stats for key are not defined in documentation, many stats are not defined

Dipti Borkar added a comment - 15/May/12 11:41 AM Chiyoung, can you provide details on stats key so that documentation can be added?

Chiyoung Seo added a comment - 24/May/12 4:28 AM Jin, We should provide the "stats key" documentation for the client library team, so that they can refer to this documentation to implement OBSERVE command in the client libraries.

Chiyoung Seo added a comment - 01/Jun/12 2:48 AM Assign it to Mike as he was mainly involved in the initial OBSERVE design.

Thuan Nguyen added a comment - 02/Jun/12 12:45 AM Integrated in github-ep-engine-2-0 #301 (See [ http://qa.hq.northscale.net/job/github-ep-engine-2-0/301/ ])      MB-5304 : Document stats key/vkey in stats.org (Revision df4dd0f9aa65b69a36a304ca76a8389090bd3297)      Result = SUCCESS Mike Wiederhold : Files : * docs/stats.org

Matt Ingenthron added a comment - 04/Jun/12 11:39 AM - edited Documentation mixes stats key and stats vkey, though the return values for each need to be understood. Also, see MB-5435 . This documentation appears to be wrong with respect to the key valid field, or there's some subtlety that is not explained by the table.

Mike Wiederhold added a comment - 04/Jun/12 12:52 PM Stats key and vkey return the same data. The only difference is the key_valid field returned by stats vkey. I have marked each stat with the command that returns it in the column on the left. I made a few minor wording changes to the documentation, but I disagree that the documentation is wrong. Please provide an explanation of what is confusing or contact me over Skype to discuss anything you think is incorrect. ** Stats Key and Vkey | key_cas | The keys current cas value |KV| | key_data_age | How long the key has waited for its |KV| | | value to be persisted (0 if clean) |KV| | key_exptime | Expiration time from the epoch |KV| | key_flags | Flags for this key |KV| | key_is_dirty | If the value is not yet persisted |KV| | key_last_modified_time | Last updated time |KV| | key_valid | See description below | V| | key_vb_state | The vbucket state of this key |KV| =key_valid= can have the following responses: this_is_a_bug - Some case we didn't take care of. dirty - The value in memory has not been persisted yet. length_mismatch - Data length in memory doesn't match the data length on disk. data_mismatch - The data in memory doesn't match the data on disk. flags_mismatch - The flags in memory don't match the flags on disk. valid - The data on disk and in memory match ram_but_not_disk - The value doesn't exist yet on disk. item_deleted - The item has been deleted.

Sergey Avseyev added a comment - 04/Jun/12 1:47 PM What is KV in the last column? I think it should be KV instead of V for key_valid

Mike Wiederhold added a comment - 04/Jun/12 2:25 PM K = Key V = VKey And key_valid is not included in stats key.

Sergey Avseyev added a comment - 04/Jun/12 2:29 PM What does mean "is not included in stats"? I can have key "x" with value "test" in the default bucket on couchbase 2.0. Here is the output of the command line tools:     $ cbc cat x     "x" Size 5 Flags 0x0 CAS 0x69cf70ff8de90000     test     $ curl -s http://localhost:8091/pools/default/buckets/default | vbuckettool - x     key: x master: 127.0.0.1:11210 vBucketId: 220 couchApiBase: http://127.0.0.1:8092/default replicas:     $ cbc stats "key x 220"     127.0.0.1:11210 key_is_dirty 0     127.0.0.1:11210 key_exptime 0     127.0.0.1:11210 key_flags 0     127.0.0.1:11210 key_cas 256796085243753     127.0.0.1:11210 key_dirtied 0     127.0.0.1:11210 key_data_age 0     127.0.0.1:11210 key_last_modification_time 1338826758     $ cbc stats "vkey x 220"     127.0.0.1:11210 key_is_dirty 0     127.0.0.1:11210 key_exptime 0     127.0.0.1:11210 key_flags 0     127.0.0.1:11210 key_cas 256796085243753     127.0.0.1:11210 key_dirtied 0     127.0.0.1:11210 key_data_age 0     127.0.0.1:11210 key_last_modification_time 1338826854     127.0.0.1:11210 key_valid valid You can see the last line says there is key "key_valid" and value "valid"

Sergey Avseyev added a comment - 04/Jun/12 2:32 PM - edited :) I see now sorry. VKey. So I have no objections any more

Mike Wiederhold added a comment - 04/Jun/12 3:43 PM Sergey appears to be okay with the documentation. PLease reopen if any other issues come up.

Matt Ingenthron added a comment - 04/Jun/12 5:16 PM My confusion too was based on the last column. Could we clarify that (if not done already)? I thought you were saying whether or not each stat returned a key and value for the stat.