pegasus/readme.pulloperations - diff

Return to readme.pulloperations CVS log

Up to [Pegasus] / pegasus

Diff for /pegasus/readme.pulloperations between version 1.1.2.16 and 1.1.2.34

version 1.1.2.16, 2013/11/21 17:59:38

version 1.1.2.34, 2014/05/29 23:55:34

Line 11

1. Binary operation from OOP. Need to add counter to binary

protocol to be able to count objects in response. Generates

warnings in things like messageserializer and does not work with

OOP right now. Corrected by converting to XML.

OOP right now. Fixed by converting to XML. Concluded that we do not

2. OpenExecQuery - Code is incorrect in that it does not include the

need to do this. The binary response is not really used often

return from the exec query function to the aggregator yet.

in the current environment So double mapping it is not a major issue.

3. Code for Pull part of OpenQueryInstancesRequest a) should be part of

Leave this as FUTURE

the common CIMOperationRequestDispatcher execCommon code.

2. OpenExecQuery - Code is incomplete in that it does not include the

4. The changes to WQLCIMOperationRequestDispatcher and CQL... for handling

return from the exec query function to the aggregator yet. Code for Pull

part of OpenQueryInstancesRequest a) should be part of the common

CIMOperationRequestDispatcher execCommon code. DONE, not tested. 29 May

3. The changes to WQLCIMOperationRequestDispatcher and CQL... for handling

pull not completed so we feed the responses back to the EnmerationContext

queues

queues. DONE Not tested 29 May

3. Lots of minor TODOs, diagnostics, etc.

4. Minor TODOs, diagnostics, etc. still in the code

4. External runtime variables. Proposing that they be fixed for this release

5. Extension to avoid double move of objects in CIMResponseData (one

rather than set by configuration. This should be discussed. Am making

this a separate bug. See bug 9819 for the changes to cover this.

5. Decision on EnumerationContext timeout (separate thread or just

checks during other operations). Can we, in fact really keep the

enumeration context table and queue under control without monitoring

with a separate thread. We must monitor for:

a. Client operation that stop requesting (i.e. inter operation time

exceeds operationTimeout). Note that if it simply exceeds the time

the next operation does the cleanup. The issue is those clients that

simply stop and do not either close or go to completion.

b. We should protect against providers that no not every finish delivering

or take to long between deliveries. This does not exist in Pegasus

today

6. Consider moving some of the code in dispatcher from templates to common

functions which would mean adding intermediate classes in CIMMessage but

would reduce code size.

7. Extension to avoid double move of objects in CIMResponseData (one

into enumerationContext queue and second to new cimResponseData for

response. Want to avoid second move by extending Open/Pull response

messages to include count and CIMResponse data to count objects out

of queue when converting (avoids the second move). Big issue here

of queue when converting (avoids the second move). This would mean

with binary data since need to extend format to count it.

extending the output writers to allow the count field to be supplied

8. NEXT TASKS:

so they would only create output for up to the count supplied.(Make

a. test the enumeration timeout thread

this future beyond bug 9676)

b. finish and test the OpenQueryInstances

6. Add more static tests (currently only OpenEnumerateInstances and

c. Clean up TODOs

OpenEnumerateInstanceNames covered).

d. Find issue when we run makepoststarttests in pullop client with

Low priority because most of this except exact XML covered in other

forceProviderProcesses = true. This causes an operation like

tests. note that we can only add very limited tests since a) no

cimcli pei CIM_ManagedElement to not complete (client timeout)

way to test multiple operations in sequence(i.e. open/pull) and

sometimes.

cannot be sure how many responses will come back in all cases even

for open.

7. Clean up the CIMOperationRequestDecoder.cpp rejectCIMException

further since there is some overlap. VERY Low priority.

8. Correct issue between operations and HTTP where we are sending

trailers with exceptions. Modify response mechanisms so that we

set non-chunked for all responses where we send error responses to

avoid the trailers.

9. It would be more efficient in dispatcher to always use exception for

rejects and change the _reject functions so that they never return

when they reject. This is VERY LOW PRIORITY and primarily saves

a few lines of code in the reject functions and their calls. Means we

would code.

_rejectIfEnumerationToBroad(...)

_rejectIfThisParameterIncorrect(...)

instead of

if (_rejectIfEnum...)

{

return true

}

It would mean that the method trace for the handlers would not return

an exit if we rejected. VERY LOW PRIORITY

10. There are still a couple of template functions around the task of

distributing requests to the multiple providers.

In fact there are two similiar but different templates for

the associators(i.e. assoc, ref, etc.) functions and the parallel

openAssoc functions. It would be nice to consolidate that logic and

further to try to create a non-template form for those functions. LOW

PRIORITY.

11. External runtime variables. This bug leaves these variables as

internal to CIMOperationRequestDispatcher.cpp.

See bug 9819 for the changes to externalize these variables, both

the runtime and buildtime variables. The Work group discussion

documents what the variables are and how they tie to statistics info.

See the OpenPegasus Wiki, Pull operations work group.

12. Not directly related to pull but now over half the code in

WQLOperationRequestDispatcher.cpp and CQLOperationRequestDispatcher

handle functions is just duplicated. Need to create a common function

and eliminate the duplicated code

13. Left almost duplicate createContext function in EnumerationContextTab.

Clean up code

NEXT TASKS:

a. finish 8, 11 2,3 above ( Finish by about 1 June)

29 May 2014

1. Correct issue in cimcli that was causing rejects of open operations when

there were no filters requested. The standard default is WQL in cimcli

so we overrode this for the pull operations.

2. Clean up OpenQueryInstances in CIMMessage.h and CIMMessage.cpp.

3. Modify CIMOperationRequestDispatcher and

EnumerationContextTable.CreateContext to pass request rather than request

parameters.

4. Remove a number of diagnostic messages. Note that there are still some.

They are marked KS_TODO with with note to remove before checkin. They are

all traces and validations.

5. Correct error in inclusion of execinfo.h include in CIMOperationRequestDis.

found by Ji Lee.

6. Cleaned up execQuery and extended it for OpenQueryInstances. ExecQuery

tested but have not tested OpenQueryInstances yet. One more checkin and we

should be able to close 2,3 above (OpenQueryInstances).

26 May 2014

1. Remove a number of diagnostics.

2. Make cache size in EnumerationContextTable dynamic and remove function

to set default values. MaximumEnums set from Dispatcher on constructor for

EnumerationContextTable and table size set to 1/2 of maximum number of

open contexts.

3. Clean up some statistics in EnumerationContextTable

4. Correct error in query and queryLanguage parameters in cimcli in that

the default for queryLanguage is WQL but must be "" for pull operations.

If WQL found they set it to "".

9 May 2104

1. Converted several template functions in CIMOperationRequestDispatcher

to real functions

2. Added intermediate CIMMessage for CIMPullOperationRequestMessage since

all of the pulls have exactly the same data. This allowed us to

convert the template function processPullRequest from template to

real function

3. Added intermediate CIMMessage for CIMOpenOperationRequestMessage betweeh\n

all of the Opens and the CIMRequestData message. This will allow

common code to test all of these variables in all 6

CIMOperationRequestDispatcher handleOpen... functions

3. Added variable in CIMOperationRequestDispatcher handleEnqueue to make

releasing the original request conditional on return from the handle...

for open... and pull... handlers.

This allowed us to drop one place where we were creating a new request msg

for the delayed response and use the original request. This meant that

we could convert the issueOpenOrPullResponse from template function to

real function.

4. Moved open and pull constructors in CIMMessage.h to CIMMessage.cpp. NOTE:

openQueryInstances not moved. Will do when we get this funciton running.

5. Added test for correct namespace in pull operations. Note that it is

never really used but since supplied with request, we test to be sure it is

same as open request namespace. added internationalized exception for this.

6. We should probably add same test to close just as an input parameter

validation.

6. Modified processPullRequest to close the enumeration context if

incorrect response type received. Before we just returned exception and

left context open for next request. Closing is much more logical since the

client really did send an erronous request.

7. Modified PullErrors to correctly test above changed in behavior.

8. Removed a significant number of diagnostic messages.

9. Added query and querylanguage parameters to cimcli.

4 May 2014

1. Clean interface with EnumerationContext timeout thread

2. Remove more diagnostics

3. Correct issue with enumerationContext Release that was causing

failures during nightly tests.

2 May 2014

1. Remove Diagnostics from ResponseData.cpp

2. Correct setup and start of EnumerationContextTimerThread

3. Change limit on max number of enum contexts to be set from Dispatcher.

4. Increase size of hashtable for enum contexts.

5. Corrected issue causing failures in removeContext. Now runs through nightly

tests most of the time.

1 May 2014

1. Cleanup of code and variable names.

2. Removed namespace variable from EnumerationContext. It was never used.

3. Added test for max number of open enumerations. Right now it is fixed

and set about 100. Note that this is tied to the size of hash table

so that there is a issue with more dynamic setting of max and the

hash table size.

4. Removed several unused functions.

5. Found one possible issue causing thread conflict but not the core one

yet.

30 April

1. Added exception response to dispatcher for Limits exceeded on

CreateEnumerationContext. Returns Exceeded server limits CIMException.

2. Fixed possible deadlock where we were not unlocking EnumerationContext

under error condition

27 April - mergeout to head of tree

1. Mergeout of the code to the head of tree. Documented that mergeout in

the bug

27 April 2014

1. Clean up in a number of areas, mostly just code without changing

functionality.

2. Modified CIMMessage.h slightly to clarify names of the pull intermediate

messages (the common open and pull response);

3. Removed a number of diagnostics from CIMResponseData

4. Added new function to ResponseStressTestProvider to be able to terminate

output after a defined number of objects with a CIMException.

5. Removed a number of other TODOs

6. Corrected a couple of errors in the server message bundle.

7. Modified ResponseEncoder to only output the error for pull errors.

It was errounously outputting the EndOfSequence, EnumerationContext and

the error. This way, only the error is sent which is spec.

8. Removed number of TODOs and cleaned up a couple of others.

9. Added static tests for OpenEnumerateInstanceNames.

10. Modified a number of the temporary diagnostics to be permanent by

changing them to PEGASUS_DEBUG_ASSERTs

22 April 2014

1. Clean up some messages.

2. Still trying to find issue that randomly keeps messages in timeout

queue long after they have been closed and returned.

3. Fixed memory loss error.

4. Added zero reject for OperationTimeout because that effectively disables

the whole timeout mechanism so that enumerations could be started, never

completed by the client and would just hang around in memory. That

would be a significant DDOS attack (start very large enum with no

timeout) and never do pulls after open.

21 April 2014

1. Reorganized code in CIMOperationRequestDecoder.cpp ...Iparam classes.

Cleaned up the rejectCIMException code.

2. Cleanup some issues in the EnumerationContextTimeout Thread.

3. Fixed issue in pullop where it was using zero as default for default

operation timeout where it should have been NULL, the DMTF specified

default.

4. Enabled the test to reject 0 as operation timeout on a request and but

controled by a #define in CIMOperationRequestDispatcher.cpp

5. Added test for excessive new OperationContext entries in table.

18 April 2014

1. Internationalize remaining messages in CIMOperationRequestDecoder.cpp and

clean up the decoders for some of the messages so they use the

...Iparam classes. Note that we did not clean up all of the old message

decoders, just the new ones and we did not clean up the Count message

since it is deprecated. TODO clean these up so the decoder is really

consistent.

2. Enabled the EnumerationContext Timer and removed code for alternate

non-thread solution.

17 April 2014

1. Reorganized the lock/unlock for handling the EnumerationContext because

there were thread windows (i.e. using the context after it was released

by other thread).

2. Modified code for the InteroperationTimerThread and deleted old code that

allowed this whole thread to by bypassed.

3. Internationalized a number of CIMOperationRequestDispatcher CIMException

response messages.

4. Fixed stress tests in Client/tests/pullop that were getting timeout. The

issue was not the pull code but that fact that the idle client timeout

appears to be measureing to the end of the operation and the operation

was taking more than 40 seconds which is the default timeout. The idle

timeout takes no account of the fact that chunks are being regularly

delivered.

5. Several minor code changes.

NOTE: At this point the code passes extensive tests of the pull operations

both in OOP and none OOP mode (but with lots of diagnostics that may

be affecting timing).

10 April 2014

1. Fix issues causing failure with repeated pull operations.

2. Consolidate some of the CIMOperationRequestDispatcher.cpp template code

7 April 2104

MergeOut of head of tree.

7 April 2014

1. Permanently fixed issue with getting correct number of objects on response.

This was introduced when did a temp fix to the OOP processing, in particular

removed the condition variable wait in the CIMOperationDispatcher handle

functions for opens and pulls. This moved the functionality to kick off

open and pull responses to the provider threads if there are not enough

objects when the handle is processed. Also added a variable to allow us to

test with either a) responses are required to satisfy the original

request size or) b, responses are required only to return some objects.

We will add statistics to see which of these works best.

2. Modified WsmProcessor EnumerationContext class since the name conflicts

with the pull Operation EnumerationContext class.

3. Created a new intermediate level of CIMMessage, the CIMPullResponseData

Message so that all of the open and Pull responses can use common code

since there are only two variables (endOfSequence and EnumerationContext)

and they are common across all the open and pull responses.

4. Removed a number of diagnostics.

5. Cleaned up the Dispcatcher so that the open and pulls have common response

code and that code can be used from the dispatcher handle functions and

the provider response functions.

31 March 2014 - Checkin

1. Fixed issues in OOP processing of pull operations, in particular

issues with cimxml output format when processed through the

*InternalXmlEncoder functions.

2. Clean up some of the internalXml functionality

3. Found issues causing timeout with a particular provider. The issue

is that the dispatcher and monitor end up using the same thread so the

condition variable in the dispatcher thread stops the monitor. Turned

off the conditionVariable in getCache for the moment which means that

we get number of responses for open... with 0 objects before the

providers can begin to respond. This is only for test.

4. Added some statistics for enumerations and display the statistics

when we close the server (same as cache statistics)

12 March 2014 - Mergeout and Mergein

1. Mergeout to head of tree for this date and mergein for patch update

to bug 9676

2. Extensions to pullop tests program and tests.

3. Added some diagnostics in looking for OOP issue.

4. Removed a number of diagnostics messages and cleaned up code in

dispatcher to simplify pull operation processing.

15 December 2013

1. Mergeout and mergein up to 15 December 2013

2. Clean up issues from tests documented in bug 9676 last week.

3. Clean up some code in dispatcher

4. Remove the filter function from ResponseStressc++Provider.

21 November 2013

1. Mergeout from head of tree to 21 November 2013.

Legend:

Removed from v.1.1.2.16
changed lines
	Added in v.1.1.2.34

No CVS admin address has been configured