AW: AW: Disaster Recovery scenario for Unix box? [Deutsche Boerse Systems: Virus checked]
Hello Stefan, I do not consider it appropriate or polite to make derogatory or insulting remarks about specific list members on this public list - in any language. By the way "deformation of character" can be very expensive for you and/or your company in a number of countries. If you wish to "rant" with Hubert about the list or other members of the list, why don't you do it among yourselves? The assumption that native English speakers would not understand your German comments is quite a risk. Regards, Carl -Ursprüngliche Nachricht- Von: Stefan Raabe [mailto:[EMAIL PROTECTED] Gesendet: Montag, 29. September 2003 10:33 An: [EMAIL PROTECTED] Betreff: Re: AW: Disaster Recovery scenario for Unix box? [Deutsche Boerse Systems: Virus checked] hallo hubert, nur falls du es nicht gemerkt haben solltest. die rebecca bullock und noch irgend so eine andere tante in der mq liste schwafeln meistens nur dummes zeugs, meistens verstehen sie die frage schon falsch und antworten dann in die total falsche richtung. oder sie antworten falsche sachen, oder dummes allgemeines zeugs, oder oder. von denen hab ich noch kaum was wertvolles gesehen, daher wuerde ich mich mit denen auch nicht auf irgendwelche diskussionen einalssen. pseudo-experten-zicken-blablabla, die sich aber fuer supertolle mq experten halten. würg. gruss stefan ps: raetsel aus der neuen bdw 6 stellige telefonnummer. wenn du die zahl * 5 nimmst dann bekommst du die gleiche Zahl nur mit der letzten nach vorne. abcdef * 5 = fabcde wenn du erstmal selbst basteln willst dann nicht unten weiterlesen... :-)) ansich auch wieder recht einfach, auch ohne programmausprobieren, wobei ich nicht fuer jede zahl (abcdef) eine randbedingung habe. so muss a = 1 sein, 4 < f < 10, ich habe mich halt von bedingung zu bedingung gehangelt (also erstmal alle f * 5 genommen und dabei halt ermittelt, das e 0 oder 5 ist und so weiter) >= 10 <= 20 x >=5 x <= 9 a = 1 abcdex * 5 = xabcde 5x = ?e x e e = 5 ! 0 5 * 5 = 25 6 * 5 = 30 7 * 5 = 35 8 * 5 = 40 9 * 5 = 45 ex de d = 7 ! 8 ! 9 ! 3 ! 4 55 * 5 = 275 56 * 5 = 280 57 * 5 = 285 58 * 5 = 290 59 * 5 = 295 05 * 5 = 025 06 * 5 = 030 07 * 5 = 035 08 * 5 = 040 09 * 5 = 045 dex cde c = 7 ! 2 ! 5 ! 0 755 * 5 = 3775 757 * 5 = 3785 759 * 5 = 3795 706 * 5 = 3530 708 * 5 = 3540 855 * 5 = 4275 857 * 5 = 4285 859 * 5 = 4295 806 * 5 = 4030 808 * 5 = 4040 955 * 5 = 4775 957 * 5 = 4785 959 * 5 = 4795 906 * 5 = 4530 908 * 5 = 4540 355 * 5 = 1775 357 * 5 = 1785 359 * 5 = 1795 306 * 5 = 1530 308 * 5 = 1540 455 * 5 = 2275 457 * 5 = 2285 459 * 5 = 2295 406 * 5 = 2030 408 * 5 = 2040 cdex bcde b = 8 ! 9 ! 4 ! 6 ! 2 7755 * 5 = 38775 7857 * 5 = 39285 7959 * 5 = 39795 7306 * 5 = 36530 7408 * 5 = 37040 2755 * 5 = 13775 2857 * 5 = 14285 2959 * 5 = 14795 2306 * 5 = 11530 2408 * 5 = 12040 5755 * 5 = 18775 5857 * 5 = 19285 5959 * 5 = 19795 5306 * 5 = 16530 5408 * 5 = 17040 0755 * 5 = 03775 0857 * 5 = 04285 0959 * 5 = 04795 0306 * 5 = 01530 0408 * 5 = 02040 bcdex abcde a = 3 ! 8 ! 1 ! 0 87755 * 5 = 438775 87959 * 5 = 439795 82857 * 5 = 414285 85306 * 5 = 416530 80408 * 5 = 402040 97755 * 5 = 488775 97959 * 5 = 489795 92857 * 5 = 464285 95306 * 5 = 4366530 90408 * 5 = 452040 47755 * 5 = 238775 47959 * 5 = 239795 42857 * 5 = 214285 45306 * 5 = 216530 40408 * 5 = 202040 67755 * 5 = 338775 67959 * 5 = 339795 62857 * 5 = 314285 65306 * 5 = 316530 60408 * 5 = 302040 27755 * 5 = 138775 27959 * 5 = 139795 22857 * 5 = 114285 25306 * 5 = 116530 20408 * 5 = 102040 abcdex 142857 * 5 = 714285 ? ok 165306 * 5 = 616530 ? no oder halt so: /* rexx */ do i = 10 to 19 x = right(i,1) y = left(i,5) if i * 5 = x!!y then do say i end end exit gruss stefan |+-+ --+| || "Kleinmanns, Hubert" | || || <[EMAIL PROTECTED]| To: [EMAIL PROTECTED] || || COM> | cc: (bcc: Stefan || || | Raabe/DBS/GDB)^ || || 29.09.2003 08:44 | Subject: AW: Disaster || || Please respond to MQSeries| Recovery scenario for Unix box? [Deutsche || || List | Boerse Systems: Virus checked] || || | || |+-+ --+| Rebecca, you are absolutely
AW: AW: Local Queue Reaches Maximum Depth
Thanks for the information Rebecca, Rick and Ronald! Using the information you provided I estbalished that the relevant "events" are enabled and the messages are be written to SYSTEM.ADMIN... queues. Unfortunately, the "event handler" triggered when messages are written to the admin. queues is basically a "NOP" i.e. it receives control/returns control, but doesn't actually do anything. I'm only covering for a couple of weeks and don't want to start a revolution, but it is something the MQ Administrator will have to look into when he returns. Thanks! Carl -Ursprungliche Nachricht- Von: Ronald Weinger [mailto:[EMAIL PROTECTED] Gesendet: Dienstag, 5. August 2003 14:52 An: [EMAIL PROTECTED] Betreff: Re: AW: Local Queue Reaches Maximum Depth The information is in the MQSeries System Management Guide and Programmable System Management manual. If the app doesn't currently discard the bad messages it should be rewritten. That s/b part of the corporate usage standards. "Mc Burnie, Carl" <[EMAIL PROTECTED]To: [EMAIL PROTECTED] RZBANK.COM> cc: Sent by: "MQSeries Subject: AW: Local Queue Reaches Maximum Depth List" <[EMAIL PROTECTED] C.AT> 08/05/2003 07:45 AM Please respond to "MQSeries List" Hi Ronald, could you elaborate on the "alert mechanism" i.e. does it result in a distinct message being produced? If it does, we could "catch" the message and take appropriate action. When I said "bad" message, I meant a message that the application wasn't expecting (completely different format). It was garbage and should, in my opinion, have been disguarded by the application rather than blocking the queue. Thanks, Carl -Ursprungliche Nachricht- Von: Ronald Weinger [mailto:[EMAIL PROTECTED] Gesendet: Dienstag, 5. August 2003 13:39 An: [EMAIL PROTECTED] Betreff: Re: Local Queue Reaches Maximum Depth You can get an alert generated when the queue reaches a percentage on the max depth. You have to determine what that percentage is, and it will depend on how rapidly messages are being put on the queue. And then have something that will report on it to someone who can act. It is also pretty simple to write an application that will periodically issue an inquiry on the queue depth, so you can generate your own alert at a specific depth, and continue to generate alerts as long as theat value is exceeded. What is a 'bad' message? A good application design is to discard garbage as spam. "Mc Burnie, Carl" <[EMAIL PROTECTED]To: [EMAIL PROTECTED] RZBANK.COM> cc: Sent by: "MQSeries Subject: Local Queue Reaches Maximum Depth List" <[EMAIL PROTECTED] C.AT> 08/05/2003 01:39 AM Please respond to "MQSeries List" Hi, I'm covering WebSphere MQ for z/OS 5.3 for a few weeks and MQ isn't exactly my speciality, so please be patient!! There was a problem yesterday with a local queue that reached maximum depth and overflowed to the "dead letter queue". There was a "bad" message in the local queue and all subsequent messages stacked up behind it until maximum depth was reached and messages began overflowing into the "dead letter queue". We would like to identify this type of problem much earlier in the future and are looking for some automation ideas? 1. The MQ API doesn't seem to return a "maximum depth reached" response for the MQPUT call, is that correct? 2. When the maximum depth for a local queue is reached MQ doesn't seem to write any sort of warning message to the system log, is that correct? 3. When messages start overflowing to the "dead letter queue" MQ doesn't seem to write any sort of message to the system log either, is that correct? My immediate idea was, that if the application performing the MQGET realises a message is "bad" it should move it off to an "error queue" or something rather than leaving it there to block the whole show, any comments? I would be grateful for any ideas or suggestions. I'm only providing emergency cover and the only manuals I could find were for MQS 2.1!!! Thanks, Carl Instructions for managing your mailing list subscription are provided in the Listserv General Users Guide available at http://www.lsoft.com Archive: http://vm.akh-wien.ac.at/MQSeries.archive The informatio
AW: Local Queue Reaches Maximum Depth
Hi Ronald, could you elaborate on the "alert mechanism" i.e. does it result in a distinct message being produced? If it does, we could "catch" the message and take appropriate action. When I said "bad" message, I meant a message that the application wasn't expecting (completely different format). It was garbage and should, in my opinion, have been disguarded by the application rather than blocking the queue. Thanks, Carl -Ursprungliche Nachricht- Von: Ronald Weinger [mailto:[EMAIL PROTECTED] Gesendet: Dienstag, 5. August 2003 13:39 An: [EMAIL PROTECTED] Betreff: Re: Local Queue Reaches Maximum Depth You can get an alert generated when the queue reaches a percentage on the max depth. You have to determine what that percentage is, and it will depend on how rapidly messages are being put on the queue. And then have something that will report on it to someone who can act. It is also pretty simple to write an application that will periodically issue an inquiry on the queue depth, so you can generate your own alert at a specific depth, and continue to generate alerts as long as theat value is exceeded. What is a 'bad' message? A good application design is to discard garbage as spam. "Mc Burnie, Carl" <[EMAIL PROTECTED]To: [EMAIL PROTECTED] RZBANK.COM> cc: Sent by: "MQSeries Subject: Local Queue Reaches Maximum Depth List" <[EMAIL PROTECTED] C.AT> 08/05/2003 01:39 AM Please respond to "MQSeries List" Hi, I'm covering WebSphere MQ for z/OS 5.3 for a few weeks and MQ isn't exactly my speciality, so please be patient!! There was a problem yesterday with a local queue that reached maximum depth and overflowed to the "dead letter queue". There was a "bad" message in the local queue and all subsequent messages stacked up behind it until maximum depth was reached and messages began overflowing into the "dead letter queue". We would like to identify this type of problem much earlier in the future and are looking for some automation ideas? 1. The MQ API doesn't seem to return a "maximum depth reached" response for the MQPUT call, is that correct? 2. When the maximum depth for a local queue is reached MQ doesn't seem to write any sort of warning message to the system log, is that correct? 3. When messages start overflowing to the "dead letter queue" MQ doesn't seem to write any sort of message to the system log either, is that correct? My immediate idea was, that if the application performing the MQGET realises a message is "bad" it should move it off to an "error queue" or something rather than leaving it there to block the whole show, any comments? I would be grateful for any ideas or suggestions. I'm only providing emergency cover and the only manuals I could find were for MQS 2.1!!! Thanks, Carl Instructions for managing your mailing list subscription are provided in the Listserv General Users Guide available at http://www.lsoft.com Archive: http://vm.akh-wien.ac.at/MQSeries.archive The information contained in this message may be CONFIDENTIAL and is for the intended addressee only. Any unauthorized use, dissemination of the information, or copying of this message is prohibited. If you are not the intended addressee, please notify the sender immediately and delete this message. Instructions for managing your mailing list subscription are provided in the Listserv General Users Guide available at http://www.lsoft.com Archive: http://vm.akh-wien.ac.at/MQSeries.archive Instructions for managing your mailing list subscription are provided in the Listserv General Users Guide available at http://www.lsoft.com Archive: http://vm.akh-wien.ac.at/MQSeries.archive
Local Queue Reaches Maximum Depth
Hi, I'm covering WebSphere MQ for z/OS 5.3 for a few weeks and MQ isn't exactly my speciality, so please be patient!! There was a problem yesterday with a local queue that reached maximum depth and overflowed to the "dead letter queue". There was a "bad" message in the local queue and all subsequent messages stacked up behind it until maximum depth was reached and messages began overflowing into the "dead letter queue". We would like to identify this type of problem much earlier in the future and are looking for some automation ideas? 1. The MQ API doesn't seem to return a "maximum depth reached" response for the MQPUT call, is that correct? 2. When the maximum depth for a local queue is reached MQ doesn't seem to write any sort of warning message to the system log, is that correct? 3. When messages start overflowing to the "dead letter queue" MQ doesn't seem to write any sort of message to the system log either, is that correct? My immediate idea was, that if the application performing the MQGET realises a message is "bad" it should move it off to an "error queue" or something rather than leaving it there to block the whole show, any comments? I would be grateful for any ideas or suggestions. I'm only providing emergency cover and the only manuals I could find were for MQS 2.1!!! Thanks, Carl Instructions for managing your mailing list subscription are provided in the Listserv General Users Guide available at http://www.lsoft.com Archive: http://vm.akh-wien.ac.at/MQSeries.archive
