[DOCS-6799] Changes to profiler entries / logs for getMore in 3.2.1 Created: 11/Dec/15  Updated: 30/Oct/23  Resolved: 09/Nov/18

Status: Closed
Project: Documentation
Component/s: Server
Affects Version/s: None
Fix Version/s: Server_Docs_20231030

Type: Task Priority: Major - P3
Reporter: David Storch Assignee: Kay Kim (Inactive)
Resolution: Duplicate Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Documented
documents SERVER-21750 getMore command does not set "nreturn... Closed
Duplicate
duplicates DOCS-6141 Doc changes to the profiler format fo... Closed
Participants:
Days since reply: 6 years, 2 weeks, 2 days ago

 Description   

In SERVER-21750, we made some small changes to the format of a getMore command profiler entry and the logs in version 3.2.1:

  • We omit ntoreturn and ntoskip from the the log line for any command, including getMore command. ntoreturn and ntoskip will still appear for log lines for legacy OP_QUERY find or legacy OP_GET_MORE. (DOCS-6141)
  • We added nreturned to the log line and profiler for getMore command (a bug fix). (no-op )
  • We now upconvert legacy OP_GET_MORE profiler entries to look similar to getMore command. (heh – no-op since we already lumped getMore = command + OP_GET_MORE)

Here is an example to demonstrate the third change. Prior to the change, OP_GET_MORE profiler entries look like this:

{ "op" : "getmore", "ns" : "test.c", "query" : {  }, "cursorid" : 22112274967,
"cursorExhausted" : true, "keyUpdates" : 0, "writeConflicts" : 0, "numYield" :
0, "locks" : { "Global" : { "acquireCount" : { "r" : NumberLong(2) } },
"MMAPV1Journal" : { "acquireCount" : { "r" : NumberLong(1) } }, "Database" : {
"acquireCount" : { "r" : NumberLong(1) } }, "Collection" : { "acquireCount" : {
"R" : NumberLong(1) } } }, "nreturned" : 2, "responseLength" : 64, "millis" : 0,
"execStats" : {  }, "ts" : ISODate("2015-12-09T22:59:16.181Z"), "client" :
"127.0.0.1", "allUsers" : [ ], "user" : "" }

After this change, the profiler entry for an identical OP_GET_MORE is "upconverted" to appear as a getMore command:

{ "op" : "getmore", "ns" : "test.c", "query" : { "getMore" :
NumberLong("24490776063"), "collection" : "c", "batchSize" : NumberLong(3) },
"cursorid" : 24490776063, "cursorExhausted" : true, "keyUpdates" : 0,
"writeConflicts" : 0, "numYield" : 0, "locks" : { "Global" : { "acquireCount" :
{ "r" : NumberLong(2) } }, "MMAPV1Journal" : { "acquireCount" : { "r" :
NumberLong(1) } }, "Database" : { "acquireCount" : { "r" : NumberLong(1) } },
"Collection" : { "acquireCount" : { "R" : NumberLong(1) } } }, "nreturned" : 2,
"responseLength" : 64, "millis" : 0, "execStats" : {  }, "ts" :
ISODate("2015-12-09T22:59:52.701Z"), "client" : "127.0.0.1", "allUsers" : [ ],
"user" : "" }

I'm not sure if there are any documentation changes required associated with this change. Since it changes user-facing output, however, I wanted to make sure the DOCS team is aware and can take action as necessary.



 Comments   
Comment by Gregory McKeon (Inactive) [ 29/Jan/18 ]

kay.kim assigning to you to clear Pavi's queue.

Generated at Thu Feb 08 07:53:00 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.