150101130000_20150101135959.tar.2~${src_uploadid}.1 default.384153.1__multipart_2015/01/01/KABR/NWS_NEXRAD_NXL2DP_KABR_20150101130000_20150101135959.tar.2~${dest_uploadid}.1 $ rados -p .rgw.buckets cp default.384153.1__shadow_2015/01/01/KABR/NWS_NEXRAD_NXL2DP_KABR_20150101130000_20150101135959.tar.2~${src_upload_id}.1_1 default.384153.1__shadow_2015/01/01/KABR/NWS_NEXRAD_NXL2DP_KABR_20150101130000_20150101135959.tar.2~${dest_upload_id}.1_1 Yehuda On Fri, Jan 15, 2016 at 1:02 PM, seapasulli@xxxxxxxxxxxx <seapasulli@xxxxxxxxxxxx> wrote:lacadmin@kh28-10:~$ rados -p .rgw.buckets ls | grep 'pcu5Hz6' lacadmin@kh28-10:~$ Nothing was found. That said when I run the command with another prefix snippet:: lacadmin@kh28-10:~$ rados -p .rgw.buckets ls | grep 'wksHvto' default.384153.1__shadow_2015/01/01/KABR/NWS_NEXRAD_NXL2DP_KABR_20150101130000_20150101135959.tar.2~wksHvto9gRgHUJbhm_TZPXJTZUPXLT2.1_1 default.384153.1__multipart_2015/01/01/KABR/NWS_NEXRAD_NXL2DP_KABR_20150101130000_20150101135959.tar.2~wksHvto9gRgHUJbhm_TZPXJTZUPXLT2.1 On 1/15/16 12:05 PM, Yehuda Sadeh-Weinraub wrote:On Fri, Jan 15, 2016 at 9:36 AM, seapasulli@xxxxxxxxxxxx <seapasulli@xxxxxxxxxxxx> wrote:Hello Yehuda, Here it is:: radosgw-admin object stat --bucket="noaa-nexrad-l2" --object="2015/01/01/PAKC/NWS_NEXRAD_NXL2DP_PAKC_20150101110000_20150101115959.tar" { "name": "2015\/01\/01\/PAKC\/NWS_NEXRAD_NXL2DP_PAKC_20150101110000_20150101115959.tar", "size": 7147520, "policy": { "acl": { "acl_user_map": [ { "user": "b05f707271774dbd89674a0736c9406e", "acl": 15 } ], "acl_group_map": [ { "group": 1, "acl": 1 } ], "grant_map": [ { "id": "", "grant": { "type": { "type": 2 }, "id": "", "email": "", "permission": { "flags": 1 }, "name": "", "group": 1 } }, { "id": "b05f707271774dbd89674a0736c9406e", "grant": { "type": { "type": 0 }, "id": "b05f707271774dbd89674a0736c9406e", "email": "", "permission": { "flags": 15 }, "name": "noaa-commons", "group": 0 } } ] }, "owner": { "id": "b05f707271774dbd89674a0736c9406e", "display_name": "noaa-commons" } }, "etag": "b91b6f1650350965c5434c547b3c38ff-1\u0000", "tag": "_cWrvEa914Gy1AeyzIhRlUdp1wJnek3E\u0000", "manifest": { "objs": [], "obj_size": 7147520, "explicit_objs": "false", "head_obj": { "bucket": { "name": "noaa-nexrad-l2", "pool": ".rgw.buckets", "data_extra_pool": ".rgw.buckets.extra", "index_pool": ".rgw.buckets.index", "marker": "default.384153.1", "bucket_id": "default.384153.1" }, "key": "", "ns": "", "object": "2015\/01\/01\/PAKC\/NWS_NEXRAD_NXL2DP_PAKC_20150101110000_20150101115959.tar", "instance": "" }, "head_size": 0, "max_head_size": 0, "prefix": "2015\/01\/01\/PAKC\/NWS_NEXRAD_NXL2DP_PAKC_20150101110000_20150101115959.tar.2~pcu5Hz6foFXjlSxBat22D8YMcHlQOBD",Try running: $ rados -p .rgw.buckets ls | grep pcu5Hz6 Yehuda"tail_bucket": { "name": "noaa-nexrad-l2", "pool": ".rgw.buckets", "data_extra_pool": ".rgw.buckets.extra", "index_pool": ".rgw.buckets.index", "marker": "default.384153.1", "bucket_id": "default.384153.1" }, "rules": [ { "key": 0, "val": { "start_part_num": 1, "start_ofs": 0, "part_size": 0, "stripe_max_size": 4194304, "override_prefix": "" } } ] }, "attrs": {} } On 1/15/16 11:17 AM, Yehuda Sadeh-Weinraub wrote:radosgw-admin object stat --bucket=<bucket> --object=<object>'
_______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
