<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=iso-8859-1">
<META content="MSHTML 6.00.2800.1505" name=GENERATOR></HEAD>
<BODY style="MARGIN: 4px 4px 1px; FONT: 10pt Tahoma">
<DIV>I noticed if I (as WF Admin with not SAP_ALL / SAP_NEW) run SWWB </DIV>
<DIV>then I get two instances running. Once from me starting it and one </DIV>
<DIV>from our BASIS folks running. Does not stop other instances from running. </DIV>
<DIV> </DIV>
<DIV>
<DIV>If you look at the program there is references to this issues about </DIV>
<DIV>multiple instances running. How and who is scheduling RSWWDHEX? </DIV></DIV>
<DIV> </DIV>
<DIV>I would have BASIS kill all scheduled instances and have BASIS </DIV>
<DIV>create the batch run and call it a day. </DIV>
<DIV> </DIV>
<DIV>I would like to know what you do to get this issue resolved.</DIV>
<DIV> </DIV>
<DIV>Rick</DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV> </DIV>
<DIV>Rick Sample<BR>SAP Workflow Analyst/Developer<BR>Graybar, Inc.<BR>11885 Lackland Rd.<BR>63146-4208<BR>314.573.5822<BR><A href="mailto:Rick.Sample@GBE.com">Rick.Sample@GBE.com</A><BR><BR>>>> wug.replies@workflowconnections.com 7/13/2005 6:57 >>><BR>Hi Sue,<BR>You've definitely got two schedules running. Get rid of ALL deadline<BR>monitoring jobs and then schedule one job & you should be OK.<BR>Cheers<BR>Mike<BR><BR>Sue Keohan wrote:<BR>> Hello all,<BR>><BR>> Aaarrghh. Tonight the system log shows...<BR>><BR>><BR>> 19:40:32 sapapp1_RP1_23 BTC 28 400 EGC > Job SWWDHEX is<BR>> currently locked<BR>> 19:40:38 sapapp1_RP1_23 BTC 28 400 EGC > Job SWWDHEX is<BR>> currently locked<BR>><BR>> Thank you Larry, but SAP_REORG_JOBS runs after midnight, still more than<BR>> 4 hours from now.<BR>><BR>> What annoys me is that the job ran 17 times today, no problem. At<BR>> 19:40, two instances kicked off (under the same userid) and killed each<BR>> other.<BR>> Probably need to escalate this to OSS. Till then, SWWB might be my new<BR>> best friend.<BR>><BR>> Thanks all for the help,<BR>> Sue<BR>> <U><A href="mailto:lianghuan.x.hu@accenture.com">lianghuan.x.hu@accenture.com</A></U> wrote:<BR>><BR>>> We had the same problem. When I found that the background job<BR>>> SWWDHEX was not working, I used SWWB to have it fixed.<BR>>><BR>>><BR>>><BR>>> Here is an OSS explanation: "The job SWWDHEX that is required for date<BR>>> monitoring, sometimes no longer reschedules itself and therefore<BR>>> disappears almost inexplicably. ... The job SWWDHEX runs at the same<BR>>> time as the job SAP_REORG_JOBS. The latter blocks the table TBTCO and<BR>>> stops the job SWWDHEX from rescheduling itself. As SAP_REORG_JOBS<BR>>> usually takes several seconds, the implicit waiting that is built-in<BR>>> to the background management (see note 41037) does not suffice for the<BR>>> lock to be released in this case. You must therefore ensure that both<BR>>> jobs are not run at the same time. As SAP_REORG_JOBS usually runs at<BR>>> the most once a day, the execution of SWWDHEX is only prevented for<BR>>> the duration of this job." Our Basis team followed the instruction and<BR>>> since then we have not had this problem anymore.<BR>>><BR>>><BR>>><BR>>> I hope that the above info can help.<BR>>><BR>>><BR>>><BR>>> Larry Hu<BR>>><BR>>><BR>>> ------------------------------------------------------------------------<BR>>> *From:* <U><A href="mailto:sap-wug-bounces@mit.edu">sap-wug-bounces@mit.edu</A></U> <U><A href="mailto:[mailto:sap-wug-bounces@mit.edu]">[mailto:sap-wug-bounces@mit.edu]</A></U> *On<BR>>> Behalf Of *Rick Sample<BR>>> *Sent:* Tuesday, July 12, 2005 9:11 AM<BR>>> *To:* <U><A href="mailto:sap-wug@mit.edu">sap-wug@mit.edu</A></U> <BR>>> *Subject:* Re: rswwdhex - job gets cancelled due to locking issue<BR>>><BR>>> Had the same issue. Here is what I found and what we did to correct.<BR>>><BR>>> If multiple instances of SWWDHEX are running we get locks. I verified<BR>>> and<BR>>> can reproduce this and fixed. Here is what I did.<BR>>> Kill all SWWDHEX and have BASIS reschedule just one.<BR>>> Schedule SWWDHEX to run every X time.<BR>>><BR>>> Still don't know why it stopped but I assume I ran the config as<BR>>> myself. SWU3<BR>>> and caused mult instances to be running.<BR>>> BASIS probably runs SWU3 for the simple fact that it SWU3 should only<BR>>> be run<BR>>> by a user that has SAP_ALL / SAP_NEW.<BR>>> Reason: WF-BATCH inherits the rights of the user running SWU3. No<BR>>> more. No less.<BR>>> (You can search for responses from Jocelyn Dart and in "Practical WF<BR>>> for SAP".)<BR>>><BR>>> WF-BATCH should not be used for batch processing like<BR>>> SWWDHEX, SWWERRE, SWWERRE, etc.<BR>>> Use some other batch user for batch processing.<BR>>><BR>>> SWWA set to run periodically. Like every 5 mins.<BR>>> SWWB runs per instance of a deadline.<BR>>> (I could have these backwards. Look them up to be sure)<BR>>><BR>>> SWWA or SWWB? Depends on what you want.<BR>>> If you have few deadlines, run per instance and it will add a batch for<BR>>> each instance. See SM37 and you will see these being added per deadline.<BR>>> Or, if you have lots of deadlines. set to just periodically, it<BR>>> should rescheduled ONE<BR>>> instance to run every X time.<BR>>><BR>>> Let me know if this helps. I am interested in what works for you.<BR>>> (And if I made any incorrect statement I am sure someone will let me<BR>>> know)<BR>>><BR>>><BR>>><BR>>><BR>>><BR>>><BR>>><BR>>><BR>>> Rick Sample<BR>>> SAP Workflow Analyst/Developer<BR>>> Graybar, Inc.<BR>>> 11885 Lackland Rd.<BR>>> 63146-4208<BR>>> 314.573.5822<BR>>> <U><A href="mailto:Rick.Sample@GBE.com">Rick.Sample@GBE.com</A></U> <<U> <A href="mailto:Rick.Sample@GBE.com">mailto:Rick.Sample@GBE.com</A></U> ><BR>>><BR>>> >>> <U><A href="mailto:keohan@ll.mit.edu">keohan@ll.mit.edu</A></U> 7/12/2005 7:34 >>><BR>>> Hi all,<BR>>><BR>>> We are running on R/3 4.6c.<BR>>><BR>>> We had RSWWDHEX running without incident until last September, when it<BR>>> stopped, I don't know why.<BR>>> When the matter was brought to my attention, we re-scheduled it via<BR>>> SWWB (or is it SWWA, I can't be<BR>>> sure because Basis does it). The interval between runs is 20 minutes,<BR>>> which should be sufficient,<BR>>> since there are probably less than 5 workitems 'waiting' for a<BR>>> deadline at any given time.<BR>>><BR>>> Anyway, the job gets finishes, but fails to reschedule itself daily<BR>>> (no specific time, it could fail<BR>>> to reschedule at 5:30, it could fail to reschedule at 7:30...). The<BR>>> system log from SM21 says<BR>>> 'Job SWWDHEX is currently locked' and the detail entry follows:<BR>>><BR>>> > Within background processing, it was found that the specified job is<BR>>> > being edited, for example, another user edits the job or the run-time<BR>>> > system of background processing is accessing the job. Please attempt<BR>>> to<BR>>> > execute the action required by you with the job at a later date.<BR>>> ><BR>>> > If, in spite of repeated attempts, this should not be successful, that<BR>>> > could be related to problems in the locking system (enqueue server):<BR>>> For<BR>>> > example, "locking corpses" could be available. Please check that with<BR>>> > the transaction SM12. You can also test the function efficiency of<BR>>> your<BR>>> > locking system there by means of the displayed diagnosis.<BR>>> ><BR>>><BR>>> I am sure we can reschedule with a longer interval between runs, but<BR>>> not sure if this will actually<BR>>> solve the problem. Has anyone out there dealt with this ? OSS, and SDN<BR>>> search of this archive<BR>>> yield no joy.<BR>>><BR>>> Happy WF-ing!<BR>>> Sue<BR>>> --<BR>>> Susan R. Keohan<BR>>> SAP Workflow Developer<BR>>> MIT Lincoln Laboratory<BR>>> 244 Wood Street<BR>>> LI-200<BR>>> Lexington, MA. 02420<BR></DIV></BODY></HTML>