1 karl 1.1.2.1 Using the CIM/XML Pull Operations
2
3 STATUS
4
|
5 karl 1.1.2.4 <<< The TODO section is being maintained during the review and checkin process
|
6 karl 1.1.2.1 to keep track of problems, errors, notes, etc. Must be deleted before
7 checkin to head of tree. Please feel free to add notes, etc in this
8 section as you review/test.>>>>>>
9
|
10 karl 1.1.2.4 TODO list:
|
11 karl 1.1.2.3 1. Binary operation from OOP. Need to add counter to binary
12 protocol to be able to count objects in response. Generates
13 warnings in things like messageserializer and does not work with
|
14 karl 1.1.2.30 OOP right now. Fixed by converting to XML.
|
15 karl 1.1.2.23 2. OpenExecQuery - Code is incomplete in that it does not include the
|
16 karl 1.1.2.15 return from the exec query function to the aggregator yet.
|
17 karl 1.1.2.4 3. Code for Pull part of OpenQueryInstancesRequest a) should be part of
18 the common CIMOperationRequestDispatcher execCommon code.
19 4. The changes to WQLCIMOperationRequestDispatcher and CQL... for handling
20 pull not completed so we feed the responses back to the EnmerationContext
21 queues
|
22 karl 1.1.2.27 3. Minor TODOs, diagnostics, etc. still in the code
|
23 karl 1.1.2.14 4. External runtime variables. Proposing that they be fixed for this release
|
24 karl 1.1.2.15 rather than set by configuration. This should be discussed. Am making
|
25 karl 1.1.2.16 this a separate bug. See bug 9819 for the changes to cover this.
|
26 karl 1.1.2.29 5. Move processPull code in dispatcher from template to
27 common function which would mean adding intermediate classes in
28 CIMMessage but would reduce code size. This createsCIMmessage
29 CIMPullRequestData... between CIMRequestDataMessage and the Pull requests
30 since they all have exactly the same input and output pararameters.
|
31 karl 1.1.2.23 6. Extension to avoid double move of objects in CIMResponseData (one
|
32 karl 1.1.2.3 into enumerationContext queue and second to new cimResponseData for
33 response. Want to avoid second move by extending Open/Pull response
34 messages to include count and CIMResponse data to count objects out
|
35 karl 1.1.2.23 of queue when converting (avoids the second move). This would mean
36 extending the output writers to allow the count field to be supplied
|
37 karl 1.1.2.30 so they would only create output for up to the count supplied.(Make
38 this future beyond bug 9676)
|
39 karl 1.1.2.28 9. Add more static tests (currently only OpenEnumerateInstances and
|
40 karl 1.1.2.30 OpenEnumerateInstanceNames covered).
|
41 karl 1.1.2.26 Low priority because most of this except exact XML covered in other
|
42 karl 1.1.2.28 tests. note that we can only add very limited tests since a) no
43 way to test multiple opertions in sequence(i.e. open/pull) and
44 cannot be sure how many responses will come back in all cases even
45 for open.
|
46 karl 1.1.2.26 10. Could clean up the CIMOperationRequestDecoder.cpp rejectCIMException
|
47 karl 1.1.2.29 further since there is some overlap. Low priority.
48 11.Correct issue between operations and HTTP where we are sending
|
49 karl 1.1.2.26 trailers with exceptions. Also correct issue where we are
50 sending endOfSequence and EnumerationContext in response when
|
51 karl 1.1.2.28 we send CIMException. DONE Except that we really need to get
52 rid of the trailers completely for pull operations. Need to
53 have flag to send as non trailer (probably non_chunked flag)
|
54 karl 1.1.2.30 12.Make enumContextTable hashtable dynamic and create entirely within
55 CIMOperationRequestDispatcher.cpp. Today it has a static size
|
56 karl 1.1.2.26
57 NEXT TASKS:
|
58 karl 1.1.2.23 b. Finish and test the OpenQueryInstances
|
59 karl 1.1.2.16 c. Clean up TODOs
|
60 karl 1.1.2.23 d. Find issue that is causing occasional failures under load. It
61 is probably a threading issue between the 3 possible threads
62 that use the EnumerationContext (request thread, provider thread,
|
63 karl 1.1.2.24 timeout thread) and a place where not mutuxed, primarily the
64 timeout thread.
|
65 karl 1.1.2.23
|
66 karl 1.1.2.30 4 May 2014
67 1. Clean interface with EnumerationContext timeout thread
68 2. Remove more diagnostics
69 3. Correct issue with enumerationContext Release that was causing
70 failures during nightly tests.
71
|
72 karl 1.1.2.29 2 May 2014
73 1. Remove Diagnostics from ResponseData.cpp
74 2. Correct setup and start of EnumerationContextTimerThread
75 3. Change limit on max number of enum contexts to be set from Dispatcher.
76 4. Increase size of hashtable for enum contexts.
77 5. Corrected issue causing failures in removeContext. Now runs through nightly
78 tests most of the time.
|
79 karl 1.1.2.28 1 May 2014
80 1. Cleanup of code and variable names.
81 2. Removed namespace variable from EnumerationContext. It was never used.
82 3. Added test for max number of open enumerations. Right now it is fixed
83 and set about 100. Note that this is tied to the size of hash table
84 so that there is a issue with more dynamic setting of max and the
85 hash table size.
86 4. Removed several unused functions.
87 5. Found one possible issue causing thread conflict but not the core one
88 yet.
89
90 30 April
91 1. Added exception response to dispatcher for Limits exceeded on
92 CreateEnumerationContext. Returns Exceeded server limits CIMException.
93 2. Fixed possible deadlock where we were not unlocking EnumerationContext
94 under error condition
95
96
|
97 karl 1.1.2.27 27 April - mergeout to head of tree
98 1. Mergeout of the code to the head of tree. Documented that mergeout in
99 the bug
100
101 27 April 2014
102 1. Clean up in a number of areas, mostly just code without changing
103 functionality.
104 2. Modified CIMMessage.h slightly to clarify names of the pull intermediate
105 messages (the common open and pull response);
106 3. Removed a number of diagnostics from CIMResponseData
107 4. Added new function to ResponseStressTestProvider to be able to terminate
108 output after a defined number of objects with a CIMException.
109 5. Removed a number of other TODOs
110 6. Corrected a couple of errors in the server message bundle.
111 7. Modified ResponseEncoder to only output the error for pull errors.
112 It was errounously outputting the EndOfSequence, EnumerationContext and
113 the error. This way, only the error is sent which is spec.
114 8. Removed number of TODOs and cleaned up a couple of others.
115 9. Added static tests for OpenEnumerateInstanceNames.
116 10. Modified a number of the temporary diagnostics to be permanent by
117 changing them to PEGASUS_DEBUG_ASSERTs
118 karl 1.1.2.27
|
119 karl 1.1.2.24 22 April 2014
120 1. Clean up some messages.
121 2. Still trying to find issue that randomly keeps messages in timeout
122 queue long after they have been closed and returned.
|
123 karl 1.1.2.25 3. Fixed memory loss error.
|
124 karl 1.1.2.26 4. Added zero reject for OperationTimeout because that effectively disables
125 the whole timeout mechanism so that enumerations could be started, never
126 completed by the client and would just hang around in memory. That
127 would be a significant DDOS attack (start very large enum with no
128 timeout) and never do pulls after open.
|
129 karl 1.1.2.25
|
130 karl 1.1.2.23 21 April 2014
|
131 karl 1.1.2.27 1. Reorganized code in CIMOperationRequestDecoder.cpp ...Iparam classes.
|
132 karl 1.1.2.23 Cleaned up the rejectCIMException code.
133 2. Cleanup some issues in the EnumerationContextTimeout Thread.
134 3. Fixed issue in pullop where it was using zero as default for default
135 operation timeout where it should have been NULL, the DMTF specified
136 default.
137 4. Enabled the test to reject 0 as operation timeout on a request and but
138 controled by a #define in CIMOperationRequestDispatcher.cpp
139 5. Added test for excessive new OperationContext entries in table.
140
141 18 April 2014
142 1. Internationalize remaining messages in CIMOperationRequestDecoder.cpp and
143 clean up the decoders for some of the messages so they use the
144 ...Iparam classes. Note that we did not clean up all of the old message
145 decoders, just the new ones and we did not clean up the Count message
146 since it is deprecated. TODO clean these up so the decoder is really
147 consistent.
148 2. Enabled the EnumerationContext Timer and removed code for alternate
149 non-thread solution.
|
150 karl 1.1.2.16
|
151 karl 1.1.2.21 17 April 2014
152 1. Reorganized the lock/unlock for handling the EnumerationContext because
153 there were thread windows (i.e. using the context after it was released
154 by other thread).
155 2. Modified code for the InteroperationTimerThread and deleted old code that
156 allowed this whole thread to by bypassed.
157 3. Internationalized a number of CIMOperationRequestDispatcher CIMException
158 response messages.
|
159 karl 1.1.2.22 4. Fixed stress tests in Client/tests/pullop that were getting timeout. The
160 issue was not the pull code but that fact that the idle client timeout
161 appears to be measureing to the end of the operation and the operation
162 was taking more than 40 seconds which is the default timeout. The idle
163 timeout takes no account of the fact that chunks are being regularly
164 delivered.
165 5. Several minor code changes.
166 NOTE: At this point the code passes extensive tests of the pull operations
167 both in OOP and none OOP mode (but with lots of diagnostics that may
168 be affecting timing).
|
169 karl 1.1.2.21
|
170 karl 1.1.2.20 10 April 2014
171 1. Fix issues causing failure with repeated pull operations.
172 2. Consolidate some of the CIMOperationRequestDispatcher.cpp template code
173
|
174 karl 1.1.2.19 7 April 2104
175 MergeOut of head of tree.
176
177 7 April 2014
178 1. Permanently fixed issue with getting correct number of objects on response.
179 This was introduced when did a temp fix to the OOP processing, in particular
180 removed the condition variable wait in the CIMOperationDispatcher handle
181 functions for opens and pulls. This moved the functionality to kick off
182 open and pull responses to the provider threads if there are not enough
183 objects when the handle is processed. Also added a variable to allow us to
184 test with either a) responses are required to satisfy the original
185 request size or) b, responses are required only to return some objects.
186 We will add statistics to see which of these works best.
187 2. Modified WsmProcessor EnumerationContext class since the name conflicts
188 with the pull Operation EnumerationContext class.
189 3. Created a new intermediate level of CIMMessage, the CIMPullResponseData
190 Message so that all of the open and Pull responses can use common code
191 since there are only two variables (endOfSequence and EnumerationContext)
192 and they are common across all the open and pull responses.
193 4. Removed a number of diagnostics.
194 5. Cleaned up the Dispcatcher so that the open and pulls have common response
195 karl 1.1.2.19 code and that code can be used from the dispatcher handle functions and
196 the provider response functions.
197
|
198 karl 1.1.2.18 31 March 2014 - Checkin
199 1. Fixed issues in OOP processing of pull operations, in particular
200 issues with cimxml output format when processed through the
201 *InternalXmlEncoder functions.
202 2. Clean up some of the internalXml functionality
203 3. Found issues causing timeout with a particular provider. The issue
204 is that the dispatcher and monitor end up using the same thread so the
205 condition variable in the dispatcher thread stops the monitor. Turned
206 off the conditionVariable in getCache for the moment which means that
207 we get number of responses for open... with 0 objects before the
208 providers can begin to respond. This is only for test.
209 4. Added some statistics for enumerations and display the statistics
210 when we close the server (same as cache statistics)
211
212 12 March 2014 - Mergeout and Mergein
213 1. Mergeout to head of tree for this date and mergein for patch update
214 to bug 9676
215 2. Extensions to pullop tests program and tests.
216 3. Added some diagnostics in looking for OOP issue.
217 4. Removed a number of diagnostics messages and cleaned up code in
218 dispatcher to simplify pull operation processing.
219 karl 1.1.2.18
|
220 karl 1.1.2.17 15 December 2013
221 1. Mergeout and mergein up to 15 December 2013
222 2. Clean up issues from tests documented in bug 9676 last week.
223 3. Clean up some code in dispatcher
224 4. Remove the filter function from ResponseStressc++Provider.
225
|
226 karl 1.1.2.16 21 November 2013
227 1. Mergeout from head of tree to 21 November 2013.
|
228 karl 1.1.2.15
229 18 November 2013
230 1. Cleanup of a bunch of minor errors and completion of all of the code for
231 the openQueryInstances except for the PullInstances in Dispatcher and
232 the aggregator function.
233 2. OpenqueryInstances added to cimcli.
|
234 karl 1.1.2.14
235 13 October 2013 CVS branch update.
236 1. Integrated bug 9786 into the branch. Note that we need to test the
237 generated statistics.
238 2. Mergeout executed to update to head of tree as of 8:00 am 13 October 2013.
239 3. Cleaned up several errors in OOP processing. Note that there is at least
240 one issue left when we to a pull on ManagedElement in at least one of the
241 namespaces.
242 4. Cleaned up some of the outstanding diagnostic code
243 5. Generally passes all tests except for one test of pullop where it is trying
244 to pull enum instances CIM_ManagedElement from a particular namespace.
|
245 karl 1.1.2.12
|
246 karl 1.1.2.16 NOTE: I did not make comments here for changes in October despite the fact
247 that I did 2 mergouts, number of fixes, and a mergein.
248
|
249 karl 1.1.2.12 30 September 2013 - CVS Update
250 Mergeout head of tree up to 29 September 2013.
|
251 karl 1.1.2.10
252 29 September 2013. CVS update.
253 1. Modified calls to statisticalData.cpp to a) directly call with request
254 type, b) incorporate the open, pull, etc. messages. However, since these
255 are not part of the CIM class, we must do something special with them.
|
256 karl 1.1.2.13 See bug 9785 for full solution to this issue.
|
257 karl 1.1.2.10 2. Corrected OOP interface to enable new flag to indicate internal operations
258 and set host, etc.
259 3. Add code to CQLOperationsDispatcher and WQLOperationDispatcher to clean
260 up CIMResponseDataCounter after filtering.
261 4. Modified ProviderAgent to set Host info for some pull operations.
262 5. Added new flag to CIMBinMsgSerializer and Deserializer.
|
263 karl 1.1.2.8
|
264 karl 1.1.2.9 17 September 2013 CVS update (Actually two different updates over 3 days)
265 1. Clean up some issues in CIMMessage.h and CIMMessage.cpp
266 2. Extend OpenExecQuery to WQL and CQL processors but return not complete
267 3. Remove memory leak in EnumerationContext and EnumerationContextTable
268 handling.
269 4. Created template functions for much of the pull operations.
270 5. Reversed order of queryLanguage and query (and changed names to match
271 execQuery) in client and server. Note that these are the execQuery
272 WQL and CQL filters and NOT FQL filters.
273 6. Some code cleanup in dispatcher
274 7. Today, passes all tests in pullop but issue in alltests. For some reason
275 not finding CIMObjectManager instance. Also, leaves enumeration contexts
276 if client terminates since cleanup thread not operating.
|
277 karl 1.1.2.11 8. XML from OOP not correctly processed.
|
278 karl 1.1.2.9
|
279 karl 1.1.2.8 14 September 2013 CVS update
280 Merged out up to 25 August. Cleaned up all operations and standardized code.
281 At this point the non pull operations code is in a set of templates but the
282 pull is not yet.
283 Fixed a significant number of problems so that it appears that the operations
284 except for OpenExecQuery run stably, at least with the pullop test program.
285 Note that there is a problem in that the Interop control provider is not
286 returning its singleton wbemserver object for some reason. Causes a test
287 failure
|
288 karl 1.1.2.7
289 Fixed for 16 June CVS Update
290 1. Cleaned up the enumerationContext and Table release functions and tested
291 to confirm that we do not lose memory in either normal sequences or
292 sequences that close early. Cleaned up pullop and added more tests
|
293 karl 1.1.2.8 Taged Before: PREAUG25UPDATE and after POSTAUG25UPDATE
|
294 karl 1.1.2.4
295 Fixed for 9 June CVS update
296 1. Cleaned up code for OpenQueryInstances. Note that this is incomplete.
297 No support in WQL or CQL Operations
298 2.
299
300 What was fixed for 5 June checkin.
|
301 karl 1.1.2.3 1. Extended ResponseTest MOF for for both CMPI and C++ subclasses
302 2. Fixed issues with pullop.
303 3. Fixed temp issue with CIMResponseData size by putting in mutex. That
304 is not a permanent fix but it gets around issue probably in the control
305 of the move logic that meant counts were off.
306 4. Fixed issues in Dispatcher so that associator code works. Still messy
307 code in the dispatcher.
308 5. Changed name of Enumerationtable.h & cpp to EnumerationContextTable.*
309 6 Changed name of ResponseStressTest module, classes, etc.
310
311 TAG: TASK_PEP317_5JUNE_2013_2
312
|
313 karl 1.1.2.2 2 June 2013
|
314 karl 1.1.2.1
315 Issues - KS
316
|
317 karl 1.1.2.15 - Still way to many TODO and KS comments and KS_TEMPS. Removing bit by bit.
|
318 karl 1.1.2.1
|
319 karl 1.1.2.15 - Runtime variable connection for the config parameters not installed. That
320 has been made into a separate bug (see bug 9819)
|
321 karl 1.1.2.1
322 5. Issue with the threaded timer. For some reason during tests it
323 eventually calls the timer thread with trash for the parm (which is
324 pointer to the EnumerationTable object). Caught because we do a valid
325 test at beginning of the function.
326
|
327 karl 1.1.2.2 6. Still using the templates in CIMOperationRequestDispatcher to simplify
328 the handle... processing.
329
330 7. I think I have a way around the double move of objects in the
331 EnumerationContext so that the outputter will just take a defined number
332 of objects directly from the gathering cache and save the second move.
333
|
334 karl 1.1.2.15 8. Not yet passing all tests but getting closer now. The major test that is
335 causing an error today is the execution of a full enumeration with the
336 forceProviders = true. This causes a client timeout sometimes.
|
337 karl 1.1.2.2
338
339
|
340 karl 1.1.2.1 ===========================================
341
342 OVERVIEW:
343
344 The operation extensions for pull operations defined in the DMTF specification
345 DSP0200 V 1.4 were implemented in Pegasus effective Pegasus version 2.11
346 including Client and Server.
347
348 These operations extend the CIM/XML individual operations to operation
349 sequences where the server must maintain state between operations in a
350 sequence and the client must execute multiple operations to get the full
351 set of instances or instance paths.
352
353 The following new CIM/XML operations as defined in DSP0200 are included;
354
355 -OpenEnumerateInstances
356 -openEnumerateInstancePaths
357 -OpenReferenceInstances
358 -OpenReferenceInstancePaths
359 -OpenAssociatiorInstances
360 -OpenAssociatorInstancePaths
|
361 karl 1.1.2.16 -OpenQueryInstances
|
362 karl 1.1.2.1 -PullInstancesWithPath
363 -PullInstancePaths
|
364 karl 1.1.2.16 -PullInstances
|
365 karl 1.1.2.1 -CloseEnumeration
366 -EnumerationCount
|
367 karl 1.1.2.2 OpenExecQuery
|
368 karl 1.1.2.1
369 The following operations have not been implemented in this version of Pegasus:
370
371 -OpenQueryInstances
372
373 The following limitations on the implementation exist;
374
375 1. The filterQueryLanguage and filterQuery parameters are processed by
376 the Pegasus client but the server returns error if there is any data in
|
377 karl 1.1.2.2 either parameter. This work does not include the development of the
378 query language. Note that a separate effort to extend Pegasus to use
379 the DMTF FQL query language is in process.
|
380 karl 1.1.2.1
381 2. The input parameter continueOnError is processed correctly by the client
382 but the Pegasus server only provides for false since the server does not
383 include logic to continue processing responses after an error is
384 encountered.
385 This is consistent with the statement in the specification that use of
386 this functionality is optional and the fact that the DMTF agrees that all
387 of the issues of continuing after errors have not been clarified.
388
389 3. The operation enumerationCount is not processed by the server today since
390 a) really getting the count would be the same cost as the corresponding
391 enumeration, b) the server does not include a history or estimating
392 mechanism for this to date.
|
393 karl 1.1.2.2 NOTE: After a through review as part of the development of the next version
394 of CMPI we have concluded that this operation is probably not worth the
395 effort. Since it is optional, Pegasus will only return the unknown status
396 at this point
|
397 karl 1.1.2.1
398 Since the concept of sequences of operations linked together (open, pull, close)
399 is a major extension to the original CIM/XML operation concept of completely
400 independent operations several new pieces of functionality are implemented
401 to control interOperationTimeouts, counts of objects to be returned, etc.
402
403 TBD - Review this
404
405 CLIENT
406
407 The new operations follow the same pattern as the APIs for existing operations
408 in that:
409
410 1. All errors are handled as CIMException and Exception
411
412 2. The means of inputting parameters are the same except that there are
413 significantly more input parameters with the open operations and for the
414 first time operations return parameters as well as objects in the
415 response. Specifically the open and pull operations return values for
416 enumerationContext which is the identity for a pull sequence and
417 endOfSequence which is the marker the server sends in open and pull
418 karl 1.1.2.1 responses when it has no more objects to send.
419
420 The significant differences include:
421
422 1. Processing of parameters on responses (i.e. the endOfSequence and
423 enumerationContext parameters are returned for open and pull operations).
424
425 2. Numeric arguments (Uint32 and Uint64 include the option of NULL in some
426 cases so they are packaged inside classes Uint32Arg and Uint64Arg in the
427 client api.
428
429 3. The association and reference operations ONLY process instances. They do
430 not include the capability to return classes like reference and associator
431 do and therefore return CIMInstance rather than CIMObject.
432
433 4. Paths are returned in all cases (i.e OpenEnumerateInstances and
434 PullInstancesWithPath where they were not with EnumeratInstances.
435
436 5. The client must maintain state between operations in a sequence (using
437 the enumerationContext parameter).
438
439 karl 1.1.2.1 TBD- Are there more differences.
440
441
442 SERVER
443
444 The Pegasus server attempts to always deliver the requested number of objects
445 for any open or pull request (the specification allows for the server to
446 deliver less than the requested number of objects and specifically to return
447 zero objects on open). We felt that it was worth any extra cost in processing
448 to provide the client with exactly what it had requested.
449
450 The pegasus server always closes an enumeration sequence upon receipt of any
451 error from the providers, repository, etc. Therefore the server will reject
|
452 karl 1.1.2.2 any request that has continueOnError = true;
453
454 Expansion to allow the continue on error may be added in a future version.
455 In any case, the whole purpose of the continue on error is really to allow
456 input from good providers to be mixed with providers that return errors so
457 that generally this would mean simply changing the logic in the return mechanism
458 to not shutdown when an error is received from any given provider.
459
460 Generally we do not believe that the providers need to do much more in the
461 future to support the continueOnError other than possibly allowing the provider
462 to continue processing after it has received an error.
|
463 karl 1.1.2.1
464 PROVIDERS
465
466 This implementation requires NO changes to the existing providers. The
467 provider APIs operate just as they do with the original operations.
468
469 Because the server processing is different however, there may be some
470 behavior differences primarily because the client now controls the speed of
471 delivery of objects.
472
473 In previous versions of Pegasus, the server attempts to deliver objects as
474 rapidly as then can be put on the network. In the case of HTTP chunked requests
475 they are delivered in chunks of about 100 objects. The primary delay for the
476 providers was the processing of each segment through the server. The server
477 is blocked so that no other segment can proceed through the server until that
478 segment is processed and sent on the network.
479 In the case of non-chunkedresponses, they are completely gathered in the serve
480 and then delivered as one non-chunked response. There were no delays for the
481 providers, just lots of possible memory use in the server.
482
483 The responses from providers (delivered through the deliver(...) interface are
484 karl 1.1.2.1 gathered into segments of about 100 objects and this group of objects is moved
485 through the server to be delivered to the client.
486
487 However with the inclusion of the pull operations, The segments of objects
488 from the providers are cached in the server response path until the
489 maxObjectCount for that request (open or pull) and that number returned in a
490 non-chunked response. Thus, if the client is slow to issue pull requests,
491 the providers might be delayed at some point to reduce memory usage in the
492 server (the delay appears as slow response tothe deliver operation).
493
494 In other words, the time to process large sets of responses from the provider
495 now depends on the speed of handling the client.
496
497 It is important to remember in developing providers that the Pegasus server
498 can most efficiently process responses if they are passed from the provider
499 to the server individually or in small arrays of objects rather than the
500 provider gathering very large arrays of objects and sending them to the
501 server.
502
|
503 karl 1.1.2.2 NEXT GENERATION PROVIDERS
504 KS_TODO
505
|
506 karl 1.1.2.1 CONFIGURATION PARAMETERS
507
508 The server includes several configuration parameters to set limits on the
509 processing of pull operations. All of these configuration parameters are
510 compile time parameters rather than runtime.
511
512 1. Maximum value of minimum interoperation time. This parameter defines the
513 maximum time allowed between the return of an open or pull response and
514 the receipt of the next pull or a close operation before the server may
515 close the enumeration. The specification allows the server to set a
516 maximum interoperation time and refuse open requests that with requested
517 operationTimeout greater than that time.
518 CIM_ERR_INVALID_OPERATION_TIMEOUT
519
520 This value is set with the Pegasus environment variable
521 PEGASUS_PULL....
522
523 2. Maximum objects returned in a single open or pull operation. The server
524 can set a maximum limit on the number of objects that can be returned in
525 a single open or pull oepration with the maxObjectCount parameter.
526
527 karl 1.1.2.1 3. Whether the server allows 0 as an interoperation timeout value. The value
528 zero is s special value for the interoperationTimeout in that it tells the
529 server to not timeout any enumeration sequence.
530
531 With this value for interoperationTimeout, the only way to close an
532 enumeration sequence is to complete all of the pulls or issue the close.
533 If for some reason the sequence is not completed, that enumeration context
534 would remain open indefinitly. Since in Pegasus any open enumeration
535 context uses resources (the context object and any provider resposnes that
536 have not yet been issued in a response) it would appear that most
537 platforms would not want to allow the existence of enumeration contexts
538 that cannot be closed by the server.
539
540 4, maximum consecutive pull requests with 0 maxObjectCount. The use of the
541 pull operation with maxObjectCount set to zero could be used to keep an
542 enumeration context open indefinitly (this tells the server to restart the
543 interoperationTimeout but not send any objects in the response). Therefore the
544 specification allows for the server setting maximum limits on this behavior
545 and returning the error CIM_ERR_SERVER_LIMITS_EXCEEDED if this limit is
546 exceeded.
547 Note that this is maximum CONSECUTIVE pulls so that issuing a pull with
548 karl 1.1.2.1 a non-zero count resets this counter.
549
550 KS-TBD - Is this really logical since we can still block by just issuing
551 lots of zero request and an occansional request for one object.
552
553 Pegaus sets the value of this limit to 1000 and allows the implementer to
554 modify it with the PEGASUS_MAXIMUM_ZERO_OBJECTCOUNT environment variable.
555
556 5. Default operationTimeout -
557
558 The default of this parameter is to refuse operat
559
560 In the current release of Pegasus these are all compile time parameters.
|
561 karl 1.1.2.11
562
563 NOTES On working with task branch.
564
565 Merge out Process
566
567 To keep our TASK branch in sync with the current head of tree we need
568 to do a regular merge out. the TaskMakefile contains the makefile
569 procedures to do this efficiently. NOTE: Following these procedures is
570 important in that you are merging out new material each time you do
571 the merge out. If you were just to repeatedly merge out, you would be
572 merging previously merged changes a second time causing a real mess.
573
574 Start with new directory and put TaskMakefile above pegasus (needed so you
575 have this file for the initial operations.
576
577 make -f TaskMakefile branch_merge_out BNAME=PEP317-pullop ## takes a long time
578
579 This checks out current head, merges it into task branch and sets tags
580 for the mergeout. Note that at the end of this step this work is
581 part of the TASK... branch.
582 karl 1.1.2.11
583 NOW check for conflicts, errors, etc. that resulted from the merge.
584 Look for conflict flags, compare the results (I use linux merge as a
585 good graphic compare tool) and build and test. When you are satisfied
586 that the merge out is clean, you can commit the results to the TASK...
587 branch
588
589 To commit the work to this into Task branch
590
591 make -f mak/TaskMakefile branch_merge_out_commit BNAME=PEP317-pullop
592
593 or manually commit and finish as follows
594
595 cvs commit
596 make -f mak/TaskMakefile branch_merge_out_finish BNAME=PEP317-pullop
597
598 ## This last step is important since it cleans up temporary tags to prepare
599 you for the next checkout
600
601 COMPARE TASKBRANCH WITH HEAD
602
603 karl 1.1.2.11 In a new pegasus work space do same as above for merge out.
604
605 make -f TaskMakefile BNAME=PEP317-pullop
606
607 This produces a result which is all of the head merged into the branch.
608 A diff of this is all the new changes to the head of tree that you will
609 include into the merge.
610
|
Return to
readme.pulloperations
CVS log
Up to [Pegasus] / pegasus